Re: [time-nuts] Measuring coax temperature coefficient with a TICC

[Scott Stobbe]

A table of a bunch of rg6 catv permutations,
http://www.texcan.com/media/import/pdf/Electronic_Cable_RG6_RG59.pdf

At least on this list if it has a solid copper core, it also has a copper
braid shield. I'm sure there is many more permutations out there.

On Wed, Apr 19, 2017 at 7:00 PM Will Kimber  wrote:

> TV co-ax these days for satellite or UHF is almost all steel wire with
> copper plating.  In fact the 'F' connector that is used is designed to
> use that stiff wire as the center pin of the connector!
>
>
> Will
>
>
> On 04/20/2017 06:57 AM, Hal Murray wrote:
> > kb...@n1k.org said:
> >> I’d want to be pretty sure what the center conductor was made out of.
> I’ve
> >> seen some stuff in coax that “one would think† should not be there
> (copper
> >> over steel …).
> > Does that effect the propagation time?
> >
> > If I gave you a good scope picture of a pulse after going through chunk
> of
> > coax, could you figure out the ratio of copper to steel?  Would you need
> to
> > know the length or could you figure that out too?
> >
> >
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] OCXO Soft-Start

[Scott Stobbe]

I can't say I have run into that issue with a MCU as most 21st century ones
have a decent POR and Brown out detect (which typically burns 10x more
current than a 32k XO + RTC, and may get switched off in battery
applications, and then problems can occur). What does seem to come up is
stuff hanging off non-always on power rails (ADC,DAC,Sensors,etc), leakage
and back-feeding onto their dedicated supply has them try to startup on
leakage but there isn't enough leakage to actually power the device. They
may issue the on die reset once and then the supply collapses all the flops
lose their reset state, but the IC dosen't try to reset again.

Fortunately, you can buy a bunch of LDOs which include a small discharge
FET to help this case as well as others.


On Thu, Apr 13, 2017 at 9:17 AM, Bob kb8tq <kb...@n1k.org> wrote:

> Hi
>
> It’s a good bet that there is at least one regulator IC inside any modern
> OCXO.
> As you slowly ramp the input voltage on a regulator, various odd things may
> or may not happen. A 1 mv / s ramp on the outside can be “who knows what”
> at the oscillator level. That said, slow voltage ramps are a really good
> way to put
> all sorts of oscillator circuits into really odd modes. Clock oscillators
> (XO’s) and MCU
> built in clock circuits very definitely fall into this category.
>
> The same sort of “who knows what” problem also gets into the rest of the
> circuit.
> A limited supply might work fine nine times out of ten or 99 out of 100.
> On the
> odd time out, something goes poof !
>
> Bob
>
> > On Apr 12, 2017, at 11:40 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
> >
> > Bob, Rick, my use of modes vs overtones was loose, you guys are spot on.
> >
> >
> > That's a fun challenge, suppressing a mode 10% higher in frequency.
> That's
> > a bit of a fussy LC filter/trap.
> >
> > So if I assume the tempCo of the B-mode to be 20 ppm/degC, that would
> mean
> > the heater servo has a sustained oscillation of about 250 udegC all day
> > long.
> >
> > Power cycling for 100 ms isn't long enough for the crystal motion to die
> > out, 200 ms seems to do it, which ~2 MCycles. At which point it runs on
> the
> > 10MHz C-Mode as desired.
> >
> > I couldn't say who Trimble went with, to me its an ebay special :), but
> > photo attached anyways.
> >
> > When the supply is current or power limited the effective supply slew
> rate
> > is something like 1 mV/s, and under these conditions, at least this
> > particular unit starts up on B-mode every time.
> >
> > Running one of these on a USB supply (5V, 500mA = 2.5W) is looking
> > plausible.
> >
> > On Wed, Apr 12, 2017 at 9:30 PM, Bob kb8tq <kb...@n1k.org> wrote:
> >
> >> Hi
> >>
> >> A SC cut crystal has multiple resonance “modes” in can operate on. This
> is
> >> in addition to the
> >> normal set of fundamental, third overtone, fifth overtone and so on. For
> >> various interesting reasons
> >> the SC modes are called A, B and C.  On a normal 10 MHz crystal, the
> “main
> >> mode” is the C mode. The
> >> A mode is well below this mode both in frequency and amplitude. The B
> mode
> >> is more problematic, It is
> >> 8 to 12% above the C mode in frequency and may be higher in amplitude
> >> (lower resistance) than the
> >> desired mode.
> >>
> >> The C mode has a “useful” third order temperature characteristic. The B
> >> mode has a fairly steep linear
> >> temp co. One of the classic design experiments on a SC based design is
> to
> >> force the oscillator onto the
> >> B mode. This lets you investigate the oven gain while running a
> >> temperature run. What you have likely
> >> done is to put the unit onto the B mode.
> >>
> >> Oscillators tend to be fractal when switching between modes. Minor
> startup
> >> issues can drive the circuit
> >> one way or the other. Once it gets onto a mode, it may be quite
> difficult
> >> to get it off of that mode ….
> >>
> >> Bob
> >>
> >>> On Apr 12, 2017, at 5:38 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> >> wrote:
> >>>
> >>> Hello,
> >>>
> >>> I wanted to see if I could soft-start a used OCXO (Trimble 34310)
> during
> >>> warm-up. By default with an appropriately rated 12 VDC supply, the OCXO
> >>> starts the heater at about 8 W, and eventually settles down to 2 W for
> >>> 20-25 degC ambient temperature. Figure Attached.
> >>>
> >

[time-nuts] OCXO Soft-Start

[Scott Stobbe]

Hello,

I wanted to see if I could soft-start a used OCXO (Trimble 34310) during
warm-up. By default with an appropriately rated 12 VDC supply, the OCXO
starts the heater at about 8 W, and eventually settles down to 2 W for
20-25 degC ambient temperature. Figure Attached.

The good news is it does startup with either a current limited or power
limited supply. Albeit, for the constant current case the start-up time is
dramatically longer. Figures Attached.

The bad news, but interesting tangent is that when this particular OCXO is
soft-started, it doesn't oscillate at 10 MHz! It starts up on a spurious
overtone at 10.9 MHz. There isn't even a hint of 10 MHz in the output
spectrum just the 10.9 MHz tone (and some noise from the banana clip-leads
loop-antenna). So this is a bit of a pain for soft-starting because,
although the oven eventually comes into regulation, the oscillator is
running on the wrong frequency.

I'm not sure what the tempCo of the 10.9 MHz crystal mode is, but its a
pretty great temperature probe of the crystal (literally). You can see the
sinusoidal frequency disturbance of this tone has the same periodicity as
that of the heater current (~50 s). So I don't know if the heater is
oscillating at the uK level or mK level, will need to find out the tempCo
of this crystal mode. Neat to see the oven dynamics.

Hopefully once the oven is settled at temp, I can load switch the OCXO for
10 or 100 ms and get it to startup on the correct 10 MHz mode. With a 10+ W
rated supply it starts up on 10 MHz every time.
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Car Clock drift - the lowly 32kHz tuning fork crystal specs

[Scott Stobbe]

The trim method that stands out from memory for generic RTC chips is to
cycle stall or double clock, x cycles every 60 seconds. Yielding 0.5 ppm
trim resolution.

On Sun, Apr 9, 2017 at 7:45 AM, Tim Shoppa  wrote:

> I've had only a few different cars over the past 25 years but I've been
> impressed with how accurate their mass-market built-in clocks are,
> especially considering the wide and completely uncontrolled temperature
> range. In the winter the interior of the car gets down below freezing most
> mornings, and in the summer the interior gets way above 120F in sunlight.
>
> (Contrast the above with the time-nuttery here where folks buy double-oven
> OCXO's and then they insist that the OCXO's have to be put in temperature
> controlled environments.)
>
> I only set the car clock twice a year, at daylight savings time changes.
> Yet between daylight savings time changes, the car clock never drifts by
> more than a minute.
>
> 60 seconds in half a year is 4ppm. So I went and looked at the specs of a
> stock 32kHz crystal, for example
> http://www.mouser.com/ds/2/77/CFS-CFV-4402.pdf
>
> 1: The crystal is speced as having a turnover point of 25C. I understand
> that.
> 2: Frequency at the turnover point is speced as being +/-20ppm. OK, that's
> not bad, most of that can be compensated for with a small trimmer cap at
> the factory to the 4ppm range. Or maybe they just program in the clock
> divider at the factory appropriate to the crystal.
> 3: The temperature coefficient of the tuning fork cut around the turnover
> point seems to always be the same: -.034ppm per deg C squared. If the temp
> goes down to 5 deg C, then, the frequency changes by 14ppm. If the temp
> goes down to -5 deg C, the frequency changes by 30ppm.
>
> With that temperature coefficient, temperatures like -5C or 5C that are
> common every winter would result in a few minutes of drift every winter.
> Yet I never observe that drift.
>
> So my conclusion, is that all these car clocks must be temperature
> compensated. And they must've been doing this for several decades at this
> point.
>
> That shouldn't be too surprising - right next to the clock display on the
> dashboard is a digital thermometer. Maybe 30 or more years ago the
> temperature compensation was done by analog circuitry, but today I'm
> guessing there's a digital chip that takes the thermometer reading and
> numerically adjusts the divider word for the 32kHz oscillator to
> temperature compensate the clock digitally.
>
> Is there a way to verify my guess at the TCXO method?
>
> I'm guessing that all the better quartz wristwatches use a similar
> technology too. Maybe they have a different crystal cut that is closer to
> body temperature for the turnover point.
>
> Tim N3QE
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Line Frequeny Stablity

[Scott Stobbe]

I would guess the tightest control loop is on the generator stator field
windings, with mechanical control being secondary. Definitely a lot of
poles and zeros to worry about.

On Wed, Apr 5, 2017 at 9:01 AM, jimlux  wrote:

> On 4/4/17 2:28 PM, Thomas D. Erb wrote:
>
>> Thanks for the info.
>>
>>
>> So that tells me how data is recorded - but not how the frequency is kept
>> stable ?
>>
>> Is the line frequency now directly tied to GPS clock - with no drift ?
>>
>
> The line frequency is adjusted, for the most part, by adjusting the prime
> power (steam valves, dam penstocks, etc.) on the generators at power
> stations. That changes the speed, slightly, although as generator 1 of N
> starts to get ahead, the electrical load increases, and it slows down.
>
> It's actually a pretty complex system, since there are a whole raft of
> "spring constants" in between the multiple generators in a system, there's
> phase shifts due to transmission line inductance and capacitance.
>
> "Stabilizing" a system in the face of changing demand is a non-trivial
> task.
>
>
>
>
>
>
>
>> Thomas D. Erb
>> t...@electrictime.com /
>> Electric Time Company, Inc.
>> Office: 508-359-4396 x 117 / Fax: 508-359-4482
>> 97 West Street Medfield, MA 02052 USA
>> www.electrictime.com
>> [Facebook]> Company-Inc/127918073950854?ref=hl>[Twitter]> twitter.com/tower_clocks>[pinterest]> pinterest.com/electrictime/>
>> [htmlsig.com]
>> ___
>> time-nuts mailing list -- time-nuts@febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/m
>> ailman/listinfo/time-nuts
>> and follow the instructions there.
>>
>>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/m
> ailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] TAPR TICC boxed

[Scott Stobbe]

Also interesting the LTC6752 is rail-rail input. Any rail-rail input opamp
I've used ends up with an ugly bump in input offset voltage transitioning
from the nmos or npn diff pair to the pmos or nmos. I'm not sure how good
or bad a rail-rail comparator may behave when common-mode biased in that
region.

On Fri, Mar 31, 2017 at 11:22 PM Bruce Griffiths <bruce.griffi...@xtra.co.nz>
wrote:

> Attempting sub nanosecond timing with an actual 1Mohm source is an
> exercise in futility. There are very few cases where one would want to
> attempt precision timing measurements with such a high impedance source.
> The 1M pulldown on the TICC input is merely intended to maintain a valid
> logic input should the user leave that input disconnected. In actual use
> with PPS signals the source impedance is in most cases a few tens of ohms.
> If one wishes to have a 1Mohm input impedance for use with AC coupled
> signals then a low noise FET input buffer preceding the comparator is
> required.
>
> Protection diodes in this application not only need to have low leakage,
>  they also need to turn on and off fast enough to be useful.
>
> The propagation delay dispersion (both vs common mode and vs overdrive)
> also need to be considered along with the comparator jitter.
>
>
> Bruce
>
>  and overdrive (both vs overdrive and vs input common modeOn 01 April 2017
> at 15:34 Scott Stobbe <scott.j.sto...@gmail.com> wrote:
>
> Fwiw, for a precision comparator you'll probably want a bipolar front end
> for a lower flicker corner and better offset stability over cmos. For
> high-speeds the diffpair is going to be biased fairly rich for bandwidth.
> So you will more than likey have input bias currents of 100's of nA to uA
> on your comparator. Which is not great with a 1 megohm source.
>
> On Fri, Mar 31, 2017 at 9:08 PM Charles Steinmetz <csteinm...@yandex.com>
> wrote:
>
> Mark wrote:
>
> I thought about using the clamp diodes as protection but was a bit
> worried about power supply noise leaking through the diodes and adding some
> jitter to the input signals...
>
> It is a definite worry even with a low-noise, 50 ohm input, and a
> potential disaster with a 1Mohm input. Common signal diodes (1N4148,
> 1N914, 1N916, 1N4448, etc.) are specified for 5-10nA of reverse current.
> Even a low-leakage signal diode (e.g., 1N3595) typically has several
> hundred pA of leakage. Note that the concern isn't just power supply
> noise -- the leakage current itself is quite noisy.
>
> For low-picoamp diodes at a decent price, I use either (1) the B-C diode
> of a small-signal BJT, or (2) the gate diode of a small-geometry JFET.
> A 2N5550 makes a good high-voltage, low-leakage diode with leakage
> current of ~30pA. Small signal HF transistors like the MPSH10 and
> 2N5179 (and their SMD and PN variants) are good for ~5pA, while the gate
> diode of a PN4417A JFET (or SMD variant) has reverse leakage current of
> ~1pA (achieving this in practice requires a very clean board and good
> layout).
>
> I posted some actual leakage test results to Didier's site, which can be
> downloaded at
> <
>
> http://www.ko4bb.com/getsimple/index.php?id=download=03_App_Notes_-_Proceedings/Reverse_leakage_of_diode-connected_BJTs_and_FETs_measurement_results.pdf
>
> .
> This document shows the connections I used to obtain the data.
>
> The TICC doesn't have the resolution for it to matter or justify a
> HP5370 or better quality front end. I'll probably go with a fast
> comparator to implement the variable threshold input.
>
> Properly applied, a fast comparator will have lower jitter than the rest
> of the errors, and is an excellent choice. Bruce suggested the LTC6752,
> which is a great part if you need high toggle speeds (100s of MHz) or
> ultra-fast edges. But you don't need high toggle rates and may not need
> ultra-fast edges. Repeatability and stability are more important than
> raw speed in this application. The LT1719, LT1720, or TLV3501 may work
> just as well for your purpose, and they are significantly less fussy to
> apply.
>
> Note that the LTC6752 series is an improved replacement for the ADCMP60x
> series, which itself is an improved replacement for the MAX999. Of
> these three, the LTC6752 is the clear winner in my tests. If you do
> choose it (or similar), make sure you look at the transitions with
> something that will honestly show you any chatter at frequencies up to
> at least several GHz. It only takes a little transition chatter to
> knock the potential timing resolution of the ultra-fast comparator way
> down. Do make sure to test it with the slowest input edges you need it
> to handle.
>
> Best regards,
>
> Charles
>
> __

Re: [time-nuts] TAPR TICC boxed

[Scott Stobbe]

Also for interest the 53131a schematic is available at
http://bee.mif.pg.gda.pl/ciasteczkowypotwor/HP/53131.pdf

HP used a low input bias current bjt opamp, the Lt1008 to bias/dc servo a
custom JFET buffer driving an AD96687 comparator.

On Fri, Mar 31, 2017 at 10:34 PM Scott Stobbe <scott.j.sto...@gmail.com>
wrote:

> Fwiw, for a precision comparator you'll probably want a bipolar front end
> for a lower flicker corner and better offset stability over cmos. For
> high-speeds the diffpair is going to be biased fairly rich for bandwidth.
> So you will more than likey have input bias currents of 100's of nA to uA
> on your comparator. Which is not great with a 1 megohm source.
>
> On Fri, Mar 31, 2017 at 9:08 PM Charles Steinmetz <csteinm...@yandex.com>
> wrote:
>
> Mark wrote:
>
> > I thought about using the clamp diodes as protection but was a bit
> worried about power supply noise leaking through the diodes and adding some
> jitter to the input signals...
>
> It is a definite worry even with a low-noise, 50 ohm input, and a
> potential disaster with a 1Mohm input.  Common signal diodes (1N4148,
> 1N914, 1N916, 1N4448, etc.) are specified for 5-10nA of reverse current.
>   Even a low-leakage signal diode (e.g., 1N3595) typically has several
> hundred pA of leakage.  Note that the concern isn't just power supply
> noise -- the leakage current itself is quite noisy.
>
> For low-picoamp diodes at a decent price, I use either (1) the B-C diode
> of a small-signal BJT, or (2) the gate diode of a small-geometry JFET.
> A 2N5550 makes a good high-voltage, low-leakage diode with leakage
> current of ~30pA.  Small signal HF transistors like the MPSH10 and
> 2N5179 (and their SMD and PN variants) are good for ~5pA, while the gate
> diode of a PN4417A JFET (or SMD variant) has reverse leakage current of
> ~1pA (achieving this in practice requires a very clean board and good
> layout).
>
> I posted some actual leakage test results to Didier's site, which can be
> downloaded at
> <
> http://www.ko4bb.com/getsimple/index.php?id=download=03_App_Notes_-_Proceedings/Reverse_leakage_of_diode-connected_BJTs_and_FETs_measurement_results.pdf
> >.
>   This document shows the connections I used to obtain the data.
>
> > The TICC doesn't have the resolution for it to matter or justify a
> HP5370 or better quality front end.   I'll probably go with a fast
> comparator to implement the variable threshold input.
>
> Properly applied, a fast comparator will have lower jitter than the rest
> of the errors, and is an excellent choice.  Bruce suggested the LTC6752,
> which is a great part if you need high toggle speeds (100s of MHz) or
> ultra-fast edges.  But you don't need high toggle rates and may not need
> ultra-fast edges.  Repeatability and stability are more important than
> raw speed in this application.  The LT1719, LT1720, or TLV3501 may work
> just as well for your purpose, and they are significantly less fussy to
> apply.
>
> Note that the LTC6752 series is an improved replacement for the ADCMP60x
> series, which itself is an improved replacement for the MAX999.  Of
> these three, the LTC6752 is the clear winner in my tests.  If you do
> choose it (or similar), make sure you look at the transitions with
> something that will honestly show you any chatter at frequencies up to
> at least several GHz.  It only takes a little transition chatter to
> knock the potential timing resolution of the ultra-fast comparator way
> down.  Do make sure to test it with the slowest input edges you need it
> to handle.
>
> Best regards,
>
> Charles
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] TAPR TICC boxed

[Scott Stobbe]

Fwiw, for a precision comparator you'll probably want a bipolar front end
for a lower flicker corner and better offset stability over cmos. For
high-speeds the diffpair is going to be biased fairly rich for bandwidth.
So you will more than likey have input bias currents of 100's of nA to uA
on your comparator. Which is not great with a 1 megohm source.

On Fri, Mar 31, 2017 at 9:08 PM Charles Steinmetz 
wrote:

> Mark wrote:
>
> > I thought about using the clamp diodes as protection but was a bit
> worried about power supply noise leaking through the diodes and adding some
> jitter to the input signals...
>
> It is a definite worry even with a low-noise, 50 ohm input, and a
> potential disaster with a 1Mohm input.  Common signal diodes (1N4148,
> 1N914, 1N916, 1N4448, etc.) are specified for 5-10nA of reverse current.
>   Even a low-leakage signal diode (e.g., 1N3595) typically has several
> hundred pA of leakage.  Note that the concern isn't just power supply
> noise -- the leakage current itself is quite noisy.
>
> For low-picoamp diodes at a decent price, I use either (1) the B-C diode
> of a small-signal BJT, or (2) the gate diode of a small-geometry JFET.
> A 2N5550 makes a good high-voltage, low-leakage diode with leakage
> current of ~30pA.  Small signal HF transistors like the MPSH10 and
> 2N5179 (and their SMD and PN variants) are good for ~5pA, while the gate
> diode of a PN4417A JFET (or SMD variant) has reverse leakage current of
> ~1pA (achieving this in practice requires a very clean board and good
> layout).
>
> I posted some actual leakage test results to Didier's site, which can be
> downloaded at
> <
> http://www.ko4bb.com/getsimple/index.php?id=download=03_App_Notes_-_Proceedings/Reverse_leakage_of_diode-connected_BJTs_and_FETs_measurement_results.pdf
> >.
>   This document shows the connections I used to obtain the data.
>
> > The TICC doesn't have the resolution for it to matter or justify a
> HP5370 or better quality front end.   I'll probably go with a fast
> comparator to implement the variable threshold input.
>
> Properly applied, a fast comparator will have lower jitter than the rest
> of the errors, and is an excellent choice.  Bruce suggested the LTC6752,
> which is a great part if you need high toggle speeds (100s of MHz) or
> ultra-fast edges.  But you don't need high toggle rates and may not need
> ultra-fast edges.  Repeatability and stability are more important than
> raw speed in this application.  The LT1719, LT1720, or TLV3501 may work
> just as well for your purpose, and they are significantly less fussy to
> apply.
>
> Note that the LTC6752 series is an improved replacement for the ADCMP60x
> series, which itself is an improved replacement for the MAX999.  Of
> these three, the LTC6752 is the clear winner in my tests.  If you do
> choose it (or similar), make sure you look at the transitions with
> something that will honestly show you any chatter at frequencies up to
> at least several GHz.  It only takes a little transition chatter to
> knock the potential timing resolution of the ultra-fast comparator way
> down.  Do make sure to test it with the slowest input edges you need it
> to handle.
>
> Best regards,
>
> Charles
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] PLL performance?

[Scott Stobbe]

Neat Project. I don't know if it will come up for you but optical or hall
rotary encoders are notorious for jitter. While a generic IC comparator may
have an open loop-gain of 100 dB, creating the mechanical equivalent is not
so easy. Hall/optical have a softer switch on/off curve. Depending what you
choose to instrument your pendulum may also introduce more jitter. The
20logN dosen't help either, 1 millideg at 0.5 Hz is 5.5 cycles at 1 MHz.

On Mon, Mar 20, 2017 at 9:07 PM, David Scott Coburn 
wrote:

> Hi All,
>
> I have built and tested a PLL circuit that will be used to generate a 1
> MHz signal locked to a 0.5 HZ signal from a pendulum.  (Details available
> upon request.)
>
> The circuit is a classic 4046 generating the 1 MHz signal which is fed
> into a 2e6 digital divider which outputs 0.5 Hz which is fed back to the
> 4046 phase comparator (PC).
>
> I take a 1 MHz signal from an HP 107A run through another 2e6 divider to
> generate a reference 0.5 Hz signal for the other 4046 PC input.
>
> I tested this by feeding the 0.5 Hz output of the PLL into a "time-stamp
> counter" board which I built to go into an HP 3582A Data Acquisition unit.
> The TSC uses the 5 MHz signal from the HP 107A to feed a free-running
> 32-bit binary counter.  The 0.5 Hz input latches the count value (on the
> rising edge of the signal), which is then logged.
>
> See the attached diagram.  The PLL under test is in the red box.  (Not
> sure what the policy is here for attachments?)
>
> If all was perfect I would get a string of values of 10,000,000 counts
> each, one every 2 seconds.
>
> Over the course of one day the average reading is, in fact, 10e6, so the
> PLL looks to be working over "long" time scales.
>
> The attached histogram plot shows the actual data for the 0.5 Hz signal,
> showing the distribution of deviations from 10e6 counts.  This is almost a
> full day of data, about 40,000 readings.
>
> The standard deviation for the data is about 55 counts.
>
> The plot looks to my eye to be a nice Gaussian shape, so I assume that the
> deviations are caused mainly by (white?) noise.  There does not look to be
> much other structure in the shape of the data.  (Comments welcome.)
>
> Sorry for the long introduction, there are some questions coming!
>
> I have looked for information on the web about others who may have done
> this kind of PLL, but did not find much.
>
> Does anyone know of any articles related to this?
>
> If so, do you know what kind of performance they got?
>
> What kind of statement could I make about the 'stability' of this
> circuit?  Simplistically: a 'stability' of ~50 counts in 10e6 is ~5e-7?
>
> By the way, this performance is WAY WAY beyond what I was expecting
>
> Cheers,
>
> Scott
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Bye-Bye Crystals

[Scott Stobbe]

I would be careful to get an all analog tcxo. The digi-trim ones while may
have a decent total error band over temperature  can hop up and down in 100
ppb steps when the temperature straddles two points on its temp comp table.

On Wed, Mar 15, 2017 at 7:06 PM Bob Camp  wrote:

> Hi
>
> > On Mar 15, 2017, at 3:55 PM, Dan Kemppainen 
> wrote:
> >
> > Hi Bob,
> >
> > If one prototypes any crystal oscillator, and runs it on a bench. Then
> builds an 'oven' of sorts and runs it on the same bench. Would you expect
> to see any improvement?
>
> Sure, the XO likely moves 50 to 100 ppm over -30 to +70. You will cut that
> down to a couple of ppm. It’s much easier these
> days to just buy (or salvage) a cheap TCXO to get the same level of
> stability.
>
> One very basic issue: If I just grab a random crystal, it likely is a cut
> that does *not* have a useful turn temperature at all. For
> a proper OCXO you need a crystal with a turn temperature in the practical
> range for your oven. There are many other
> issues.
>
> The key point (just as in the previous message) is that you must have good
> frequency vs temperature data to know if
> you are improving things or not. That involves having a real temperature
> test chamber than can be slewed in a controlled
> fashion and repeatably set to a sequence of temperatures. A typical run
> starts at room, steps down to cold (or up to hot).
> It then steps to the other end and finally steps back to room. Data is
> taken every 10C or so and analyzed to be sure that
> things are not all messed up. One obvious problem / issue would be drift
> during the run. A typical run takes several hours to
> most of a day.
>
> Bob
>
> >
> > In other words for a given oscillator (crystal and electronics, etc),
> will there be any improvements in an oven compared to not in an oven? Or,
> are there other things that outweigh the gains by temp controlling the
> whole thing.
> >
> > Yeah, this is a pretty open question, but I don't really have a feel for
> how good an oscillator needs to be before an oven starts to improve
> things...
> >
> > Dan
> >
> >
> >
> >
> > On 3/15/2017 12:00 PM, time-nuts-requ...@febo.com wrote:
> >> Hi
> >>
> >> Where do you plan on getting an OCXO grade crystal at an odd frequency
> like
> >> that? Much of the performance of a good OCXO is in the crystal. Doing a
> proper
> >> design on one is a lot of work. You *might* think that having a design
> for 5.00
> >> MHz would give you a good design for 5.50 MHz. I have empirical
> evidence that
> >> this isn’t the case. Many years later, I’m still utterly amazed that
> this is the way things
> >> work in the crystal business ….( = it’s not just a design issue, it’s
> also a business decision)
> >>
> >> More or less the crystal needs to be:
> >>
> >> 1) Cut specifically to have a turn at a temperature that makes sense
> for your application.
> >> 2) A “large blank” design (for it’s frequency)
> >> 3) In a cold weld package (most of the normal crystals are resistance
> weld)
> >> 4) Run through a high vacuum / high temperature process
> >> 5) Be plated with gold rather than something like silver or aluminum
> (unless it’s at VHF).
> >> 6) Have a motional capacitance that makes sense for your EFC range (
> normally = minimize)
> >> 7) Preferably be an SC or modified SC cut.
> >>
> >> This is for a high stability part. The list does keep going on for a
> while, but that should
> >> give you a pretty good idea.
> >>
> >> Bob
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Bye-Bye Crystals

[Scott Stobbe]

Not only that. Good luck finding a datasheet with *any* analog
specifications for its internal oscillator. Here are the pins for an
external crystal. The microchip PICs are nice, they give you the goldilocks
selection for drive level a little cool, a little hot, maybe just right.

On Tue, Mar 14, 2017 at 8:04 AM, Bob Camp  wrote:

> Hi
>
> > On Mar 14, 2017, at 4:44 AM, Hal Murray  wrote:
> >
> >
> > artgod...@gmail.com said:
> >> I'm not after quality - I do have an application in mind but it doesn't
> need
> >> to compete with mass production. Just wondering if it's feasible to make
> >> something crude that will resonate.
> >
> > Are you doing this for fun or ???
> >
> > Feasible?  Sure.  Cheaper?  That depends.
> >
> > The cost difference between a complete oscillator package and a simple
> > crystal is tiny.  The osc is often cheaper if you include board space or
> > engineering time.
>
> Purchased in volume, the difference it the price of a crystal vs a
> complete XO
> is enormous. You will see at least a 10:1 cost savings on the crystal and
> likely
> more than that.  Simply attaching a crystal to the internal oscillator
> inside a
> chip is nearly zero engineering cost.  If your product is cost sensitive
> and
> not super tight tolerance … you go with the crystal.
>
> Bob
>
> >
> > Is your background digital or analog?  Do you want a sine wave or a
> clock?
> >
> > My background is primarily digital.  If the chip you are using has 2 pins
> > setup to drive a crystal, you can probably get it to run reliably by
> > following the data sheet and/or app notes.  The usual recipe is 2 tiny
> caps
> > and a big resistor.  (big in resistance, not physically big)
> >
> > An advantage of using a crystal with the on-chip amplifier that I didn't
> > mention last time is that you save the osc power if you power down that
> > corner of the chip.
> >
> > If you want a sine wave, you are out of my comfort zone.  I'd probably
> look
> > in ham radio literature.
> >
> > They make logic chips like a 74HCU04, U for unbuffered.  One of their
> uses is
> > for making oscillators.  I've never done it.  Try google.
> >
> > --
> > These are my opinions.  I hate spam.
> >
> >
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] time/sampling behavior of RTL-SDR

[Scott Stobbe]

I bought one awhile back from a company called NooElec as more of a novelty
item than anything. But I can say it runs quite hot and if you tune to the
3rd harmonic of a 10 MHz reference it takes a while to thermally stabilize.
In hindsight spending the extra 3-4 dollars for one with a TCXO would have
been worthwhile.

On Tue, Mar 7, 2017 at 4:08 PM, jimlux  wrote:

> Has anyone ever looked at the timing consistency/behavior of the cheap
> RTL-SDR dongles?
>
> That is, if you feed a sine wave in and get the samples out via the USB
> interface, are there missing samples?  What's the ADC clock look like?
>
> If you had two of them, and you determine the relative phase of them at
> some time t0, and you let it run for minutes, how far will it have moved?
>
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/m
> ailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

[time-nuts] Fwd: Dusty 53131A

[Scott Stobbe]

Recently the fan in my 53131 has been making a pretty awful sound, so its
time to replace it. When I disassembled the counter I found the trim cap
had turned into a fur ball. Not that it really matters since it runs on an
external reference, but its interesting to see if the dust has any effect.
See https://www.flickr.com/photos/147407087@N06/32427230764/. So with the
unit disassembled I logged the effect of removing the dust by logging a
GPSDO during the cleaning process.

Must of the dust was around the shroud of the trim cap. Near the mounting
terminals there was fairly minimal amounts of dust. At least in this
instance, if the dust had any impact on frequency, its orders of magnitude
below the thermal instability of the AT crystal.

The crystal appears to be in a to-39 3-lead package. It is also
freestanding directly in front of the exhausting cooling fan...
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP5328A + NI GPIB-ENET/1000

[Scott Stobbe]

If your counter is showing up in the NI device tree, you should be pretty
close to getting it working.

I haven't used a 5328A, but I suspect the command you have shown above is
sample code for a basic program, the actual HPIB command may just be a 'T'.

On Thu, Feb 23, 2017 at 1:37 AM, James Peroulas  wrote:

> Does anyone have experience using older HP equipment with the NI
> GPIB-ENET/1000 GPIB to ethernet box? Is this even possible with such an old
> piece of equipment? I'm very familiar with using SCPI commands over telnet
> but not so familiar with HPIB.
>
> So far I've installed the NI 488.2 software which has successfully found
> both the GPIB-ENET/1000 box and the HP5328A connected to it. I've opened
> the NI-488.2 communicator which should allow me to send commands to the
> counter and read back results. However, anything I type in the "send
> string" returns a timeout.
>
> The manual for the 5328A shows example commands such as:
> CMD"?U*","T"
> But it's not clear to me what portion (all?) is the string that I should be
> sending to the 5328a.
>
> The 5328a is set to "Talk Only" mode via dip switch. I would expect that I
> should be able to just 'read' the results, but 'read' also times out...
>
> Thanks for any pointers,
> James
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Vintage Frequency Measurement

[Scott Stobbe]

Neat bit of kit. Was it common to get IMD with harmonics out of the
vibrator power supply, and end up on tuning in on the wrong frequency?

On Wed, Feb 15, 2017 at 2:28 AM, Robert Atkinson via time-nuts <
time-nuts@febo.com> wrote:

> Hi,Sorry if I caused confusion by calling the SCR/BC221 a wavemeter.
> Clearly it's not in the wider usage of the term, and the manual and front
> panel call it a frequency meter. However the similar British device was
> called a wavemeter "Wavemeter Class D" http://www.vmarsmanuals.co.uk/
> archive/724_Wavemeter_Class_D_No2_Working_Instructions.pdfSo here in the
> UK the 221 was often also called a wavemeter. Classic wavemeters were also
> available for example the Marconi TF975.
> Robert G8RPI.
>
>   From: Bob Camp 
>  To: Discussion of precise time and frequency measurement <
> time-nuts@febo.com>
>  Sent: Monday, 13 February 2017, 23:46
>  Subject: Re: [time-nuts] Vintage Frequency Measurement
>
> Hi
>
> With a VFO running, you have a heterodyne frequency meter. That is (at
> least to me)
> a very different device than an absorption wave meter. I know way to put
> power into
> a BC-221 and use it as an absorption device.
>
> I’m not in any way saying that the LM or the 221 are less useful. They are
> still to this day
> great little boxes. The just aren’t (by my understanding) wave meters.
> That term describes
> a different device that works a different way.
>
> Bob
>
> > On Feb 13, 2017, at 5:52 PM, John Miles  wrote:
> >
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] ``direct'' RS-232 vs. RS-232 via USB vs. PPS decoding cards

[Scott Stobbe]

Something like this would make a great NTP server.
https://www.digikey.com/products/en?keywords=P0286-ND

Too bad they didn't include a PTP 1588 capable PHY...

On Tue, Feb 14, 2017 at 12:33 PM, Chris Albertson  wrote:

> Here is a something that could work.  It has a real serial port and you
> could add more ethernet controllers, uses very little power and cost only
> $60.
> www.newegg.com/
>  N82E16813157497_re=j1900-_-13-157-497-_-Product>
>
> There are other boards like this that use the same J1900 CPU.   I'm
> thinking about using this as th machine tool (milling machine) controller.
>
> On Tue, Feb 14, 2017 at 4:26 AM, Bob Camp  wrote:
>
> > Hi
> >
> > A direct port might be a +/- 100 ns sort of thing most of the time and a
> > +/-10 us
> > thing every so often under some OS’s. Most desktop operating systems are
> > not
> > designed to prioritize random pin interrupts. A dirt cheap MCU coded with
> > a few
> > (hundred) lines of assembly code may be a better option than a typical
> > desktop.
> > Complicating this further is the degree to which some OS’s can be
> directly
> > or
> > indirectly optimized. Install *this* package and it all goes nuts.
> Install
> > that package
> >  and not much happens ….
> >
> > Bob
> >
> > > On Feb 13, 2017, at 11:07 AM, Ruslan Nabioullin  >
> > wrote:
> > >
> > > Hi, generally speaking, what are the performance differences between
> the
> > following: 1. direct RS-232 (i.e., what I believe is a standard PCI card
> > offering RS-232---essentially UARTs interfaced more-or-less directly to
> the
> > PCI bus); 2. RS-232 via USB; 3. PPS decoding PCI cards (which might also
> > have an IRIG input or even an onboard GNSS receiver).
> > >
> > > Thanks in advance,
> > > Ruslan
> > > ___
> > > time-nuts mailing list -- time-nuts@febo.com
> > > To unsubscribe, go to https://www.febo.com/cgi-bin/
> > mailman/listinfo/time-nuts
> > > and follow the instructions there.
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> > mailman/listinfo/time-nuts
> > and follow the instructions there.
> >
>
>
>
> --
>
> Chris Albertson
> Redondo Beach, California
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Vintage Frequency Measurement

[Scott Stobbe]

On Mon, Feb 13, 2017 at 6:41 PM, Bob Camp  wrote:

>
>
> I think what you would find is that it *is* a fairly normal AT cut and the
> data book
> that came with the instrument plotted out the data for the specific
> crystal in
> the device. The usable temperature range was fairly small, so the plot will
> be pretty linear.
>

Attached is a plot of crystal calibrators temperature stability. Span is
roughly 65 degC.

One of the other aspects I think is intriguing is the DC PSRR of a vacuum
tube crystal oscillator. In the case of a bjt based oscillator you have the
C-V relation for depletion capacitance and the base-emitter dynamic
capacitance as a function of collector current. I would suspect that for a
one active device oscillator, tube vs bjt, a tube crystal oscillator would
be less sensitive to small power supply variations (+- 10% ). Which is a
convenient attribute for a poorly/unregulated battery supply in the vacuum
tube case. Unless filament current has an appreciable impact on frequency,
I wouldn't think so...
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Vintage Frequency Measurement

[Scott Stobbe]

On Sun, Feb 12, 2017 at 8:39 AM, Dan Rae <dan...@verizon.net> wrote:

> On 2/11/2017 10:08 PM, Scott Stobbe wrote:
>
>> I was inspired recently coming across a Lampkin 105 frequency meter, as to
>> how  frequency measurement was done before counters.
>>
>> Certainly zero-beating a dial calibrated oscillator, would be one
>> approach.
>>
>>
>> Google BC-221 and you may get some idea of how those worked.  I just wish
> I could find the one hidden in my garage :^)


 To they very long list of individuals who pointed out the BC-221, many
thanks. What a endless treasure trove of great material.

Double thanks, to those who have made commitments to document and share the
history online, such as Alan M, Tom B, Brooke C, and others
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

[time-nuts] Vintage Frequency Measurement

[Scott Stobbe]

I was inspired recently coming across a Lampkin 105 frequency meter, as to
how  frequency measurement was done before counters.

Certainly zero-beating a dial calibrated oscillator, would be one approach.

Is there a standout methodology or instrument predating counters?
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Sub-ps delay line

[Scott Stobbe]

I would also advise you take a look at how well you can maintain your
system impedance, say 50 Ohms. For example, I have seen about 100's ps
phase difference on a 10 MHz reference, using one BNC female-female coupler
versus another, a small part is due to TOF, but most of that is due to
subtle differences in the impedance of each coupler, thus by causing
reflections. The same is true for one cable versus another.

On Tue, Feb 7, 2017 at 11:13 AM, Mattia Rizzi 
wrote:

> Hello,
> I'm looking/designing a sub-ps delay line with very high stability.
> Basically it has microwave requirements on phase matching.
> The main features that such delay line should have are:
> - sub-ps resolution and about 1 ns range
> - High stability, must not drift more than 2ps/year, preferably 1ps/year
> - Temperature coefficient (tempco) below 1 ps/celsius
> - Low phase noise floor, target random jitter below 100 fs RSM from 100Hz
> to 1MHz.
> - flicker noise below -90dBc at 1Hz (100MHz carrier)
> - cheap (below 50 euros) and PCB integrable
> - optional: autocalibration or a way to check calibration health over time
> (checking the oscillation frequency of the delay line connected as loop?)
>
> Operating conditions: The delay line will be used for RF distribution,
> where the clock signals (100-200MHz) must stay in +/- 10 ps error window.
> Since timing jitter (wander) is 1.6ps RMS, the delay line must be very
> accurate, with maximum of +/- 3ps of delay uncertainty. The delay line is
> used to phase-match the  clock outputs at factory, hence will not be
> anymore modified (or for only fine corrections, tens of picoseconds). The
> factory calibration compensates for the delay line and PCB
> process/production variations. The boards will operate at almost same
> temperature and humidity levels over years of continuous running.
>
> Circuit constrains: The delay line is fed with an AC-coupled LVPECL clock
> output (only P output used) and should provides a single-ended AC clock
> output signal.
>
> Indeed, no commercial chip fits into these requirements.
>
> My idea is to use an RC filter to delay the input clock signal and then to
> restore the clock edges with a LTC6957-1 (LVPECL outputs).
> The RC filter would be realized using a varactor (Skyworks SMVA1470-004LF)
> and a 16-bit DAC to control the voltage bias (+ stable voltage reference).
> I already checked the values, and sub-ps resolution seems easily
> achievable. The solution requires a factory calibration due to the
> non-linear behavior of the varactor, but since I only need small
> adjustments, this is not a problem.
>
> The problem is to guarantee the calibration over years of operation.
> Since a femtofarad parasitic capacitance can change the delay, I already
> thought about protecting the delay line with some kind of resin (Epoxy?)
> and/or a RF cage to protect it from dirt and moisture.
> One of the issue is aging. I derived a typical varactor aging from VCTCXO
> oscillators (no varactor manufacturer knows the effects of aging on its
> products, apparently) and it's still good. But the aging of LTC6957 is not
> known.
> Is the PCB fabrication using microwave requirements on the dielectric fine?
>
> Based on your experience, do you think that such delay line can respect the
> requirements listed above, especially stability?
> Am I missing something?
> Thank you!
>
> cheers,
> Mattia Rizzi
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Dropbox is cool, but...

[Scott Stobbe]

I've had similar results with LTSpice, by default it tosses the simulations
results to the current working directory. Fortunately, you can tell LTSpice
to use a specific temp folder for simulation results.

On Sun, Feb 5, 2017 at 7:42 PM, Didier Juges  wrote:

> Yes, I noticed that before.
> I have a number of tools that don't like running off a Dropbox folder,
> including several software development tools for starter. Too many files
> opened at the same time.
> Don't assume that because it looks like a normal folder, it works like one,
> even though for many things, it does work remarkably well.
>
> On Feb 5, 2017 2:32 PM, "John Ackermann N8UR"  wrote:
>
> > So I was clever and decided to log some PPS data to a folder within my
> > "Dropbox" folder.  Strange results followed... the whole system just
> bogged
> > down, and even fairly slow serial data dropped characters.
> >
> > It turns out that the culprit was the Dropbox daemon continuously trying
> > to sync the file as it changed every second.  It didn't manifest as CPU
> > overload or anything obvious; the problem was apparently thrashing in the
> > I/O system.  Once I started dumping the data to a "normal" directory, the
> > problem went away.  (This was on Linux, by the way).
> >
> > So, a lesson learned -- don't stream unbuffered data, even at a low rate,
> > into a sync'd folder!
> >
> > John
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/m
> > ailman/listinfo/time-nuts
> > and follow the instructions there.
> >
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Distribution amplifier (again!) - now mostly ok but has gain peaking

[Scott Stobbe]

Nice project. The gain peaking is more than likely from your high speed
opamp. Parasitic capacitance on the inverting terminal from routing and the
input capacitance of the opamp itself, adds another pole to your opamps
loopgain, burning phase margin.

A small compensation cap across the top leg of your feedback divider, would
boost your phase margin.

Actually, Jim Williams has a monster app note, N47 dedicated to high speed
amplifiers :)
On Sat, Jan 28, 2017 at 7:10 AM Anders Wallin 
wrote:

> Hi all,
> I've been tinkering with another distribution amplifier design and made
> some measurements earlier this week.
> The goal is roughly 1:8 fan-out, gain of 0 dB, for good quality (Cs, maser,
> OCXO) 5 or 10 MHz signals in the range of maybe +0 dBm to +15 dBm - in a 1U
> form-factor.
>
> Earlier I made an SMD version of the TADD-1 design [1] which showed about
> -156 dBc/Hz far-out phase-noise but was quite sensitive to external noise
> and required 12VDC power from a lead-acid battery as well as shielding in
> aluminium foil for a 'quiet' PN-spectrum.
> I then did some SPICE simulations [2] (never trust them without testing ;)
> which indicated ADA4899 would be a good op-amp. In practice the
> slew-rate/distortion was limiting and the AD4899 version didn't show better
> PN.
>
> This new version is inspired by looking inside a 6502[3] - and in the
> mean-time I also measured and Ettus Octoclock [4] - but its performance
> isn't so exciting..
>
> My current design is now here: https://goo.gl/photos/WB8fYd4jzba7nXH18
> So far my observations are:
> - phase noise around -162 dBc/Hz at 10 MHz
> - nice quiet PN-spectrum when unshielded and powered from lab-supplies
> - this probably means the supply-section with common-mode choke, BNX025
> filter and LT1963/LT3015 is working OK. I should probably build a
> 10Hz-100kHz LNA (e.g. [5]) to verify. I've used 2k@100MHz ferrites a lot
> and an RC-filter on all supply pins - maybe overkill?
> - an undesired feature is gain-peaking which increases from output ch1 to
> ch8, shown here: https://goo.gl/photos/6QkoKakSPDdT7Acj7
> I tried to improve it a bit by adding a 100pF cap at the start of the long
> trace that feeds the output stages, but some gain-peaking still remains:
> https://goo.gl/photos/qrkLzZ21ptcHxFsw6
> - reverse isolation around 120 dB
> - channel-to-channel isolation around 80 dB
> - at 10MHz 1dB compression between +14 and +15 dBm
> - IP3 perhaps +27dBm to +30 dBm.
>
> Any ideas on how to deal with the long 'feeder-trace' that seems to be the
> cause of the gain-peaking?
> Anyway if not used at 100MHz perhaps my next version will have reduced BW
> where the feeder-trace is not an issue..
> Another issue is that the voltage regulators get quite hot when fed at
> +/-12V and producing +/-6V. They should probably be positioned as far away
> from the input/output amps and thermally disconnected if possible. I have a
> +/-12V AC/DC brick on order - but a DIY linear PSU producing e.g. +/-8VDC
> for the regulators might be better.
>
> The picture gallery also shows a pulse distribution amp for 1PPS. It has an
> LT1711 comparator feeding an 74AC14 buffer with length-matched traces to
> 74AC04's at the outputs. So far my length-matching didn't give zero
> output-skew between the outputs - I see around 150-200ps skew which I tried
> to tune a bit with wires and 0R resistors - without very much success.. any
> ideas for improving this - or just leave it at 200ps skew?
>
> cheers,
> Anders
>
> [1]
>
> http://www.anderswallin.net/2015/12/frequency-distribution-amplifier-first-tests/
> [2]
>
> http://www.anderswallin.net/2015/12/frequency-distribution-amplifier-v2-simulations/
> [3]
>
> http://www.anderswallin.net/2016/02/symmetricom-6502-distribution-amplifier/
> [4]
> http://www.anderswallin.net/2016/09/ettus-octoclock-distribution-amplifier/
> [5] http://cds.linear.com/docs/en/application-note/an83f.pdf
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal effects on cables

[Scott Stobbe]

A google search for digiwave turned up cheap coax at walmart. I don't know
how well copper clad steel holds up in the kHz, low MHz.

On Mon, Jan 23, 2017 at 5:01 PM, Bob Camp  wrote:

> Hi
>
> Back a long time ago the people I was working with spent time looking
> at the impedance of a variety of coax cables. The data they came up
> with on some varieties of cable would suggest that cable is not an outlier…
>
> Bob
>
>
> > On Jan 23, 2017, at 3:08 PM, jimlux  wrote:
> >
> > On 1/23/17 9:16 AM, Bob Camp wrote:
> >> Hi
> >>
> >> Nothing is ever simple if you dig deep enough:
> >>
> >> http://www.jensign.com/RG58U/
> >>
> >> http://www.cliftonlaboratories.com/the_curious_case_of.htm
> >>
> >
> > here's my guess on the curious case cable: it was mismarked by accident
> at the factory (e.g. 93 ohm coax, marked as 50 because someone forgot to
> push the button on the marking machine), and dumped in the scrap bin, and
> then sold by someone who bought the surplus and just read the markings.
> >
> > Or, even sold as a whole spool, likely at a discount, to a
> "knowledgeable buyer" who was willing to take it, mismarkings and all.
> >
> >
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] purpose of time of day display units

[Scott Stobbe]

There is a lot to be said about a tool that just works. In the advent of a
piece of gear failing whether that be a firmware bug or a cooling fan, or
more severe, having a known diagnostic tool during that time is priceless
(well maybe not priceless but extremely nice to have).

On Sun, Jan 22, 2017 at 1:31 AM, Ruslan Nabioullin 
wrote:

> Hi, looking at pictures of various time metrology equipment setups for
> best practices and inspiration, I have commonly seen time of day display
> unit(s) installed in racks containing processing or time transfer
> equipment, e.g., http://www.xyht.com/wp-content/uploads/2015/05/Powers_
> Master_Clock.jpg. All that these units do is merely display the time of
> day and sometimes the date, typically by means of seven segment LED
> displays, of the time code inputted to them (typically IRIG-B, I'm
> guessing).  Any ideas why such a unit is necessary when one can simply look
> at the time displayed by timing receivers and time code generators (and
> even some standards), and the interface of some fusor, defined in this
> context as a system which performs timing data fusion (by implementing a
> paper clock or a more primitive algorithm) and timekeeping, either by means
> of a direct shell, or via something like NTP?
>
> Thanks in advance,
> Ruslan
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/m
> ailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] 10MHz to 25MHz

[Scott Stobbe]

Is there any reason why you wouldn't be able to run the same drive level on
say the fifth overtone versus the fundamental? I would guess at 100 MHz it
may be 3rd or 5th, or are they fundamental?

The comments one drivelevel are simply based on snr, larger signal with
same noise, better snr

On Thu, Jan 19, 2017 at 7:06 PM Bob Camp <kb...@n1k.org> wrote:

> Hi
>
>
>
>
>
> > On Jan 19, 2017, at 3:03 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
>
> >
>
> > Wouldn't crystal drive level be one of the important specifications for
> far
>
> > out phase noise?
>
>
>
> It would, but you can get the same floor at 10 MHz as you can get at 100
> MHz.
>
>
>
> Bob
>
>
>
> >
>
> > On Thu, Jan 19, 2017 at 1:33 PM, Bob Camp <kb...@n1k.org> wrote:
>
> >
>
> >> HI
>
> >>
>
> >> A lot of your evaluation of the term “better” will depend on your
> intended
>
> >> use. One of the limits on phase noise
>
> >> is the thermal noise floor. Because of that, starting at a higher
>
> >> frequency will always give you an edge on broadband
>
> >> phase noise. ADEV / short term stability is linked to the Q of your
>
> >> resonator. In a quartz crystal, maximum Q is
>
> >> roughly proportional to frequency. The other limit on Q is blank
> geometry
>
> >> (size). One other limit is practicality -
>
> >> is a $250,000 OCXO that is 1 cubic meter in size appropriate for your
>
> >> application? The answer to that one is
>
> >> universally - NO :) Somewhere along the line of larger size and cost,
>
> >> other technologies make more sense.
>
> >>
>
> >> So, if better = phase noise floor, 100 MHz is better than 10 MHz. If
>
> >> better = ADEV, 5 MHz in a large package is
>
> >> likely better than 100 MHz. Indeed these are only two variables. There
> are
>
> >> *many* others you could look at.
>
> >>
>
> >> Lots of fun
>
> >>
>
> >> Bob
>
> >>
>
> >>
>
> >>
>
> >>> On Jan 19, 2017, at 7:13 AM, Charles Steinmetz <csteinm...@yandex.com>
>
> >> wrote:
>
> >>>
>
> >>> Chris wrote:
>
> >>>
>
> >>>> I have always wondered why we build our "standard" with such a low
>
> >>>> frequency.   Why not a 100MHz GPSDO?   Why 10MHz
>
> >>>
>
> >>> Quartz crystals work better at lower frequencies, predominantly because
>
> >> they have higher Q.  10MHz was chosen because it is low enough for
>
> >> excellent performance but high enough to be directly useful (since an
>
> >> accident of biology gave us ten fingers, we've created a base-10 world
> and
>
> >> powers of 10 are favored in almost everything).
>
> >>>
>
> >>> In prior times, 5MHz crystals held this position, and before that,
>
> >> 1MHz.  There is a good argument even today that the best 2.5MHz or 5MHz
>
> >> crystals are better than the best 10MHz crystals, but not by enough to
> make
>
> >> 2.5MHz or 5MHz standards popular any longer.
>
> >>>
>
> >>> One lonely data point, which proves nothing:  My best crystal
> oscillator
>
> >> is a Symmetricom clone of the double-oven HP 10811s (it came out of an
> HP
>
> >> GPSDO, so apparently HP at one time used them interchangeably with the
>
> >> 10811).  That OCXO uses a 5MHz crystal and a frequency doubler to
> produce
>
> >> its 10MHz output.
>
> >>>
>
> >>> Best Regards,
>
> >>>
>
> >>> Charles
>
> >>>
>
> >>
>
> >> ___
>
> >> time-nuts mailing list -- time-nuts@febo.com
>
> >> To unsubscribe, go to https://www.febo.com/cgi-bin/
>
> >> mailman/listinfo/time-nuts
>
> >> and follow the instructions there.
>
> >>
>
> > ___
>
> > time-nuts mailing list -- time-nuts@febo.com
>
> > To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
> > and follow the instructions there.
>
>
>
> ___
>
> time-nuts mailing list -- time-nuts@febo.com
>
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
> and follow the instructions there.
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] 10MHz to 25MHz

[Scott Stobbe]

Wouldn't crystal drive level be one of the important specifications for far
out phase noise?

On Thu, Jan 19, 2017 at 1:33 PM, Bob Camp  wrote:

> HI
>
> A lot of your evaluation of the term “better” will depend on your intended
> use. One of the limits on phase noise
> is the thermal noise floor. Because of that, starting at a higher
> frequency will always give you an edge on broadband
> phase noise. ADEV / short term stability is linked to the Q of your
> resonator. In a quartz crystal, maximum Q is
> roughly proportional to frequency. The other limit on Q is blank geometry
> (size). One other limit is practicality -
> is a $250,000 OCXO that is 1 cubic meter in size appropriate for your
> application? The answer to that one is
> universally - NO :) Somewhere along the line of larger size and cost,
> other technologies make more sense.
>
> So, if better = phase noise floor, 100 MHz is better than 10 MHz. If
> better = ADEV, 5 MHz in a large package is
> likely better than 100 MHz. Indeed these are only two variables. There are
> *many* others you could look at.
>
> Lots of fun
>
> Bob
>
>
>
> > On Jan 19, 2017, at 7:13 AM, Charles Steinmetz 
> wrote:
> >
> > Chris wrote:
> >
> >> I have always wondered why we build our "standard" with such a low
> >> frequency.   Why not a 100MHz GPSDO?   Why 10MHz
> >
> > Quartz crystals work better at lower frequencies, predominantly because
> they have higher Q.  10MHz was chosen because it is low enough for
> excellent performance but high enough to be directly useful (since an
> accident of biology gave us ten fingers, we've created a base-10 world and
> powers of 10 are favored in almost everything).
> >
> > In prior times, 5MHz crystals held this position, and before that,
> 1MHz.  There is a good argument even today that the best 2.5MHz or 5MHz
> crystals are better than the best 10MHz crystals, but not by enough to make
> 2.5MHz or 5MHz standards popular any longer.
> >
> > One lonely data point, which proves nothing:  My best crystal oscillator
> is a Symmetricom clone of the double-oven HP 10811s (it came out of an HP
> GPSDO, so apparently HP at one time used them interchangeably with the
> 10811).  That OCXO uses a 5MHz crystal and a frequency doubler to produce
> its 10MHz output.
> >
> > Best Regards,
> >
> > Charles
> >
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thunderbolt Harmonics

[Scott Stobbe]

Harmonic traps are another avenue to explore since the frequency is "fixed"
at well below ppm. Which leaves the fundamental untouched.

On Wed, Jan 18, 2017 at 5:21 AM, Charles Steinmetz 
wrote:

> Rhys wrote:
>
> I was looking at the output of my Trimble Thunderbolt GPSDO and was rather
>> surprised to see really "loud" harmonics in there. ~ 60dB down from the
>> 10Mhz signal.
>>
>
> Welcome to the world of RF.  Loudest harmonic at ~ -60dBc (dB with respect
> to carrier) is actually pretty good for a commercial product. Very few
> distribution amplifiers do this well.  For that matter, many good
> laboratory RF generators are specified with harmonics only below -35 to -45
> dBc.  We do not generally expect RF sources or amplifiers to get down to
> the -80 to -90 dBc range (although amplifiers with harmonics < -80dBc at
> 10MHz/1Vrms/50 ohms are possible), and certainly not the -100 to -120dBc
> that we expect from high fidelity audio sources and amplifiers.
>
> Even harmonics (which make the carrier asymmetrical) can cause phase
> errors that are harmful in high-precision systems [1], so I am a vocal
> supporter of distribution amplifiers with harmonics < -80dBc.
>
> Note that cleaning up the Tbolt output to < -80dBc would probably require
> a crystal filter (a filter with a sharp corner very close to 10MHz, in any
> case), which means its phase response changes very rapidly with the filter
> frequency.  Sharp filters shift frequency with temperature, which causes
> temperature-dependent phase shifts.  Unless the filter is maintained in an
> isothermal environment (like a good oven), this can cause problems in
> sensitive applications.
>
> Best regards,
>
> Charles
>
>
> [1]   F.L. Walls (NIST), F.G. Ascarrunz (SpectraDynamics), The Effect of
> Harmonic Distortion on Phase Errors in Frequency Distribution and Synthesis
> (year unknown, probably late '90s).
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/m
> ailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] time-nuts Digest, Vol 150, Issue 36

[Scott Stobbe]

As an aside, some materials can change their magnetic properties after cold
working/forming like stainless steel for example.

On Tue, Jan 17, 2017 at 10:17 AM, Bob Camp <kb...@n1k.org> wrote:

> Hi
>
> At least at the “it stops working” level, Rb’s are not as sensitive to
> residual magnetic fields as what you
> describe on H-masers. The scrap guys routinely mangle the shielding around
> small Rb’s and the parts
> still work when we receive them.
>
> On another level, the answer is (of course) yes. Any residual field  will
> have an effect on a magnetically tuned
> frequency source. That would definitely drive the material selection for
> anything in or near the physics package.
>
> Bob
>
> > On Jan 17, 2017, at 9:53 AM, John Ponsonby <jebponso...@gmail.com>
> wrote:
> >
> > Re: Low Cost Temperature sensor
> > The ZNI1000 sensor is based on nickel. As such it must be ferromagnetic.
> It would thus be absolutely unacceptable in an H-maser where every single
> small item inside the magnetic shields must be tested for residual
> ferromagnetism. If ferromagnetism is detectable the item is unacceptable.
> This applies to every small screw, thermistor, varactor etc. Unfortunately
> manufacturers of thermistors etc don't state what the wires are made of and
> it may vary between batches of the same nominal item. This is one of the
> problems with making H-masers. Though nominally made of copper and zinc,
> common brass often shows residual ferro magnetism because it is "recovered"
> metal and is  contaminated with iron. That brass must never be used in a
> magnetic instrument have been known for a very long time. I don't know
> off-hand if Caesium or Rubidium standards are as sensitive in this regard
> as H-masers. I guess I could find out from Vanier and Audoin.
> > John P
> >
> >
> > On 17 Jan 2017, at 14:14, time-nuts-requ...@febo.com wrote:
> >
> >> Send time-nuts mailing list submissions to
> >>  time-nuts@febo.com
> >>
> >> To subscribe or unsubscribe via the World Wide Web, visit
> >>  https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> >> or, via email, send a message with subject or body 'help' to
> >>  time-nuts-requ...@febo.com
> >>
> >> You can reach the person managing the list at
> >>  time-nuts-ow...@febo.com
> >>
> >> When replying, please edit your Subject line so it is more specific
> >> than "Re: Contents of time-nuts digest..."
> >>
> >>
> >> Today's Topics:
> >>
> >>  1. Re: Looking for GPS module (Exactime ET6000/Datum 9390-6000)
> >> (ziggy9+time-n...@pumpkinbrook.com)
> >>  2. Re: wifi with time sync (David)
> >>  3. Re: wifi with time sync (David)
> >>  4. Looking for GPS module (Exactime ET6000/Datum 9390-6000
> >> (Mark Sims)
> >>  5. HP 5061A/B Cesium tube conundrum (cdel...@juno.com)
> >>  6. Re: HP 5061A/B Cesium tube conundrum (paul swed)
> >>  7. Fluke/Pendulum Counters - Rubidium Timebase (Ed Palmer)
> >>  8. Re: Fluke/Pendulum Counters - Rubidium Timebase (Magnus Danielson)
> >>  9. Looking for GPS module (Exactime ET6000/Datum 9390-6000)
> >> (Mark Sims)
> >> 10. How to create a super Rb standard (Perry Sandeen)
> >> 11. Low CostTemperature sensor (Perry Sandeen)
> >> 12. Who has a hm H Maser? (Perry Sandeen)
> >> 13. Who has a hm H Maser? (Perry Sandeen)
> >> 14. TICC update? (Scott Newell)
> >> 15. Re: Low CostTemperature sensor (Bill Hawkins)
> >> 16. Re: Looking for GPS module (Exactime ET6000/Datum 9390-6000)
> >> (ziggy9+time-n...@pumpkinbrook.com)
> >> 17. Re: Low CostTemperature sensor (Jason Ball)
> >> 18. Re: Low CostTemperature sensor (Tom Miller)
> >> 19. Re: Low CostTemperature sensor (Charles Steinmetz)
> >> 20. Re: How to create a super Rb standard (Bob Camp)
> >> 21. Re: Low CostTemperature sensor (Scott Stobbe)
> >> 22. Re: TICC update? (John Ackermann N8UR)
> >>
> >>
> >> --
> >>
> >> Message: 1
> >> Date: Mon, 16 Jan 2017 13:17:32 -0500
> >> From: ziggy9+time-n...@pumpkinbrook.com
> >> To: Discussion of precise time and frequency measurement
> >>  <time-nuts@febo.com>
> >> Subject: Re: [time-nuts] Looking for GPS module (Exactime ET6000/Datum
> >>  9390-6000)
> >> Message-ID: <de337749-1b3f-e367-06b0-03b0f9c46...@pumpkinbrook.com>
> >> Content-Type: text/plain; charset=utf

Re: [time-nuts] Low CostTemperature sensor

[Scott Stobbe]

 Thermometry based on Diode leakage current wouldn't be impossible I
suppose, you might loose some hair in the process.

The signal levels on the opamp are goofed too.

On Tue, Jan 17, 2017 at 4:19 AM Charles Steinmetz 
wrote:

> Tom wrote:
>
>
>
> > That article has a major error. Anyone know what it is?
>
>
>
> Well, the author says the reverse current of a diode is "directly"
>
> proportional to temperature.  This could suggest that he means the
>
> relationship is linear (the relationship is actually exponential with
>
> absolute temperature).  But that's not really an *error* -- just sloppy.
>
>   "Direct" does not necessarily imply "linear."  An exponential
>
> relationship is "direct" in the sense that it is what mathematicians
>
> call "injective" (every temperature corresponds to exactly one value of
>
> reverse current).
>
>
>
> Then, in discussing the LM95235, he says that it can use the
>
> "collector-emitter junction diode" of a transistor as the sense element.
>
>   Of course, a bipolar transistor has no collector-emitter junction.
>
> His diagram correctly shows a diode-connected NPN operating in the
>
> active region (forward biased, not reverse biased as the rest of his
>
> article discusses) as the sensor for the LM95235.
>
>
>
> Are any of these what you had in mind, or is there more?
>
>
>
> Charles
>
>
>
> ___
>
> time-nuts mailing list -- time-nuts@febo.com
>
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
> and follow the instructions there.
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] time transfer over wifi

[Scott Stobbe]

Pretty much every wifi transceiver is adc sampled so the frames are
"timestamped", but the adc sample time may not get pushed up.

The rtt/tof for the large umbrella of localization applications, I would
imagine will be impented even farther back in the rx chain.

On Sun, Jan 15, 2017 at 11:42 AM jimlux  wrote:

> On 1/15/17 8:26 AM, Bob Camp wrote:
>
> > Hi
>
> >
>
> > Ok, that’s a pretty good paper. At least it shows data and digs into the
> details.
>
> > It also would lead one to believe that a “Time Nuts” grade sync system
> might
>
> > be a hackable sort of thing …… hmmm…..Given how highly integrated these
>
> > WiFi chip sets have become, that probably is a fantasy.
>
> >
>
>
>
> maybe, maybe not..
>
>
>
> They might have implemented the time stamping feature (like in that WiFi
>
> Arduino thing) but the details are poorly documented, and of course,
>
> everyone is different.
>
>
>
> But it's a start.
>
>
>
> and 802.11v appears to be the "standard" for how to do it in a
>
> "standards compliant" way.
>
>
>
>
>
>
>
> ___
>
> time-nuts mailing list -- time-nuts@febo.com
>
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
> and follow the instructions there.
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] time transfer over wifi

[Scott Stobbe]

Here is a ti app note with timestamping hardware wl8 but ordinary ap's with
no special protocol just timestamping the beacon frame.

http://www.ti.com/lit/an/swaa162a/swaa162a.pdf

On Sun, Jan 15, 2017 at 10:06 AM jimlux  wrote:

> Returning to the OP
>
> "A TimeSync certification program will appear later this year, but
>
> semiconductor firms will have to create new Wi-Fi chips including the
>
> feature."
>
>
>
> so this "new thing" will be hardware of some TBD form.
>
> https://www.wi-fi.org/discover-wi-fi/wi-fi-timesync
>
>
>
>
>
>
>
> But more interesting to time-nuts, I think, is how do you do it without
>
> the new hardware.
>
>
>
> http://www.cse.msu.edu/~glxing/docs/WizSync.pdf
>
> says, in part:
>
>
>
> 802.11  requires  all  APs  to  broadcast  periodic beacon frames that
>
> carry important management information (e.g., supported  rates  and
>
> security  settings).  The  default  beacon period is 102.4 ms, which is
>
> rarely changed on production APs. ...However, as defined in 802.11,
>
> whether a  beacon  frame  is  delayed  or  not,  the  subsequent  beacon
>
> frame shall always  be scheduled at the undelayed  nominal beacon interval.
>
>
>
> so this is the "use a 1pps, but throw out outliers" kind of strategy...
>
>
>
> And there would need to be some sort of measurement of the AP's timing
>
> error - they make the assumption that the timing of the beacons is
>
> driven by a clock with max 25ppm error (as required by the 802.11 std),
>
> although they've measured <5ppm normally
>
>
>
> Ultimately, they got on the order of 0.1 0.2 ms.
>
>
>
>
>
> That's a few orders of magnitude worse than "microsecond", but it's also
>
> an interesting read.
>
>
>
>
>
>
>
> an older presentation (2006) might be useful
>
>
> http://www.ieee802.org/1/files/public/docs2006/avb-stanton-wifi-timesync-intro-060613.pdf
>
>
>
> discusses 802.11v
>
>
>
> there's been a lot of stuff on time sync/distribution over 802.11 links
>
> for the last decade.. maybe this CES announcement is more about "we at
>
> WiFi alliance are ready to market it".   Has anyone gone through the
>
> 802.11 standards list recently?  It might well be that the standard is
>
> already there.
>
>
>
> 802.11aa says "Amendment 2: MAC Enhancements for Robust Audio Video
>
> Streaming" in the description...   although that might just be things
>
> like QoS and access control-digital rights management
>
>
>
> ___
>
> time-nuts mailing list -- time-nuts@febo.com
>
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
> and follow the instructions there.
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] wifi with time sync

[Scott Stobbe]

Yes, it will be interesting to see how well wifi rtt/tof does indoors with
plenty of multipath. But for sure sub microsecond.

On Sat, Jan 14, 2017 at 6:32 PM Bob Camp <kb...@n1k.org> wrote:

> Hi
>
>
>
>
>
> > On Jan 14, 2017, at 5:29 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
>
> >
>
> > I don't think wifi is ever going to be a real-time system, as it shares
> the
>
> > ether with all other ISM devices. That said even 1 ms of variation is
> still
>
> > 4 orders of magnitude greater than the actual time of flight.
>
> >
>
> > The precision time aspect will most certainly be done in hardware, even
> if
>
> > it's just as simple as a timestamp of receiving the beacon frame.
>
>
>
> My concern *is* that it’s going to be like 1588 in that respect. Off we
> all have
>
> to buy new time stamping hardware. Until that’s all up and running
>
> you don’t get the new timing stuff. Based on what I see, there’s not a lot
>
> of hope for it otherwise.
>
>
>
> Bob
>
>
>
> >
>
> > On Sat, Jan 14, 2017 at 3:35 PM, Bob Camp <kb...@n1k.org> wrote:
>
> >
>
> >> Hi
>
>
>
> ___
>
> time-nuts mailing list -- time-nuts@febo.com
>
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
> and follow the instructions there.
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] wifi with time sync

[Scott Stobbe]

I don't think wifi is ever going to be a real-time system, as it shares the
ether with all other ISM devices. That said even 1 ms of variation is still
4 orders of magnitude greater than the actual time of flight.

The precision time aspect will most certainly be done in hardware, even if
it's just as simple as a timestamp of receiving the beacon frame.

On Sat, Jan 14, 2017 at 3:35 PM, Bob Camp  wrote:

> Hi
>
> Here’s what I am seeing:
>
> 64 bytes from 192.168.2.2: icmp_seq=3700 ttl=64 time=5.025 ms
> 64 bytes from 192.168.2.2: icmp_seq=3701 ttl=64 time=4.579 ms
> 64 bytes from 192.168.2.2: icmp_seq=3702 ttl=64 time=1.511 ms
> 64 bytes from 192.168.2.2: icmp_seq=3703 ttl=64 time=1.601 ms
> 64 bytes from 192.168.2.2: icmp_seq=3704 ttl=64 time=2.370 ms
> 64 bytes from 192.168.2.2: icmp_seq=3705 ttl=64 time=4.376 ms
> 64 bytes from 192.168.2.2: icmp_seq=3706 ttl=64 time=2.503 ms
> 64 bytes from 192.168.2.2: icmp_seq=3707 ttl=64 time=4.923 ms
> 64 bytes from 192.168.2.2: icmp_seq=3708 ttl=64 time=4.458 ms
> 64 bytes from 192.168.2.2: icmp_seq=3709 ttl=64 time=33.322 ms
> 64 bytes from 192.168.2.2: icmp_seq=3710 ttl=64 time=2.006 ms
> 64 bytes from 192.168.2.2: icmp_seq=3711 ttl=64 time=1.750 ms
> 64 bytes from 192.168.2.2: icmp_seq=3712 ttl=64 time=122.948 ms
> 64 bytes from 192.168.2.2: icmp_seq=3713 ttl=64 time=9.869 ms
> 64 bytes from 192.168.2.2: icmp_seq=3714 ttl=64 time=24.545 ms
> 64 bytes from 192.168.2.2: icmp_seq=3715 ttl=64 time=1.944 ms
> 64 bytes from 192.168.2.2: icmp_seq=3716 ttl=64 time=63.656 ms
> 64 bytes from 192.168.2.2: icmp_seq=3717 ttl=64 time=126.056 ms
> 64 bytes from 192.168.2.2: icmp_seq=3718 ttl=64 time=99.767 ms
> 64 bytes from 192.168.2.2: icmp_seq=3719 ttl=64 time=72.922 ms
> 64 bytes from 192.168.2.2: icmp_seq=3720 ttl=64 time=4.168 ms
> 64 bytes from 192.168.2.2: icmp_seq=3721 ttl=64 time=3.995 ms
> 64 bytes from 192.168.2.2: icmp_seq=3722 ttl=64 time=5.065 ms
> 64 bytes from 192.168.2.2: icmp_seq=3723 ttl=64 time=2.609 ms
> 64 bytes from 192.168.2.2: icmp_seq=3724 ttl=64 time=4.355 ms
> 64 bytes from 192.168.2.2: icmp_seq=3725 ttl=64 time=4.979 ms
> 64 bytes from 192.168.2.2: icmp_seq=3726 ttl=64 time=4.551 ms
> 64 bytes from 192.168.2.2: icmp_seq=3727 ttl=64 time=1.315 ms
> 64 bytes from 192.168.2.2: icmp_seq=3728 ttl=64 time=3.747 ms
> 64 bytes from 192.168.2.2: icmp_seq=3729 ttl=64 time=4.426 ms
> 64 bytes from 192.168.2.2: icmp_seq=3730 ttl=64 time=4.243 ms
> 64 bytes from 192.168.2.2: icmp_seq=3731 ttl=64 time=4.202 ms
> 64 bytes from 192.168.2.2: icmp_seq=3732 ttl=64 time=4.382 ms
>
> Each ping is about one second.
>
> A 64 second spacing on the round trip “check signals” would likely
> miss this sort of issue. On the other hand, if you are trying to send
> PPS time info *and* see the same sort of “bump” things are likely
> to go tilt pretty quickly.
>
> The range of the bump can go up to over half a second, but only
> does that rarely. Timing between bumps is in the “hours” range.
>
> Is this the nanoseconds or picoseconds that we normally work in?
> Certainly not. It *is* something that could really mess up time
> transfer via WiFi if (note the if) it applies to other traffic as well.
> There are a lot of people running around trying to move from wired
> LAN’s to full WiFi.
>
> Bob
>
> > On Jan 14, 2017, at 3:08 PM, Hal Murray  wrote:
> >
> >
> > kb...@n1k.org said:
> >> Ok, what I see is that every few hours, I get a “rogue delay” on a
> single
> >> ping. How would NTP help me spot a single transit with a 250 ms round
> trip
> >> and identify the  time it occured? Keep in mind that NTP is going to
> >> throttle back to a very low level of “chat” quite quickly…..
> >
> > If you turn on ntpd's rawstats, it will write an entry for each packet
> > exchange with 4 time stamps.  If you assume the clocks on both systems
> are
> > accurate, you can get the transit times in each direction.  That will
> tell
> > you which direction is having troubles.  That may or may not be useful
> > information.
> >
> > You can make ntpd poll more frequently with maxpoll on the server line.
> I
> > think the normal default min is 64 seconds.  You can get more by using
> more
> > servers.  If that's not fast enough, poke me off list and I'll write a
> hack
> > that will do it faster and/or write the log files in a format you like.
> >
> >
> > --
> > These are my opinions.  I hate spam.
> >
> >
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to 

Re: [time-nuts] Thermal effects on cables --> ADEV

[Scott Stobbe]

I think their advice was to limit the ADEV calculation for some tau to 300
bins. The standard error on estimating the standard deviation is ~ +- 5%
for 200 samples. So loosely speaking in the neighborhood of 100-300 bins
the resulting adev will have an rms uncertainty of roughly 5%. So limiting
the number of bins to 300 for any particular tau you wish to monitor, you
will see the ADEV wonder up and down over time, but if it exceeds say 5
sigma, 25% something is up.

On Fri, Jan 13, 2017 at 12:30 PM, Ole Petter Rønningen <
opronnin...@gmail.com> wrote:

> That IS interesting.. It reads to me that the advice is to keep a "moving
> 300 pt ADEV" when continously monitoring a (pair of) frequency source in
> e.g a VLBI site - the reason for limiting it to 300 pts being that much
> more than that is likely to average out potential issues..
>
> Does that make sense?
>
> > Den 13. jan. 2017 kl. 17.04 skrev Bob Camp :
> >
> > Hi
> >
> > There’s an interesting comment buried down in that paper about limiting
> ADEV to
> > < 300 samples per point. Their objective is apparently to better
> highlight “systematic
> > errors”. I certainly agree that big datasets will swamp this sort of
> thing. I’m not quite
> > sure that I’d recommend ADEV to find these things in the first place. My
> guess is that
> > it’s the only spec they have to call the device good or bad in this case
> …They don’t seem
> > to have Hadamard in their list of variances. If I was going after
> systematics with a deviation,
> > that’s the one I’d use. Of course I probably would not use a
> something-dev in the first place.
> >
> > Bob
> >
> >
> >> On Jan 13, 2017, at 1:52 AM, Ole Petter Ronningen <
> opronnin...@gmail.com> wrote:
> >>
> >> Hi, all
> >>
> >> The question of phase shifts in cables pops up every now and then on
> this
> >> list - I stumbled across a good table of measured phase shifts with
> >> temperature in different cable types in this paper:
> >> http://www.ira.inaf.it/eratec/gothenburg/presentations/ERATEC_2014_
> PresentationWSchaefer.pdf
> >> that I though would be of interest to others.
> >>
> >> A quick summary given below, see pdf for full details. Lots of other
> >> interesting stuff in there also.
> >>
> >> Values in ppm/K, for 10 Mhz except when otherwise stated. (The paper
> gives
> >> values for 5, 10 and 100Mhz)
> >>
> >> Huber-Suhner Multiflex 141: -6
> >> RG-223: -131.9
> >> Semiflex Cable: -11.5
> >> Huber-Suhner: -8.6
> >> Times Microwave LMR-240: -3.4
> >> Times Microwave SFT-205: 7.7
> >> Meggitt 2T693 SiO2: 30.6
> >> Andrew FSJ-1 (@5Mhz): 25
> >> Andrew FSJ-4 (@5Mhz): 10
> >> Andrew LDF-1P-50-42: 2.8
> >> Andrew LDF4-50A: 4.7
> >> Times Microwave TF4FLEX (@100Mhz):6.4
> >> Phasetrack PT210 (@100Mhz): 2
> >>
> >> Ole
> >> ___
> >> time-nuts mailing list -- time-nuts@febo.com
> >> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> >> and follow the instructions there.
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal effects on cables --> ADEV

[Scott Stobbe]

You are certainly justified to be cautious of only using an xDEV for state
of health. I don't know what GPS does for example to mark SV's as healthy
or not healthy.

On Fri, Jan 13, 2017 at 1:11 PM, Bob Camp <kb...@n1k.org> wrote:

> Hi
>
> I do agree with their point that systematics will get buried in giant data
> blocks.
> What I’m not quite as sure of is the utility of even 300 sample blocks to
> spot
> systematic issues.
>
> Bob
>
> > On Jan 13, 2017, at 1:08 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
> >
> > I think you might be overthinking their point, that if you plan to use an
> > xDEV as a measure for state of health, don't use years worth of data.
> > Otherwise it could be days before the xDEV visually changes.
> >
> > On Fri, Jan 13, 2017 at 11:04 AM, Bob Camp <kb...@n1k.org> wrote:
> >
> >> Hi
> >>
> >> There’s an interesting comment buried down in that paper about limiting
> >> ADEV to
> >> < 300 samples per point. Their objective is apparently to better
> highlight
> >> “systematic
> >> errors”. I certainly agree that big datasets will swamp this sort of
> >> thing. I’m not quite
> >> sure that I’d recommend ADEV to find these things in the first place. My
> >> guess is that
> >> it’s the only spec they have to call the device good or bad in this case
> >> …They don’t seem
> >> to have Hadamard in their list of variances. If I was going after
> >> systematics with a deviation,
> >> that’s the one I’d use. Of course I probably would not use a
> something-dev
> >> in the first place.
> >>
> >> Bob
> >>
> >>
> >>> On Jan 13, 2017, at 1:52 AM, Ole Petter Ronningen <
> opronnin...@gmail.com>
> >> wrote:
> >>>
> >>> Hi, all
> >>>
> >>> The question of phase shifts in cables pops up every now and then on
> this
> >>> list - I stumbled across a good table of measured phase shifts with
> >>> temperature in different cable types in this paper:
> >>> http://www.ira.inaf.it/eratec/gothenburg/presentations/ERATEC_2014_
> >> PresentationWSchaefer.pdf
> >>> that I though would be of interest to others.
> >>>
> >>> A quick summary given below, see pdf for full details. Lots of other
> >>> interesting stuff in there also.
> >>>
> >>> Values in ppm/K, for 10 Mhz except when otherwise stated. (The paper
> >> gives
> >>> values for 5, 10 and 100Mhz)
> >>>
> >>> Huber-Suhner Multiflex 141: -6
> >>> RG-223: -131.9
> >>> Semiflex Cable: -11.5
> >>> Huber-Suhner: -8.6
> >>> Times Microwave LMR-240: -3.4
> >>> Times Microwave SFT-205: 7.7
> >>> Meggitt 2T693 SiO2: 30.6
> >>> Andrew FSJ-1 (@5Mhz): 25
> >>> Andrew FSJ-4 (@5Mhz): 10
> >>> Andrew LDF-1P-50-42: 2.8
> >>> Andrew LDF4-50A: 4.7
> >>> Times Microwave TF4FLEX (@100Mhz):6.4
> >>> Phasetrack PT210 (@100Mhz): 2
> >>>
> >>> Ole
> >>> ___
> >>> time-nuts mailing list -- time-nuts@febo.com
> >>> To unsubscribe, go to https://www.febo.com/cgi-bin/
> >> mailman/listinfo/time-nuts
> >>> and follow the instructions there.
> >>
> >> ___
> >> time-nuts mailing list -- time-nuts@febo.com
> >> To unsubscribe, go to https://www.febo.com/cgi-bin/
> >> mailman/listinfo/time-nuts
> >> and follow the instructions there.
> >>
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal effects on cables --> ADEV

[Scott Stobbe]

I think you might be overthinking their point, that if you plan to use an
xDEV as a measure for state of health, don't use years worth of data.
Otherwise it could be days before the xDEV visually changes.

On Fri, Jan 13, 2017 at 11:04 AM, Bob Camp  wrote:

> Hi
>
> There’s an interesting comment buried down in that paper about limiting
> ADEV to
> < 300 samples per point. Their objective is apparently to better highlight
> “systematic
> errors”. I certainly agree that big datasets will swamp this sort of
> thing. I’m not quite
> sure that I’d recommend ADEV to find these things in the first place. My
> guess is that
> it’s the only spec they have to call the device good or bad in this case
> …They don’t seem
> to have Hadamard in their list of variances. If I was going after
> systematics with a deviation,
> that’s the one I’d use. Of course I probably would not use a something-dev
> in the first place.
>
> Bob
>
>
> > On Jan 13, 2017, at 1:52 AM, Ole Petter Ronningen 
> wrote:
> >
> > Hi, all
> >
> > The question of phase shifts in cables pops up every now and then on this
> > list - I stumbled across a good table of measured phase shifts with
> > temperature in different cable types in this paper:
> > http://www.ira.inaf.it/eratec/gothenburg/presentations/ERATEC_2014_
> PresentationWSchaefer.pdf
> > that I though would be of interest to others.
> >
> > A quick summary given below, see pdf for full details. Lots of other
> > interesting stuff in there also.
> >
> > Values in ppm/K, for 10 Mhz except when otherwise stated. (The paper
> gives
> > values for 5, 10 and 100Mhz)
> >
> > Huber-Suhner Multiflex 141: -6
> > RG-223: -131.9
> > Semiflex Cable: -11.5
> > Huber-Suhner: -8.6
> > Times Microwave LMR-240: -3.4
> > Times Microwave SFT-205: 7.7
> > Meggitt 2T693 SiO2: 30.6
> > Andrew FSJ-1 (@5Mhz): 25
> > Andrew FSJ-4 (@5Mhz): 10
> > Andrew LDF-1P-50-42: 2.8
> > Andrew LDF4-50A: 4.7
> > Times Microwave TF4FLEX (@100Mhz):6.4
> > Phasetrack PT210 (@100Mhz): 2
> >
> > Ole
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] General questions about making measurements with time interval counter.

[Scott Stobbe]

The pesudo code for the Adev is quite easy to interpret.

For a frequency record of N samples

For each tau=M samples
Reshape(N/M,M)
Mean
Diff
Rms
End

On Thu, Jan 12, 2017 at 4:11 PM Bob Stewart  wrote:

> Hi Bob,
>
> OK, like Bugs Bunny, I'll venture out on the limb, cut the limb, and see
> whether I fall or the tree falls:
>
> Wouldn't it take 1801 samples to get 18 seconds at 100S tau?  Maybe I
> didn't state that properly, but I think you get my meaning.  Also, I've
> never actually taken the time to look at the formula or the code to see how
> the ADEV is calculated.  But doesn't it use a sliding boxcar type of
> calculation?  Or is that some other *DEV?  My point is that for 1801
> seconds, aren't there a lot more than 18 samples put in the 100S bin?  And
> I've probably stated that incorrectly, too.
>
>
>
> Bob
>
>
>
>
>
>   From: Bob Camp 
>
>  To: Bob Stewart ; Discussion of precise time and
> frequency measurement 
>
>  Sent: Thursday, January 12, 2017 2:38 PM
>
>  Subject: Re: [time-nuts] General questions about making measurements with
> time interval counter.
>
>
>
> Hi
>
>
>
> Keep in mind that when you do 1800 samples at 1 second, that data will
> only meet the
>
> 100 sample requirement out to tau = 18 seconds. Past that you are in the
> “under 100 samples
>
> region”.
>
>
>
> Bob
>
>
>
> > On Jan 12, 2017, at 2:32 PM, Bob Stewart  wrote:
>
> >
>
> > Hi Bob,
>
> > OK, thanks for explaining.  When you and others use highly technical
> terms like "small number of samples" it's not always clear to me what you
> mean.  =)  Ten samples?  That's not enough for anything.  Normally I run at
> least 1800 samples; at least if I plan to share them with someone.
>
> >
>
> > Bob -
>
> > AE6RV.com
>
> >
>
> > GFS GPSDO list:
>
> > groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>
> >
>
> >  From: Bob Camp 
>
> > To: Bob Stewart ; Discussion of precise time and
> frequency measurement 
>
> > Sent: Thursday, January 12, 2017 1:03 PM
>
> > Subject: Re: [time-nuts] General questions about making measurements
> with time interval counter.
>
> >
>
> > Hi
>
> >
>
> > It varies from 5370 to 5370. You see a lot of plots that run out to 10
> samples or less. Anything below 100 samples
>
> > is risky in some senses.
>
> >
>
> > Bob
>
> >
>
> >> On Jan 12, 2017, at 12:25 PM, Bob Stewart  wrote:
>
> >>
>
> >> Hi Bob,
>
> >> OK, what's a small number of data points?  Attached is a screencap of
> captures for 25, 50, 75, 100, and 150 seconds.  Yeah, at 25 seconds, the 1S
> tau is up at 4.56E-11, but it falls pretty quickly.  I will mention that
> this particular 5370 is much better than my other one.  So, maybe this one
> is an exceptional example?
>
> >>
>
> >> Just for grins, I also included a screencap of the phase points.
>
> >>
>
> >> Bob -
>
> >> AE6RV.com
>
> >>
>
> >> GFS GPSDO list:
>
> >> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>
> >>
>
> >>  From: Bob Camp 
>
> >> To: Bob Stewart ; Discussion of precise time and
> frequency measurement 
>
> >> Sent: Thursday, January 12, 2017 11:04 AM
>
> >> Subject: Re: [time-nuts] General questions about making measurements
> with time interval counter.
>
> >>
>
> >> Hi
>
> >>
>
> >> There is a big difference between RMS and single shot. Single shot, the
> 5370 is a very different beast.
>
> >> That’s not a big deal when you have a few thousand readings and it all
> averages down. Unfortunately
>
> >> we all love to do runs with a very small number of points and then draw
> conclusions from them. As the
>
> >> sample size goes down, you no longer have a 2 to 4 x 10^-11 beast, it’s
> more like 5X that.
>
> >>
>
> >> Bob
>
> >>
>
> >>
>
> >>> On Jan 12, 2017, at 11:31 AM, Bob Stewart  wrote:
>
> >>>
>
> >>> Hi Bob,
>
> >>> Normally I see somewhere between 2E-11 and 4E-11 at 1S tau on my
> 5370A, as in the blue trace on the attached plot.  Am I misunderstanding
> your meaning?  Granted, I am clocking the 5370A with a GPSDO, but I believe
> I see about the same thing with the HP10811.  This test was 1PPS vs 1PPS on
> two different units.
>
> >>> The plot also has a test run by Tom, in orange, using his H Maser and
> a Timepod to show how poor the 5370 is compared to the Timepod below about
> 60S tau.  These are essentially apples vs apples tests.
>
> >>>
>
> >>> Bob
>
> >>>
>
> >>>
>
> >>>
>
> >>>
>
> >>>  From: Bob Camp 
>
> >>> To: Discussion of precise time and frequency measurement <
> time-nuts@febo.com>
>
> >>> Sent: Thursday, January 12, 2017 8:27 AM
>
> >>> Subject: Re: [time-nuts] General questions about making measurements
> with time interval counter.
>
> >>>
>
> >>> Hi
>

Re: [time-nuts] σ vs s in ADEV

[Scott Stobbe]

I could be wrong here, but it is my understanding that Allan's pioneering
work was in response to finding a statistic which is convergent to 1/f
noise. Ordinary standard deviation is not convergent to 1/f processes. So I
don't know that trying to compare the two is wise. Disclaimer: I could be
totally wrong, if someone has better grasp on how the allan deviation came
to be, please correct me.

On Wed, Jan 4, 2017 at 3:12 PM, Attila Kinali  wrote:

> Hi,
>
> A small detail caught my eye, when reading a paper that informally
> introduced ADEV. In statistics, when calculating a variance over
> a sample of a population the square-sum is divided by (n-1)(denoted by s in
> statistics) instead of (n) (denoted by σ) in order to account for a small
> bias
> the "standard" variance introduces
> (c.f. https://en.wikipedia.org/wiki/Unbiased_estimation_of_
> standard_deviation )
> In almost all literature I have seen, ADEV is defined using an average,
> i.e. dividing by (n) and very few use (n-1).
>
> My question is two-fold: Why is (n) being used even though it's known
> to be an biased estimator? And why do people not use s when using (n-1)?
>
> Attila Kinali
>
> --
> It is upon moral qualities that a society is ultimately founded. All
> the prosperity and technological sophistication in the world is of no
> use without that foundation.
>  -- Miss Matheson, The Diamond Age, Neil Stephenson
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] FE-5650A option 58 tuning word for 10 MHz output

[Scott Stobbe]

It very well could be. Based on Marks comments, it sounds like the DDS tone
after being squared up is directly driving a 23-bit counter for the 1 PPS
output.

On Mon, Jan 9, 2017 at 12:17 PM, Bob kb8tq <kb...@n1k.org> wrote:

> Hi
>
> In most Rb’s (including the FE 56xx’s) the DDS is mixed with a fixed
> microwave
> frequency signal. The DDS only has to make up “part” of the total offset.
> You get
> roughly a three orders of magnitude improvement because of this. Rick has
> gone
> into all the gory details of why it gets done this way in talking about
> the 5071. It
> is the same thing on an Rb.
>
> So, your basic math is correct about a normal DDS. In this case you are in
> the
> PPT rather than PPB range due to the multiplication.
>
> Bob
>
>
> > On Jan 9, 2017, at 10:40 AM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
> >
> > A 32-bit DDS synthesizing at 1/5 Fs, yields a tuning resolution of ~ 1
> ppb.
> > So, I would imagine a slightly lower frequency is programmed into the DDS
> > and the c-field is trimmed to yield a higher precision. If the new
> > synthesized tone you wish to generate is an integer number of DDS codes
> you
> > could start by assuming the c-field is trimmed to be on frequency, but if
> > the new tone is a fractional number of 32-bit DDS codes you will have to
> > manually trim if you want higher precision.
> >
> > On Mon, Jan 9, 2017 at 4:48 AM, wb6bnq <wb6...@cox.net> wrote:
> >
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] FE-5650A option 58 tuning word for 10 MHz output

[Scott Stobbe]

A 32-bit DDS synthesizing at 1/5 Fs, yields a tuning resolution of ~ 1 ppb.
So, I would imagine a slightly lower frequency is programmed into the DDS
and the c-field is trimmed to yield a higher precision. If the new
synthesized tone you wish to generate is an integer number of DDS codes you
could start by assuming the c-field is trimmed to be on frequency, but if
the new tone is a fractional number of 32-bit DDS codes you will have to
manually trim if you want higher precision.

On Mon, Jan 9, 2017 at 4:48 AM, wb6bnq  wrote:

> Hi Matt,
>
> Well, after rereading Mark’s paragraph in question, I think he did not
> properly develop his complete thoughts.  The first statement about the
> Hydrogen Maser is absolute.  The second statement is the one that is really
> vague.  The third statement is the clue taken with the fact that the first
> sentence states the purpose of being used as a general purpose programmable
> frequency synthesizer.
>
> So the answer is leaving the C-field pot untouched and taking the
> difference between the “R” value and the “needed” input frequency
> associated with the current “F” value to produce the original output
> frequency gives a correction term to be applied to the “R” value to produce
> the value you use to come up with the new “F” value used for determining
> the wanted output signal.
>
> {After thought sentence} The above is not all that clear either, oh well.
> Read on it becomes clearer.
>
> So lets go through the process and see if I can do this without screwing
> up.  The formula for the DDS chip to produce a desired output for a given
> system clock frequency is the following :
>
> FTW (in decimal) = (Desired Output Frequency x 2^n) / SYSCLK
>
> However, the need is to determine what the proper input frequency is to
> produce the 8388608 Hz with the given “F” value as the FTW (Frequency
> Tuning Word).  So the formula is the following:
>
> SYSCLK = (Desired Output Frequency x 2^n) / FTW (in decimal)
>
> In your reported numbers this produces :
> first 2^32 = 4,294,967,296 times desired output of 8,388,608 Hz =
> 36,028,797,018,963,968
>
> SYSCLK = (Desired Output Frequency x 2^n) / FTW
> (in decimal)
> 50,255,055.809934059845495428970822 Hz = 36,028,797,018,963,968 /
> 716918854
>
> So the above 50 MHz number (SYSCLK) is the result of adjusting the C-field
> so the unit is “ON” frequency for the expected 1 Hz output from the
> factory.  This is the SYSCLK value that should be used to find the new “F”
> value for the DDS upon selecting a new output frequency such as 10 MHz (or
> as close as possible without touching the C-field) if that is your wanted
> output value.
>
> Actually, now that I have done the exercise, computing the delta between
> the “R” value and the above 50 MHz makes no sense and serves no purpose.  I
> cannot stress enough.  This is all predicated on not touching the C-field
> adjustment and assuming the 1 Hz signal is precisely on frequency.
>
> This method does not give a lot of confidence as to preciseness.  The real
> value in these Rb units is they have a much lower drift rate than a
> reasonably good quality Quartz oscillator.  Typically less then parts in 10
> to the minus 10th or minus 11th per month.
>
> BillWB6BNQ
>
>
>
>
>
> Mathias Weyland wrote:
>
> On 2017-01-04 10:16, wb6bnq wrote:
>>
>> Hello Bill
>>
>> Thanks for re-iterating over this.
>>
>>
>>  Yes, I do think the outer can covering is a MU-metal shield.  The
>>> bottom plate where the connector is located is not.
>>>
>>
>>
>> That is reassuring thank you!
>>
>>
>>  I know the calculator that comes with Windows XP will produce the
>>> correct mathematical results.  I think the Windows version 7 does the
>>> same.  I do not have Windows 10 and therefore cannot address that
>>> one, if there is one.  Even EXCEL spreadsheet does not do the job
>>> properly.  So use caution with your calculations.
>>>
>>
>>
>> OK noted. The original calculations were done with a calculator that
>> was designed for high precision (in the floating point sense). I did
>> re-run the calculations in windows calculator for kicks, and the
>> result is different, although the difference is too small to have an
>> effect on the integer phase accumulator increment (fingers crossed!)
>>
>>
>>  However, with all that said, it means nothing if you cannot properly
>>> measure the final value against an external standard of greater
>>> accuracy.  Acquiring the equipment to do the external measurements is
>>> where the real cost comes in.
>>>
>>
>>
>> Yes, I think that I am aware of that and I have the opportunity to
>> do that with somebody else's gear. I also understand that I'm supposed
>> to do that on a regular basis.
>>
>>
>>  Hopefully the above helps to clear up your query ?
>>>
>>
>>
>> Yes most of it is clear, thank you. Unfortunately though my original
>> question, i.e. how to incorporate the reported R value into the
>> calculation, is still kind of open. I'm 

Re: [time-nuts] Temperature (environmental) sensors

[Scott Stobbe]

The good news is if the dataloger you get has a mems pressure sensor, you
will have a high precision temperature sensor, whether or not the product
software provides that resolution to you is another matter. In addition to
the piezo-resistive bridge being mechanically sensitive to diaphragm
strain, they are also great thermistors. Which ends up requiring a high
resolution temperature sensor to temperature compensate the bridge readings.

Two related but off target products I would personally recommend is picking
up a used EeePC as a usb data sink. I bought one of the originals back
almost 10 years ago, and I still find it to come in handy once in awhile
versus a rasperry pi, you could likely pick one up for nothing today. A USB
labjack works quite well to tack up a few thermistors to a DUT for logging,
it also has full driver support for linux/windows with a python interface
(other languages too).

On Tue, Jan 3, 2017 at 10:08 AM, Tom Van Baak  wrote:

> I have some high-end temperature and pressure instruments. But for casual
> use in my home and mobile timing lab I use Sparkfun Weather Stations. The
> old URL is:
>
> https://www.sparkfun.com/products/retired/10586
>
> It's USB, talk-only, one reading a second, temperature, pressure, humidity
> -- about as simple as you can get. Perfect for data logging along with
> frequency standards, GPS, counters and such.
>
> But they don't make 'em anymore. My question is what similar
> well-engineered, talk-only, serial or USB, temperature-pressure-humidity
> sensors have you run across and could recommend? Not to be picky bit no
> cheapo 1C or 0.5C sensors; 0.1C or better is ok.
>
> I know it's "easy" to throw one together with an Arduino, but I'm looking
> for something pre-packaged, something that reliably works, out-of-the-box.
> I have backup plans but hope someone on the list knows some products they
> have used and would recommend.
>
> We could extend the discussion to voltage and power monitors too. Or some
> kind of universal sensor TAPR project. But for now, let's just keep it to
> simple air / environmental sensing.
>
> Thanks,
> /tvb
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] WWVB Simulator

[Scott Stobbe]

Fwiw, an xor gate would make a simple bpsk modulator, carrier input A
modulation B, B is low A is buffered, when B is high, carrier is inverted
(180 degree phase shift).

On Mon, Jan 2, 2017 at 7:20 PM David  wrote:

> I see generally how it should work but did not draw out a truth table.
>
>
>
> Why use the 74CB3T3253 instead of a low voltage 4052 variant?
>
>
>
> It think you could buffer just the two references and save two
>
> operational amplifiers and 2 or 4 capacitors.
>
>
>
> On Mon, 2 Jan 2017 21:18:12 + (UTC), you wrote:
>
>
>
> >I have come up with a ridiculously simple WWVB simulator that simulates
> both the AM modulation and the BPSK modulation.
>
> >
>
> >A simple explanation and a link to the core schematic can be found here:
> http://spacetimepro.blogspot.com/2017/01/wwvb-simulator.html
>
> >I'm working on fleshing out the design and producing a PCB. Any comments
> or suggestions before I complete the design are welcome. I will try to
> incorporate useful suggestions in the final design.
>
> >
>
> >Note: I have looked all over the 'net for a similar design or any thing
> hinting at it. I found nothing. Once you see the design it becomes obvious.
>
> >
>
> >What clarified it for me is that I looked at the last diagram in this
> file http://www.wparc.us/presentations/SDR-2-19-2013/Tayloe_mixer_x3a.pdf
> and then plotted out the digital version of what the one cycle of the
> carrier would look like given that the signal is just made up of ones and
> zeros.
>
> >The final design (as far as I have gone) incorporates a synchronizing
> latch so the signals only change at the "zero" of the modulator.
>
> >
>
> >Simon
>
> ___
>
> time-nuts mailing list -- time-nuts@febo.com
>
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
> and follow the instructions there.
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Using GPSDO as a Refrence for Protable Amateur Radio Microwave Operations

[Scott Stobbe]

The other interesting aspect, is that if the transceiver is mobile, even at
a lazy pedestrian walking speed of 1 m/s, the resulting Doppler shift is 3
E-9 deltaF.

On Thu, Dec 22, 2016 at 4:36 PM, Bob Camp <kb...@n1k.org> wrote:

> Hi
>
> For close in phase noise (< 10 Hz) the 10 MHz still wins over the 100 MHz
> after multiplication.
>
> ADEV of the 10 MHz (with or without frequency scale) will be better on the
> higher Q resonator.
> That will always be the low frequency overtone rather than the VHF crystal.
>
> Indeed, a large blank 5 MHz would beat the 10 MHz. It’s a good bet that if
> a 2.5 MHz cold weld
> SC with a 30 mm blank diameter existed, it would beat either one of them
> (Q would be much higher).
> Given the cost of coming up with that part …. not going to happen.
>
> Bob
>
> > On Dec 22, 2016, at 3:26 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
> >
> > Well for the same Q a competing oscillator will still take a 20 dB phase
> > noise increase for every frequency decade you scale up to. If Q*f is
> > approximately constant, you take another 20 dB hit in phase noise from
> > degraded Q, totaling 40 dB/decade. Compared to 20 dB/decade plus the
> noise
> > introduced by the phase detector and loop-filter of the PLL.
> > https://en.wikipedia.org/wiki/Leeson's_equation
> >
> > On Thu, Dec 22, 2016 at 10:53 AM, Attila Kinali <att...@kinali.ch>
> wrote:
> >
> >> On Wed, 21 Dec 2016 18:59:20 -0800
> >> Chris Albertson <albertson.ch...@gmail.com> wrote:
> >>
> >>> Why to people always build 10MHz GPSDOs?   If the use of the GPSDO is
> to
> >>> drive a microwave, why not build a MUCH higher frequency GPSDO.Is
> the
> >>> reason that 10MHz crystals just happen to be very good and there are
> not
> >>> good 100MHz ovenized crystals?  Or for portable use could you not use
> the
> >>> 1PPS signal to discipline a microwave oscillator.
> >>
> >> Short answer:
> >> GPSDOs are mostly about high stability, not about low phase noise.
> >> The 10MHz just happend to be a good compromise on stability, phase noise
> >> and usefulnes.
> >>
> >> Long answer:
> >> A GPSDO has to exhibit good stability up to several 100 s to a few 1000
> s.
> >> This dictates that the OCXO used has to have as high long term stability
> >> as possible. To get there you need an as thick crystal lab as possible.
> >> The lower the frequency and the higher the overtone, the better.
> >> Quartz resonators exhibit a nearly constant Q*f, so in first order
> >> approximation, there is no point in choosing a higher frequency
> >> crystal, as the Q will then decrease and thus increase the phase noise
> >> would have been the same as the increased phase noise of a frequency
> >> multiplier. Of course, frequency multiplication is not exactly perfect
> and
> >> the Q*f is not 100% flat. There is a sweet spot where Q*f is maximal
> >> between
> >> 5MHz and 10MHz. For historical reasons, 10MHz has been deemed the more
> >> useful
> >> value and that's the reason we have a lot of 10MHz OCXO. If you go for
> high
> >> stability oscillators, you will see a lot 5MHz OCXOs being used (for the
> >> increased stability). Of course nobody says that these are the only
> >> frequencies that can be used. For example, for specialized use cases you
> >> will find GPSDOs with "odd" frequencies (like the 30.72MHz/61.44MHz used
> >> for LTE).
> >>
> >> As others have already commented, when using GPSDOs as a frequency
> >> reference
> >> for an GHz link, one would use some high frequency oscillator in the
> lower
> >> 100MHz range (using a BAW quartz) or somewhere between 500MHz and
> 1000MHz
> >> (using an SAW quartz) as a low phase noise reference and upconvert this.
> >> Yes, it is possible to discipline such an oscillator directly using GPS,
> >> but for the sake of stability (see above), design reuse and ease of
> >> building/testing, using an 10MHz input is generally the better solution.
> >> This allows to use any device that can produce an 10MHz signal, like
> >> e.g. an Rb vapor cell standard.
> >>
> >>
> >>Attila Kinali
> >>
> >> --
> >> It is upon moral qualities that a society is ultimately founded. All
> >> the prosperity and technological sophistication in the world is of no
> >> use without that foundation.
> >> -- 

Re: [time-nuts] Using GPSDO as a Refrence for Protable Amateur Radio Microwave Operations

[Scott Stobbe]

Well for the same Q a competing oscillator will still take a 20 dB phase
noise increase for every frequency decade you scale up to. If Q*f is
approximately constant, you take another 20 dB hit in phase noise from
degraded Q, totaling 40 dB/decade. Compared to 20 dB/decade plus the noise
introduced by the phase detector and loop-filter of the PLL.
https://en.wikipedia.org/wiki/Leeson's_equation

On Thu, Dec 22, 2016 at 10:53 AM, Attila Kinali  wrote:

> On Wed, 21 Dec 2016 18:59:20 -0800
> Chris Albertson  wrote:
>
> > Why to people always build 10MHz GPSDOs?   If the use of the GPSDO is to
> > drive a microwave, why not build a MUCH higher frequency GPSDO.Is the
> > reason that 10MHz crystals just happen to be very good and there are not
> > good 100MHz ovenized crystals?  Or for portable use could you not use the
> > 1PPS signal to discipline a microwave oscillator.
>
> Short answer:
> GPSDOs are mostly about high stability, not about low phase noise.
> The 10MHz just happend to be a good compromise on stability, phase noise
> and usefulnes.
>
> Long answer:
> A GPSDO has to exhibit good stability up to several 100 s to a few 1000 s.
> This dictates that the OCXO used has to have as high long term stability
> as possible. To get there you need an as thick crystal lab as possible.
> The lower the frequency and the higher the overtone, the better.
> Quartz resonators exhibit a nearly constant Q*f, so in first order
> approximation, there is no point in choosing a higher frequency
> crystal, as the Q will then decrease and thus increase the phase noise
> would have been the same as the increased phase noise of a frequency
> multiplier. Of course, frequency multiplication is not exactly perfect and
> the Q*f is not 100% flat. There is a sweet spot where Q*f is maximal
> between
> 5MHz and 10MHz. For historical reasons, 10MHz has been deemed the more
> useful
> value and that's the reason we have a lot of 10MHz OCXO. If you go for high
> stability oscillators, you will see a lot 5MHz OCXOs being used (for the
> increased stability). Of course nobody says that these are the only
> frequencies that can be used. For example, for specialized use cases you
> will find GPSDOs with "odd" frequencies (like the 30.72MHz/61.44MHz used
> for LTE).
>
> As others have already commented, when using GPSDOs as a frequency
> reference
> for an GHz link, one would use some high frequency oscillator in the lower
> 100MHz range (using a BAW quartz) or somewhere between 500MHz and 1000MHz
> (using an SAW quartz) as a low phase noise reference and upconvert this.
> Yes, it is possible to discipline such an oscillator directly using GPS,
> but for the sake of stability (see above), design reuse and ease of
> building/testing, using an 10MHz input is generally the better solution.
> This allows to use any device that can produce an 10MHz signal, like
> e.g. an Rb vapor cell standard.
>
>
> Attila Kinali
>
> --
> It is upon moral qualities that a society is ultimately founded. All
> the prosperity and technological sophistication in the world is of no
> use without that foundation.
>  -- Miss Matheson, The Diamond Age, Neil Stephenson
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] u-blox LEAH M8F

[Scott Stobbe]

There was an interesting comment in the Lea-m8f datasheet,
"In strong signal clear-sky applications the best time pulse consistency
between neighbouring receivers is
achieved when using a single GNSS because of the small time offsets between
different GNSS systems. "


On Thu, Dec 22, 2016 at 8:01 AM John Haine 
wrote:

> I was slightly imprecise in my previous posting about ubx timing modules
>
> and a few people have raised queries in the list or by email.
>
>
>
> What I should have said is that this module is designed as a precise
>
> low-noise /frequency /reference so it directly generates the 30.72 MHz
>
> needed in LTE infrastructure from a GPS disciplined VCTCXO.  As I
>
> understand it, the usual 10 MHz output is digitally generated and
>
> jitters, so can't be used as a reference in high performance radio
>
> systems.  One could use a phase-locked synth to multiply up from the 1
>
> PPS signal but given the large ratio needed this is prone to generating
>
> its own noise.  The M8F does it all internally and is therefore
>
> preferred if what you want is a low noise frequency reference, though
>
> its frequency may not be ideal if for example you wanted 10 MHz - you
>
> could get 10.24 MHz by dividing by 3 but 10 MHz is trickier.
>
>
>
> --
>
> John Haine
>
> 9 Parkway
>
> Shudy Camps
>
> Cambridge
>
> CB21 4RQ
>
>
>
> 01799 584773 (T)
>
> 07841 322831 (M)
>
>
>
> ___
>
> time-nuts mailing list -- time-nuts@febo.com
>
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>
> and follow the instructions there.
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] u-blox timing modules

[Scott Stobbe]

The LEA-M8F is a complete GPSDO in module, the TCXO is EFC steered, no
sawtooth needed.

On Wed, Dec 21, 2016 at 8:30 PM, Bob Camp  wrote:

> Hi
>
> Ummm…. because you need a 30.72 MHz reference in a 4G base station rather
> than the
> TCXO that normally is in the uBlox modules. You can then “carry” the
> sawtooth into the
> SDR part of the basestation without a lot of muss and fuss.
>
> Bob
>
> > On Dec 21, 2016, at 5:48 PM, Bob Stewart  wrote:
> >
> > Hi John,
> > Could you tell me why one would use choose an LEA-M8F over an LEA-M8T?
> > Bob - AE6RV
> >  -
> > AE6RV.com
> >
> > GFS GPSDO list:
> > groups.yahoo.com/neo/groups/GFS-GPSDOs/info
> >
> >  From: John Haine 
> > To: time-nuts@febo.com
> > Sent: Wednesday, December 21, 2016 11:34 AM
> > Subject: [time-nuts] u-blox timing modules
> >
> > It's worth noting, in case people don't know, that u-blox have a
> > specific chip & module (the latter being the LEA-M8F) for precision
> > timing, developed for the LTE (4G) base station market. This disciplines
> > a low phase noise 30.72 MHz reference to received satellite signals,
> > GPS/GLONASS/BeiDou.  (30.72 MHz is the standard LTE sampling rate.)
> >
> > John.
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
> >
> >
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] 1PPS users?

[Scott Stobbe]

Part of the reason 1PPS needs to be so clean is because you are
continuously integrating phase noise of the LO (hopefully an OCXO). While
10 uS is pretty trivial off a gps receiver. Without gps, 10 us over 24 hrs
with a plane jane AT-cut crystal subject environmental dynamics becomes
ludicrous. More than likely the short term stability of GPSDO is cleaner
than it needs to be for many applications, but that is so holdover over an
hour or day remains reasonable.

On Sun, Dec 18, 2016 at 6:16 PM, Bob Stewart  wrote:

> One thing I've never really understood is who actually uses the
> high-quality 1PPS output from a GPSDO.  I have spent a lot of time, effort,
> and money on developing my GPSDO without a whole of thought to the user
> base.  It was just a quest for the best result I could obtain with a
> particular technology.  The frequency standard users was a no brainer.
> Everyone who wants a frequency standard eventually understands they need to
> get a GPSDO, or an Rb, or a Cs.  And that's all I thought I had: a good
> frequency standard.  And then Tom prodded me a bit and showed me the
> shortcomings of what I was doing, and I did something about it.  So, if an
> NTP user can get his time fix directly from a noisy receiver, who actually
> needs a time-accurate, low jitter 1PPS pulse?
> Bob - AE6RV
>  -
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Totally unrelated, but..

[Scott Stobbe]

That's a really nice part, like the old LM723 was bring your own pass
element, the 3042 is essentially bring your own reference, I hope that is a
trend that continues. And a 10uF capacitor is pretty quiet at 1KHz plus.

The other neat thing about the 3042 is the error amplifier is run unity
gain with a compensation network. For a typical 12V ldo you loose 20 dB
loopgain and 1 decade of loop bandwidth gaining up 1.2 VRef to 12 V on the
feedback divider. At 12 V I'm sure the 3042 would blow a typical LDO out of
the water in line/load regulation.

On Wed, Dec 7, 2016 at 11:36 PM, jimlux <jim...@earthlink.net> wrote:

> On 12/7/16 4:20 PM, Scott Stobbe wrote:
>
>> Yes, the short hand I like to use is 4 nV*sqrt(R/1000).
>>
>> 2 nV/rthz off a bandgap is pretty darn impressive, that includes a delta
>> vbe gained up ~10x.
>>
>> On Wed, Dec 7, 2016 at 7:03 PM Bob Camp <kb...@n1k.org> wrote:
>>
>> H
>>>
>>
> the cool thing about those parts is that their PSRR extends up to several
> MHz.  A lot of LDOs have good PSRR to kHz.
>
> Someone at LT did a good job on that design.
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/m
> ailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Totally unrelated, but..

[Scott Stobbe]

Yes, the short hand I like to use is 4 nV*sqrt(R/1000).

2 nV/rthz off a bandgap is pretty darn impressive, that includes a delta
vbe gained up ~10x.

On Wed, Dec 7, 2016 at 7:03 PM Bob Camp <kb...@n1k.org> wrote:

> Hi
>
>
>
> Just for reference, KTB in 1 Hz at 50 ohms is roughly 0.9 nV at room
>
> temperature. The previously mentioned 2 nV is equivalent to about 250 ohms.
>
>
>
> Bob
>
>
>
> > On Dec 7, 2016, at 6:58 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
>
> >
>
> > You can buy opamps off the shelf with an input referred noise less than a
>
> > 50 ohm resistor to build up a preamp (of course you can build up
> something
>
> > even lower with discretes, but then it's a time commitment over a basic
>
> > opamp preamp)
>
> >
>
> > Even if your regulator is low noise if it's running with low phase margin
>
> > you can get noise peaking.
>
> >
>
> > On Wed, Dec 7, 2016 at 3:06 PM Van Horn, David <
>
> > david.vanh...@backcountryaccess.com> wrote:
>
> >
>
> >>
>
> >>
>
> >> I replaced the original caps, and I added caps, I substituted good
>
> >> Jonhansen RF caps, and Tanceram caps.
>
> >>
>
> >> No help at all.
>
> >>
>
> >>
>
> >>
>
> >> The layout and routing is as good as I could do, and the only
> improvement
>
> >> I could see possible would be to move one cap closer to the reg.
>
> >>
>
> >> The difference would be less than the tolerance of part placement on the
>
> >> existing pads.
>
> >>
>
> >>
>
> >>
>
> >> Fortunately this isn't a critical piece of equipment, but I want it
>
> >> working RIGHT before I put it back in service.
>
> >>
>
> >> It's a custom receiver for 457 kHz.
>
> >>
>
> >>
>
> >>
>
> >>
>
> >>
>
> >> -Original Message-
>
> >>
>
> >> From: time-nuts [mailto:time-nuts-boun...@febo.com] On Behalf Of
> Graham /
>
> >> KE9H
>
> >>
>
> >> Sent: Wednesday, December 7, 2016 10:41 AM
>
> >>
>
> >> To: Discussion of precise time and frequency measurement
>
> >>
>
> >> Subject: Re: [time-nuts] Totally unrelated, but..
>
> >>
>
> >>
>
> >>
>
> >> Remember that the internal Voltage reference in the original three
>
> >> terminal regulator designs is a Zener.
>
> >>
>
> >> (Zeners are also useful as RF white noise sources.) The regulator is
>
> >> generally an amplifier with DC feedback.
>
> >>
>
> >> If you look at the application notes on the early regulators, they
> require
>
> >> capacitors to ground on both the input and outputs.
>
> >>
>
> >> If these capacitors are missing, or too low in value, or not good
>
> >> capacitors at RF frequencies, then the Zener noise is amplified by the
>
> >> regulator amplifier and pushed out the output port.
>
> >>
>
> >> I would experiment by putting a good ceramic 0.1uF cap to ground, right
> at
>
> >> the regulator output port.
>
> >>
>
> >>
>
> >>
>
> >>
>
> >>
>
> >> --- Graham
>
> >>
>
> >>
>
> >>
>
> >> ==
>
> >>
>
> >>
>
> >>
>
> >>
>
> >>
>
> >> On Wed, Dec 7, 2016 at 10:43 AM, Bob Camp <kb...@n1k.org> wrote:
>
> >>
>
> >>
>
> >>
>
> >>> Hi
>
> >>
>
> >>>
>
> >>
>
> >>> You probably have proven one of the most basic design truths: Parts
>
> >>
>
> >>> will
>
> >>
>
> >>> *always* oscillate just
>
> >>
>
> >>> outside the bandwidth of your test gear” :). A few other possible
> issues:
>
> >>
>
> >>>
>
> >>
>
> >>> 1) Something else is oscillating and it is simply interacting with the
>
> >>
>
> >>> regulator in an odd way.
>
> >>
>
> >>> 2) The oscillation / noise is at a very low level and it’s below your
>
> >>
>
> >>> test gear’s noise floor
>
> >>
>
> >>> 3) Testing stops the oscillation
>
> >>
>
> >>>
>
> >>
&

Re: [time-nuts] Totally unrelated, but..

[Scott Stobbe]

You can buy opamps off the shelf with an input referred noise less than a
50 ohm resistor to build up a preamp (of course you can build up something
even lower with discretes, but then it's a time commitment over a basic
opamp preamp)

Even if your regulator is low noise if it's running with low phase margin
you can get noise peaking.

On Wed, Dec 7, 2016 at 3:06 PM Van Horn, David <
david.vanh...@backcountryaccess.com> wrote:

>
>
> I replaced the original caps, and I added caps, I substituted good
> Jonhansen RF caps, and Tanceram caps.
>
> No help at all.
>
>
>
> The layout and routing is as good as I could do, and the only improvement
> I could see possible would be to move one cap closer to the reg.
>
> The difference would be less than the tolerance of part placement on the
> existing pads.
>
>
>
> Fortunately this isn't a critical piece of equipment, but I want it
> working RIGHT before I put it back in service.
>
> It's a custom receiver for 457 kHz.
>
>
>
>
>
> -Original Message-
>
> From: time-nuts [mailto:time-nuts-boun...@febo.com] On Behalf Of Graham /
> KE9H
>
> Sent: Wednesday, December 7, 2016 10:41 AM
>
> To: Discussion of precise time and frequency measurement
>
> Subject: Re: [time-nuts] Totally unrelated, but..
>
>
>
> Remember that the internal Voltage reference in the original three
> terminal regulator designs is a Zener.
>
> (Zeners are also useful as RF white noise sources.) The regulator is
> generally an amplifier with DC feedback.
>
> If you look at the application notes on the early regulators, they require
> capacitors to ground on both the input and outputs.
>
> If these capacitors are missing, or too low in value, or not good
> capacitors at RF frequencies, then the Zener noise is amplified by the
> regulator amplifier and pushed out the output port.
>
> I would experiment by putting a good ceramic 0.1uF cap to ground, right at
> the regulator output port.
>
>
>
>
>
> --- Graham
>
>
>
> ==
>
>
>
>
>
> On Wed, Dec 7, 2016 at 10:43 AM, Bob Camp  wrote:
>
>
>
> > Hi
>
> >
>
> > You probably have proven one of the most basic design truths: Parts
>
> > will
>
> > *always* oscillate just
>
> > outside the bandwidth of your test gear” :). A few other possible issues:
>
> >
>
> > 1) Something else is oscillating and it is simply interacting with the
>
> > regulator in an odd way.
>
> > 2) The oscillation / noise is at a very low level and it’s below your
>
> > test gear’s noise floor
>
> > 3) Testing stops the oscillation
>
> >
>
> > Bob
>
> >
>
> > > On Dec 6, 2016, at 4:24 PM, Van Horn, David 
> > backcountryaccess.com> wrote:
>
> > >
>
> > > Lots of discussion on here about low noise regulation so someone may
>
> > know what to look for.
>
> > >
>
> > > I have a receiver which is getting a lot of interference from
> somewhere.
>
> > > Antenna disconnected, interference still high.
>
> > > After much poking around, we found that replacing a voltage
>
> > > regulator
>
> > with a slightly different part cures the problem.
>
> > > Running that section on external battery is also fine, so it appears
>
> > > the
>
> > original regulator causes some problem.
>
> > > We tried various batteries over a range of voltages within the chip
>
> > spec, and couldn't make it have a problem.
>
> > >
>
> > > I looked at the reg input and output with scope and spectrum
>
> > > analyzer,
>
> > and I don't see anything that indicates excessive noise or oscillation.
>
> > > The PCB layout is as tight as you could ask for. Fat tracks, lots of
>
> > ground, I couldn't lay it out any better.
>
> > > Replacing the input and output caps didn't change anything.
>
> > > Replacing the input and output caps with parts that should be
>
> > > "better",
>
> > like Johanson Tancerams or tantalums has no effect.
>
> > >
>
> > > Just for laughs, we tried a number of different regulator chips, all
>
> > > new
>
> > from the reel.
>
> > > The parts with the quietest and with the most noisy specs caused
>
> > problems.
>
> > > One part, with a noise spec more or less in the middle of the spread
>
> > > is
>
> > the one that works.
>
> > >
>
> > > So what is it that a monolithic regulator (linear) can do which is
>
> > > not
>
> > observable on a scope or SA, which would cause a receiver to think
>
> > it's getting a signal or significant noise in band?
>
> > > Everything else in the system is shut down, I am sure the regulator
>
> > > chip
>
> > is the culprit, but so far I don't see how it's causing the problem.
>
> > > I could just use the quiet chip and move on, but experience tells me
>
> > that I'd just have problems again down the road.  That's voodoo, not
>
> > science.
>
> > >
>
> > >
>
> > > Ideas?
>
> > >
>
> > >
>
> > >
>
> > > --
>
> > > David VanHorn
>
> > > Lead Hardware Engineer
>
> > >
>
> > > Backcountry Access, Inc.
>
> > > 2820 Wilderness Pl, Unit H
>
> > > Boulder, CO  80301 USA
>
> > > phone: 303-417-1345 x110
>
> > > email: 

Re: [time-nuts] Switching regulator replacement for 7805

[Scott Stobbe]

FWIW, you can snub the switch node to dampen parasitic ringing in exchange
of a loss in efficiency of a couple percent. I'm pretty sure I have seen
this outlined in a few app notes.

On Mon, Dec 5, 2016 at 1:22 PM, jimlux  wrote:

> On 12/5/16 9:13 AM, Dan Kemppainen wrote:
>
>>
>>
>>> Depending on the application, another possible application is to
>>> sync up the DC/DC converter to the "main" clock source. This makes
>>> the switching noise then coherent to the system, which either makes
>>> it average out completely, or possible to filter it out in the digital
>>> domain using a deep notch-filter in receiver applications.
>>>
>>>
>>>
>> You are assuming the switching noise is directly caused by the
>> fundamental switching frequency, this may not always be the case. Often
>> the fast edges can excite parasitic reactances in the board and
>> components, which in turn oscillate and radiate at their own frequency.
>> This is not necessarily related to the switching frequency.
>>
>>
>>
>
> On big science spacecraft this is pretty common (which have all sorts of
> sensitive science instruments, some of which may not have the best EMI/EMC
> characteristics from both a source and susceptability standpoint).  On the
> other hand, as switching rates have gone up, it's not clear that
> distributing a 20kHz "PWM reference clock" around is a good idea.
>
> On a radio I was recently working with, it used to radiate quite nicely at
> about 15-19 MHz, even though the switching rate was in the hundreds of kHz
> range - we figured that this is where the "antenna efficiency" (better with
> higher freq) of the traces matched with the "harmonic power" (lower with
> higher freq)
>
>
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/m
> ailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Switching regulator replacement for 7805

[Scott Stobbe]

Well a sigma-delta modulator in loose terms is an error amplifier around a
quantizer, so you get 1/loopgain rejection of quantization noise (in other
words the noise is shaped out in frequency). Resulting in a noise spectrum
that converges in 1/N versus 1/sqrt(N) for flat Gaussian noise, versus
lobes and nulls for a single sinusoid.

I totally agree that the first step is to reduce the switching residual
that is generated, even half a bond wire at say 1 nH is 13 mOhms at 2 MHz,
combine that with a power converter running 30% ripple current of a 1A
output is 300mA ripple current resulting in 4 mV just on half a bondwire.
Equivalently a 10 uF MLCC should be able to hit 10 mOhms at 2 MHz before
hitting its SRF.

On Sun, Dec 4, 2016 at 5:58 PM, Attila Kinali <att...@kinali.ch> wrote:

> On Sun, 4 Dec 2016 16:22:02 -0500
> Scott Stobbe <scott.j.sto...@gmail.com> wrote:
>
> > If you wanted to be nutty you wouldn't go PWM at all, just like
> > fractional-N sythns don't just mash 2 divider values. You would
> sigma-delta
> > modulate your power stage. I don't know if you can buy one COTS, but
> there
> > are plenty of papers on rolling your own.
>
> I guess you are refering to spread-spectrum techniques.
> Such DC/DC converters exist, but are usually those with high power
> ratings. IMHO it is also not worth the effort, as its main use is
> to meet EMI emission requirements. The only application that comes
> to my mind where spread-spectrum actually helps are high sensitiv
> radio receivers where every spur is a nuisance. For most other
> use, and time-nuts use in particular, it is much less useful.
> The noise energy is not gone. It is still there, just spread over
> a large bandwidth. In time measurement applications, noise is
> integrated over time _and_ frequency. Thus even if the noise is
> spread over a large bandwidth, the energy will still contribute
> to the uncertainty and degrade the ADEV. It will be just harder
> to identify as the peak is now much smaller and wanders in frequency.
>
> It is much better to the design such, that as little as possible
> of the switching energy leaks out of the DC/DC converter and filter
> out the rest.
>
> Depending on the application, another possible application is to
> sync up the DC/DC converter to the "main" clock source. This makes
> the switching noise then coherent to the system, which either makes
> it average out completely, or possible to filter it out in the digital
> domain using a deep notch-filter in receiver applications.
>
>
> Attila Kinali
> --
> Malek's Law:
> Any simple idea will be worded in the most complicated way.
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Switching regulator replacement for 7805

[Scott Stobbe]

If you wanted to be nutty you wouldn't go PWM at all, just like
fractional-N sythns don't just mash 2 divider values. You would sigma-delta
modulate your power stage. I don't know if you can buy one COTS, but there
are plenty of papers on rolling your own.

On Sun, Dec 4, 2016 at 4:09 PM, Attila Kinali  wrote:

> On Sun, 4 Dec 2016 15:45:55 -0500
> Bob Camp  wrote:
>
> > Given that the parts to build one are fairly easy to get and that we
> likely have “nutty”
> > EMI requirements. Maybe a small board that drops into a 78xx footprint
> is the
> > better solution.
>
> I don't fully agree. To go full "nutty" you would want to use the chips
> in the QFN packages, as those have shorter leads and a large ground pad
> to keep EMI down. But soldering an 0.5mm QFN by hand is impossible
> (believe me, I tried) and not everyone has an oven. Also a lot of the
> EMI performance depend on what output current you expect. Using a 1A
> design for 10mA is a good way to get poor EMI performance as the
> chip will go from PWM (aka fixed frequency) to PFM (aka fixed on time)
> mode under low load to increase efficiency. The same goes for input
> voltage. So we would need to do multiple designs for different
> output current and input voltage ranges... which kind of defeats
> the purpose of "a small board that drops into".
>
> What could be done, though, is a small board with all components,
> but the inductor on it. So then one could select the right inductor
> for the application and solder it on. Using an inductor that is easy
> to solder (like e.g. WE-PD3) with a wide selection of values should
> do the trick.
>
> Attila Kinali
>
> --
> It is upon moral qualities that a society is ultimately founded. All
> the prosperity and technological sophistication in the world is of no
> use without that foundation.
>  -- Miss Matheson, The Diamond Age, Neil Stephenson
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP53131A RS232 question

[Scott Stobbe]

The 53131A is approximately 0.9 Sa/S with a 1 Sec Gate time. With 0.1 s
gate time things get really ugly... His issue is not gate time his setup
works with print turned off.

On Sat, Dec 3, 2016 at 2:42 PM, Bob Camp <kb...@n1k.org> wrote:

> Hi
>
> Ok, it’s 0.9992 samples per second … I guess that is so close to
> 5.0 seconds and so far from 1. seconds that it
> explains the issue
>
> Bob
>
> > On Dec 3, 2016, at 2:01 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
> >
> > Well, that's only true if your counter has zero dead-time... I know you
> > know this, 1 second gate time will give slightly less than 1 Sa/S.
> >
> > On Fri, Dec 2, 2016 at 1:15 PM, Bob Camp <kb...@n1k.org> wrote:
> >
> >> Hi
> >>
> >> With a 1 second gate time, you will get one data output per second with
> the
> >> settings you have described. I suspect there is something else wrong
> with
> >> your
> >> setup. The same settings will work for 1 pps and give you one reading
> >> output
> >> per second.  The input level should have no impact as long as it will
> >> properly
> >> trigger the counter.
> >>
> >> Bob
> >>
> >>> On Dec 2, 2016, at 12:54 PM, <cdel...@juno.com> <cdel...@juno.com>
> >> wrote:
> >>>
> >>> Hi,
> >>>
> >>> I have not played with my 53131A RS232 for a couple years and have a
> >>> question.
> >>> I want to configure it to "talk" to Hyperterminal so I can log some
> >>> readings.
> >>> I did the 5 items to configure it:
> >>> HW pacing (disabled)
> >>> print on
> >>> baud rate (19200)
> >>> parity
> >>> SW pacing (none)
> >>>
> >>> with a 5Mhz input I only get data sent about once every 3 seconds, I
> >>> thought with  a 1 second gate I'd get an  output measurement every
> >>> second?
> >>>
> >>> I would think there is plenty of time to output the first measurement
> >>> before the next one is read!
> >>>
> >>> Actual setup I'm wanting to do is time interval (at a 1PPS rate) and
> log
> >>> the results.
> >>>
> >>> I remember I had it working once and it would be OK as long as I did
> not
> >>> exceed around 80% of full scale.
> >>> If that happened I would only get an output logged every other second.
> >>>
> >>> I can live with that part but need to get back to it working below
> around
> >>> 80%.
> >>>
> >>> Any advice is most welcome!
> >>>
> >>> I'm going to configure for the TI and <80% and see what happens.
> >>>
> >>> Cheers,
> >>>
> >>> Corby
> >>>
> >>> ___
> >>> time-nuts mailing list -- time-nuts@febo.com
> >>> To unsubscribe, go to https://www.febo.com/cgi-bin/
> >> mailman/listinfo/time-nuts
> >>> and follow the instructions there.
> >>
> >> ___
> >> time-nuts mailing list -- time-nuts@febo.com
> >> To unsubscribe, go to https://www.febo.com/cgi-bin/
> >> mailman/listinfo/time-nuts
> >> and follow the instructions there.
> >>
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Best Chance GPS module

[Scott Stobbe]

Hi Lewis,

Here is a sample data-point related to processor load, on the RPI 2.
Stepping from Idle to full load on 4 cores resulted in a temp rise near the
XO of approximately 14 degC, and correspondingly the XO shifted 3.6 PPM.

On Sat, Dec 3, 2016 at 10:29 AM, MLewis  wrote:

> So much to absorb and learn from what people have responded with.
>
> Thanks all!
>
> On 01/12/2016 12:01 PM, Chris Albertson wrote:
>
>> OK, now I know what you need.   Millisecond level time on the data
>> processing machine. ... Let's assume you were able to set up a local NTP
>> server that runs off it's
>> own GPS reference clock. ... about 100x better then you
>> need. ...  Ethernet is not perfect but good enough for what you want.
>>
> That's the take I have on it.
>
> I really doubt varying processing load is an issue with NTP. ... What
>> happens is
>> the PPS causes an interrupt and inside the handler the nanosecond clock is
>> sampled and copied to memory.   The handler has something like 8 lines of
>> code and runs very fast.
>>
> That's good to know about PPS, 'cause the computer's load while polling
> hosts over the internet is getting me horrible variance in offsets.
>
> The other thing you might look at is NOT using NTP but using PTP.
>>
> Looks like fun, but way better than I need. More than I can afford too.
>
> I don't think your measurements are measuring correctly. ... Any offset is
>> perfectly fine, that is simple the communications delay and is accounted
>> for by NTP. ... If you were looking at
>> offset, just don't do that.
>>
> O.K.. I'm pretty sure I don't understand that, but the issue I found was
> not that different hosts offsets varied from one another, but that the
> offset reported by "a" host would jump around. And when the load on the
> computer was changing, low-to-high or high-to-low, several hosts',
> sometimes all hosts, offsets jump around, with that multi-host variance
> continuing for some minutes after the computer was running at the new load.
> This settled down a little when I turned core parking off, but only a
> little. I've attached a sample of the offsets I'm getting to show this
> variance. Oddly, a sustained higher load would often settle the variance
> and give the most consistent results: one such period is between poll 6 &
> 10 on the "test run N24" attachment, and the graph shows the offset slope
> and hosts' offset variances as the load moves from heavy to medium and then
> light.
>
> After giving up on SMAs to tame individual host's offsets, to get a usable
> offset from the reported offsets I implemented a cascade of filters:
> applying factors on standard deviation to implement truncation means to
> remove outliers, then winsorizing means, with independent bias factors
> applied to selection and winsorization. This all worked rather well once I
> tuned the filters' parameters. Then the variance got worse, so I added some
> increasing attenuation on increasing corrected offset values and that made
> my corrected offset usable within the tolerance I needed, until from a
> certain date the reported offsets went all over the map.
>
> But we shouldn't go down this road, other than curiosity (and I wish I had
> the time to explore the why), as going to a separate machine as an NTP host
> removes all of those types of issues. And I don't have to grow my own
> code...
>
> I think your only problem is finding a GPS with PPS output that works at
>> your location.   Don't worry much more.  If it works and has PPS it is
>> good
>> enough
>>
> Exactly.
> Any module that can get a usable GPS signal can discipline time and be
> delivered over my local Ethernet to better than I need.
>
> You might have a "Plan B", ...
>>
> Thanks for those. Good to know.
>
>
>
> I believe the location issues narrows it down to the MAX-M8Q or the
> NEO-M8T.
>
> Both have great sensitivity, but their firmware varies to address intent.
> The M8Q can be explicitly set to Dynamic-Mode-Stationary (as it should go
> to automatically with an unmoving antenna) and the M8T will set there as it
> moves to focusing on a better time solution after establishing a location
> fix. In comparing their product descriptions, the M8T seems the better
> choice while sitting still for obtaining usable results in questionable
> locations, but - speculating - that wording may be marketing wording in
> response to prior issues with earlier T series modules. And so far, I've
> not found any accounts of first hand experience with a M8T.
>
> The other issue is what breakout board the modules are available on.
> - With the M8Q, there's hats or boards that can connect direct to a Pi or
> such, but lack protection with supply voltage or outputs if I want to feed
> them to another computer.
> - And the M8T is available on a board that provides power regulation and
> some protection, so that should be able to feed NEMA & PPS to any suitable
> computer without risk.
> - And I found a board that accepts GPS 

Re: [time-nuts] HP53131A RS232 question

[Scott Stobbe]

Well, that's only true if your counter has zero dead-time... I know you
know this, 1 second gate time will give slightly less than 1 Sa/S.

On Fri, Dec 2, 2016 at 1:15 PM, Bob Camp  wrote:

> Hi
>
> With a 1 second gate time, you will get one data output per second with the
> settings you have described. I suspect there is something else wrong with
> your
> setup. The same settings will work for 1 pps and give you one reading
> output
> per second.  The input level should have no impact as long as it will
> properly
> trigger the counter.
>
> Bob
>
> > On Dec 2, 2016, at 12:54 PM,  
> wrote:
> >
> > Hi,
> >
> > I have not played with my 53131A RS232 for a couple years and have a
> > question.
> > I want to configure it to "talk" to Hyperterminal so I can log some
> > readings.
> > I did the 5 items to configure it:
> > HW pacing (disabled)
> > print on
> > baud rate (19200)
> > parity
> > SW pacing (none)
> >
> > with a 5Mhz input I only get data sent about once every 3 seconds, I
> > thought with  a 1 second gate I'd get an  output measurement every
> > second?
> >
> > I would think there is plenty of time to output the first measurement
> > before the next one is read!
> >
> > Actual setup I'm wanting to do is time interval (at a 1PPS rate) and log
> > the results.
> >
> > I remember I had it working once and it would be OK as long as I did not
> > exceed around 80% of full scale.
> > If that happened I would only get an output logged every other second.
> >
> > I can live with that part but need to get back to it working below around
> > 80%.
> >
> > Any advice is most welcome!
> >
> > I'm going to configure for the TI and <80% and see what happens.
> >
> > Cheers,
> >
> > Corby
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP53131A RS232 question

[Scott Stobbe]

I have a HP53131A and don't have this issue, the only thing I can think of
is that you have the cables swapped i.e. the start edge on ch2 and stop
edge on ch1.

I have firmware REV:3703, If there are any settings you want me to try I am
happy to help.

On Sat, Dec 3, 2016 at 12:47 PM,  wrote:

> I've tried all the parameters and the update rate still drops to about
> once every 3 seconds when I turn print on!
> You still see the 1pps strobing the input channels every transition even
> with print on.
> I was using just pins 2, 3 and 5 on the rs232 and switched to a "full"
> cable with no improvement.
> Since the problem  occurs even without the laptop connected I don't think
> it's a cable issue.
> Do I need to have the laptop issue a GET command or something after each
> transmission?
> Heeelp!
>
> Cheers,
>
> Corby
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] RPi/ beagle bone-like computer without video

[Scott Stobbe]

As far as I know Matlab is x86 only, of course you can generate c code from
Matlab but that has its own challenges.

On Fri, Dec 2, 2016 at 2:14 PM jimlux  wrote:

> On 12/2/16 10:08 AM, David J Taylor wrote:
> > From: jimlux
> >
> > We're processing several thousand samples, received over a serial port
> > or USB in a few seconds.  The algorithm (in Matlab, hence the need for
> > Linux) grinds for around 30 seconds to produce the output.
> > ===
> >
> >
> > The Raspberry Pi OS includes a free MATLAB licence, as I expect you
> > already know.
> >
>
>
>
> yes, but
>
> "Note that you cannot install MATLAB as a standalone application or
> execute MATLAB code locally on a Raspberry Pi. To develop applications
> for the Raspberry Pi to be used in standalone operation, please see the
> Raspberry Pi Support from Simulink®."
>
>
> and of course, we're not using Simulink
>
> > Cheers,
> > David
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] RPi/ beagle bone-like computer without video

[Scott Stobbe]

As much as it pains me to recommend them some of the iot modules sound like
a good fit, like the Intel atom one.

On Fri, Dec 2, 2016 at 12:13 PM jimlux  wrote:

> On 12/2/16 8:51 AM, Attila Kinali wrote:
> > On Fri, 2 Dec 2016 08:05:17 -0800
> > jimlux  wrote:
> >
> >>> I'm measuring 0.350A with max cpu usage on all cores and the following
> >>> settings:
> >>>
> >> @ 5V, right, so 1.75W going full out.  That's a useful number to know.
> >
> > I once did an board with an i.m233 (arm9 at 400MHz) and it did less
> > than 0.5W IIRC. But I think this is about the lower you can do without
> > tweaking the OS.
> >
> >>> Apart from that I read somewhere that the beaglebones have terrible EM
> >>> characteristics. But I can't find my source so take it with a grain of
> >>> salt :-)
> >
> > Yes. The BBB is badly designed in this regard... The RPI is better there.
> >
> >> I must say, there are more of these things available in almost too wide
> >> a variety.. You could spend days going through all the datasheets and
> >> websites - I suspect that they ALL have about the same power consumption
> >> for a given amount of processing horsepower - same feature size on the
> >> die, after all - so it's more about peripherals and ease of use
> >
> > There are way too many, IMHO. And a lot of them are not usefull for
> > a lot of stuff or have very bad support. The Odroid are a prime example
> > of this. There is a "community" around them, yet getting them to do
> > anything usefull is a major pain. There is a handfull of companies
> > I know of, who do provide good support and those are the ones I am
> > usually sticking to (unless I have special needs).
> >
> >> And, it's more likely that idiosyncracies in the distros have been
> >> identified and it's more likely that the software will run on them after
> >> its built.
> >
> > It's actually better to go with a company who is invested in giving
> > you a working board than using something popular. Especially one that
> > cares to push all its patches upstream.
> >
> > Beside the mentioned Toradex, and Aries Embedded, there is also Olimex
> > which is known for it's wide variety of boards with good support.
> > Depending on your exact requirements, I would probably go for one of
> > the i.mx233 boards (the imx233 nano is quite neat) or A10 or A33 board.
> > Especially the i.mx233 is nice as it has an on-chip Li-poly
> charger/controller.
> > All you need to do is to supply it with 5V and it does the rest.
> >
> > A note of warning: a lot of the boards from Olimex have not enough ground
> > pins for the high speed singals they provide. If you are transfering data
> > with high-speed (several 10MHz) over the headerpin connectors, you will
> need
> > to add some additional ground connections.
> >
> > What are the exact requirements you have? How much computational power
> > do you need? How do you interface the sensors? How many boards will
> > you need? Is it out of question to build your own processor board using
> > one of the ARM9's in QFP? What is your budget?
>
> We're processing several thousand samples, received over a serial port
> or USB in a few seconds.  The algorithm (in Matlab, hence the need for
> Linux) grinds for around 30 seconds to produce the output.
>
> we're not sensitive on the "board cost" - labor to design a board is
> expensive, so a board that has low power, and the right connectors, so
> it's <1 day to make cables, etc. is a better deal than several weeks to
> design a board and spin it, etc.
>
>
>
> >
> > The reason why I'm asking the last two questions is, it is often more
> efficient
> > to do your own CPU board if you have to design a PCB anyways for the
> sensors,
> > need more than 10-20 boards and you can live with one of the "small"
> ARM9's
> > that come in QFP packages (like the i.mx233 or AM1705).
>
>
>
> That would come later, and be "someone else's problem" - We do the proof
> of concept, "demonstrate that it works in a relevant environment", and
> then it goes from there.
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] RPi/ beagle bone-like computer without video

[Scott Stobbe]

Having a full blown os is nice when all the processor is responsible for is
house keeping and storage. You also get plenty of RAM for buffering prior
to writing to persistent storage, like an SD card. That said I have found
SD cards to be fusy, at least the microchip FAT libraries. And write cycle
times are random sometimes being 0.5 seconds, which means you need to be
able to allocate memory the equivalent of a few seconds worth of data for
buffering (not an issue on linux systems).

On Wed, Nov 30, 2016 at 10:05 PM, Graham / KE9H 
wrote:

> If you are going to go battery powered, I would also recommend staying away
> from Linux, go with something like a 32 bit PIC32MX or PIC32MZ.  Full
> Ethernet stack, RTOS if you need it, can do deep sleep down into the
> microamp range when not active.
>
> --- Graham.
>
> On Wed, Nov 30, 2016 at 8:35 PM, Gary Chatters <
> gcarlis...@garychatters.com>
> wrote:
>
> > There do seem to be a lot of small SBCs out there.  I have used boards
> from
> > Technologic Systems, http://www.embeddedarm.com
> > and EMAC, Inc, http://www.emacinc.com
> >
> > They have numerous models for you to look though.  Some various features:
> > - SBC or SoM/CoM on baseboard
> > - Various form factors including PC/104
> > - Usually with RS-232, USB, GPIO, Ethernet.
> > - Many without video.
> > - ARM processor
> > - Many with industrial temperature range
> > - Linux and development environment provided (may not be latest)
> > - Lower power then BBB.  One model runs at 0.5 watts.  Many around 1.
> >
> > Prices generally 2 or 3 x BBB prices.
> >
> > Gary
> >
> > On 11/30/2016 03:42 PM, jimlux wrote:
> >
> >> I'm looking for a small linux single board - similar to RPi or
> >> Beaglebone Black, but don't need the HDMI, or video stuff.
> >> Preferably without weird connectors, and available for wide temperature
> >> ranges (it's for a data logger/collector in the field)
> >>
> >> What's out there?
> >>
> >> There's BBB in industrial flavor (-40 to +85C ) for $60-70
> >>
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/m
> > ailman/listinfo/time-nuts
> > and follow the instructions there.
> >
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thunderbolt Schematic

[Scott Stobbe]

If the unit is low on amplitude, there isn't likely much more than a buffer
amp between the ocxo and 10 MHz output.

On Tue, Nov 29, 2016 at 7:54 PM Bob Camp  wrote:

> Hi
>
> There is no schematic for the TBolt available. Based on Trimble’s approach
> to keeping
> all their IP private (look at the survey side …), there will not be a
> schematic coming out
> from them. Since a good chunk of the functionality is in the FPGA’s and
> CPU’s, a schematic
> may be a tough thing to come up with by tracing this and that on the board.
>
> Bob
>
> > On Nov 29, 2016, at 6:38 PM, Richard W. Solomon 
> wrote:
> >
> > From the last e-mail I saw in 2014, there was no schematic available.
> >
> >
> >
> > Has anything changed lately, is one available ?
> >
> >
> >
> > I have one with low output that I would like to take a crack at fixing.
> >
> >
> >
> > Thanks, Dick, W1KSZ
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Excel logarithmic function (was Thermal impact on OCXO)

[Scott Stobbe]

Hi Lars,

There are a few other pieces I have yet to fully appreciate. One of which
is that Aln(Bt+1) isn't a time-invariant model. In the most common case
(for the mfg) the time scale aligns with infancy of the OCXO, when it's hot
off the line. However after pre-aging, perhaps some service life, what time
reference is best? Sometime I will try adding an additional parameter for
infancy time and see how that goes.

A fit of the full ten year data-set, attached in the two plots
"Lars_10Year.png", "Lars_10Year_45Day.png".

I would agree to your description of 1/sqrt(t) aging for the first 1000
days, but sometime after, it follows 1/t. Attached is plot of age rate
"Lars_AgeRate.png". You can see during the first 1000 days the age rate
declines at 1 decade for 2 decades time indicating t^(-1/2), but eventually
it follows 1/t.

On Wed, Nov 23, 2016 at 3:57 PM, Lars Walenius <lars.walen...@hotmail.com>
wrote:

> Hi Scott.
>
>
>
> Here is a textfile with data for the 10 years (As in the graph 2001-2011).
>
>
>
> Also the ln(bt+1) fit, as Magnus said, has the derivate b/(b*t+1) that
> with b*t >>1 is 1/t. But my data has the aging between 1 and 10 years more
> like 1/sqrt(t) If I just have a brief look on the aging graph.
>
>
>
> Lars
>
>
>
> *Från: *Scott Stobbe <scott.j.sto...@gmail.com>
> *Skickat: *den 19 november 2016 04:11
>
> Hi Lars,
>
>
>
> I agree with you, that if there is data out there, it isn't easy to find,
>
> many thanks for sharing!
>
>
>
> Fitting to the full model had limited improvements, the b coefficient was
>
> quite large making it essentially equal to the ln(x) function you fitted in
>
> excel. It is attached as "Lars_FitToMil55310.png".
>
>
>
> So on further thought, the B term can't model a device aging even faster
>
> than it should shortly after infancy. In the two extreme cases either B is
>
> large and (Bt)>>1 so the be B term ends up just being an additive bias, or
>
> B is small, and ln(x) is linearized (or slowed down) during the first bit
>
> of time.
>
>
>
> You can approximated the MIL 55310 between two points in time as
>
>
>
> f(t2) - f(t1) = Aln(t2/t1)
>
>
>
> A = ( f(t2) - f(t1) )/ln(t2/t1)
>
>
>
> Looking at some of your plots it looks like between the end of year 1 and
>
> year 10 you age from 20 ppb to 65 ppb,
>
>
>
> A ~ 20
>
>
>
> The next plot "Lars_ForceAcoef", is a fit with the A coefficient forced to
>
> be 2 and 20. The 20 doesn't end-up fitting well on this time scale.
>
>
>
> Looking at the data a little more, I wondered if the first 10 day are going
>
> through some behavior that isn't representative of long-term aging, like
>
> warm-up, retrace (I'm sure bob could name half a dozen more examples). So
>
> the next two plots are fits of the 4 data points after day10, and seem to
>
> fit well, "Lars_FitAfterDay10.png", "Lars_1Year.png".
>
>
>
> If you are willing to share the next month, we can add that to the fit.
>
>
>
> Cheers,
>
>
>
> On Fri, Nov 18, 2016 at 1:26 PM, Lars Walenius <lars.walen...@hotmail.com>
>
> wrote:
>
> >
>
> > Hopefully someone can find the correct a and b for a*ln(bt+1) with
>
> stable32 or matlab for this data set:
>
> > Days ppb
>
> > 2   2
>
> > 4   3.5
>
> > 7   4.65
>
> > 8   5.05
>
> > 9   5.22
>
> > 12 6.11
>
> > 13 6.19
>
> > 25 7.26
>
> > 32 7.92
>
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Do reflections up/down the antenna cable cause a problem with GPS?

[Scott Stobbe]

Ouch! Relay stepped attenuator? Or solid state components?

On Mon, Nov 21, 2016 at 11:00 PM Bob Camp <kb...@n1k.org> wrote:

> Hi
>
> There are packages you can put on a fairly standard HP signal generator
> that will
> do the mismatch stuff without spending all the money Sperient wants for
> one of
> their machines. One suggestion: If you *do* go with the HP solution,
> running the
> attenuator up and down to simulate fast fades (think urban canyon) will
> fry the generator
> in about 30 days … I have empirical data ….
>
> Bob
>
> > On Nov 21, 2016, at 8:53 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
> >
> > I haven't used one personally but a spirent gps simulator would let you
> do
> > a try it and see.
> >
> > It will be interesting to see if out of the growing sdr community an open
> > source gps simulator emerges.
> >
> > On Mon, Nov 21, 2016 at 1:01 PM Dr. David Kirkby (Kirkby Microwave Ltd) <
> > drkir...@kirkbymicrowave.co.uk> wrote:
> >
> >> People state it is desirable to have a GPS antenna well clear of
> >> obstructions, which I believe is to stop reflections. But there is
> another
> >> source of reflections which I suspect could be just as problematic.
> >>
> >> Whilst the input impedance of the antenna input terminal on a GPS
> receiver
> >> is probably marked 50 Ohms, I'd be somewhat surprised if it was very
> close
> >> to 50 Ohms. Antenna cables have an impedance, which is typically 50 +/-
> 2
> >> Ohms, but this varies, not only between different makes/models of
> cables,
> >> but even on the same real of cable.The output of the pre-amp is most
> >> unlikely to have a 50 Ohm source impedance. In fact, the output
> impedance
> >> might be close to 0 Ohms, as it may be driven by a voltage source,
> without
> >> any 50 Ohm resistor.
> >>
> >> Anything not immediately absorbed by the GPS receiver is going to be
> >> reflected back up the coax, and could be reflected multiple times.
> >>
> >> I just looked on my HP 8720D VNA, and see I can reduce the output power
> to
> >> -70 dBm, which would should not do any damage. It will be interesting to
> >> see just what the input impedance of the GPS receiver is. I'm tied up
> with
> >> doing my accounts over the next few days, but later I will look.
> >>
> >> If reflections on the antenna/cable/receiver are a problem, then
> >> attenuators can improve the match, but of course they reduce the signal
> >> level too. A more intelligent, but more difficult solution, is to build
> a
> >> matching network. For that one would need a VNA to measure the
> impedance in
> >> the first place.
> >>
> >> Dr. David Kirkby Ph.D CEng MIET
> >> Kirkby Microwave Ltd
> >> Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, Essex, CM3
> 6DT,
> >> UK.
> >> Registered in England and Wales, company number 08914892.
> >> http://www.kirkbymicrowave.co.uk/
> >> Tel: 07910 441670 / +44 7910 441670 (0900 to 2100 GMT only please)
> >> ___
> >> time-nuts mailing list -- time-nuts@febo.com
> >> To unsubscribe, go to
> >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> >> and follow the instructions there.
> >>
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Do reflections up/down the antenna cable cause a problem with GPS?

[Scott Stobbe]

I haven't used one personally but a spirent gps simulator would let you do
a try it and see.

It will be interesting to see if out of the growing sdr community an open
source gps simulator emerges.

On Mon, Nov 21, 2016 at 1:01 PM Dr. David Kirkby (Kirkby Microwave Ltd) <
drkir...@kirkbymicrowave.co.uk> wrote:

> People state it is desirable to have a GPS antenna well clear of
> obstructions, which I believe is to stop reflections. But there is another
> source of reflections which I suspect could be just as problematic.
>
> Whilst the input impedance of the antenna input terminal on a GPS receiver
> is probably marked 50 Ohms, I'd be somewhat surprised if it was very close
> to 50 Ohms. Antenna cables have an impedance, which is typically 50 +/- 2
> Ohms, but this varies, not only between different makes/models of cables,
> but even on the same real of cable.The output of the pre-amp is most
> unlikely to have a 50 Ohm source impedance. In fact, the output impedance
> might be close to 0 Ohms, as it may be driven by a voltage source, without
> any 50 Ohm resistor.
>
> Anything not immediately absorbed by the GPS receiver is going to be
> reflected back up the coax, and could be reflected multiple times.
>
> I just looked on my HP 8720D VNA, and see I can reduce the output power to
> -70 dBm, which would should not do any damage. It will be interesting to
> see just what the input impedance of the GPS receiver is. I'm tied up with
> doing my accounts over the next few days, but later I will look.
>
> If reflections on the antenna/cable/receiver are a problem, then
> attenuators can improve the match, but of course they reduce the signal
> level too. A more intelligent, but more difficult solution, is to build a
> matching network. For that one would need a VNA to measure the impedance in
> the first place.
>
> Dr. David Kirkby Ph.D CEng MIET
> Kirkby Microwave Ltd
> Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, Essex, CM3 6DT,
> UK.
> Registered in England and Wales, company number 08914892.
> http://www.kirkbymicrowave.co.uk/
> Tel: 07910 441670 / +44 7910 441670 (0900 to 2100 GMT only please)
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] How phase stable is rg59 or alternate coax

[Scott Stobbe]

When I first took a look at some of the coax datasheets I couldn't find
anything. I was able to find the following paper "phase stability of
typical navy radio frequency coaxial cables"
http://www.dtic.mil/dtic/tr/fulltext/u2/628682.pdf I attached the table
from the last page. They estimate RG59 to have a tempCo of -330 PPM/degC
for electrical length. They also estimated RG-58 at -480 PPM/degC.

On Mon, Nov 21, 2016 at 2:44 PM, John Ackermann N8UR  wrote:
>
> I can't find the data right now, but will keep digging.  There's also a
> short paper from the early 2000s from Haystack on their measurement of
> LMR400 in an environmental chamber.  They came to the same conclusion, but
> I can't find that paper either. :-


John, many thanks for the Haystack tip! That is a wonderful paper, I
believe the one you are quoting is "Dispersion and temperature effects in
coax cables" http://www.haystack.mit.edu/tech/vlbi/mark5/mark5_memos/067.pdf
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

[time-nuts] How phase stable is rg59 or alternate coax

[Scott Stobbe]

If you had 30 ft of rg59 outdoors seeing maybe 10 degC swings everyday,
would the propagation time be stable to ps? ns?

On Sun, Nov 20, 2016 at 7:04 PM Hal Murray  wrote:

>
> Is that even a sensible question?  Is there a better way to phrase it?
>
>
> The problem I'm trying to avoid is that the weather and the satellite
> geometry change over time so I can't just collect data for X hours, switch
> to
> the other antenna or move the antenna to another location, collect more
> data,
> then compare the two chunks of data.
>
> The best I can think of would be to setup a reference system so I can
> collect
> data from  2 antennas and 2 receivers at the same time.  It would probably
> require some preliminary work to calibrate the receivers.  I think I can do
> that by swapping the antenna cables.
>
>
> If I gave you a pile of data, how would you compute a quality number?  Can
> I
> just sum up the S/N slots for each visible/working satellite?
>
>
> --
> These are my opinions.  I hate spam.
>
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Excel logarithmic function (was Thermal impact on OCXO)

[Scott Stobbe]

Hi Lars,

I agree with you, that if there is data out there, it isn't easy to find,
many thanks for sharing!

Fitting to the full model had limited improvements, the b coefficient was
quite large making it essentially equal to the ln(x) function you fitted in
excel. It is attached as "Lars_FitToMil55310.png".

So on further thought, the B term can't model a device aging even faster
than it should shortly after infancy. In the two extreme cases either B is
large and (Bt)>>1 so the be B term ends up just being an additive bias, or
B is small, and ln(x) is linearized (or slowed down) during the first bit
of time.

You can approximated the MIL 55310 between two points in time as

f(t2) - f(t1) = Aln(t2/t1)

A = ( f(t2) - f(t1) )/ln(t2/t1)

Looking at some of your plots it looks like between the end of year 1 and
year 10 you age from 20 ppb to 65 ppb,

A ~ 20

The next plot "Lars_ForceAcoef", is a fit with the A coefficient forced to
be 2 and 20. The 20 doesn't end-up fitting well on this time scale.

Looking at the data a little more, I wondered if the first 10 day are going
through some behavior that isn't representative of long-term aging, like
warm-up, retrace (I'm sure bob could name half a dozen more examples). So
the next two plots are fits of the 4 data points after day10, and seem to
fit well, "Lars_FitAfterDay10.png", "Lars_1Year.png".

If you are willing to share the next month, we can add that to the fit.

Cheers,

On Fri, Nov 18, 2016 at 1:26 PM, Lars Walenius 
wrote:
>
> Hopefully someone can find the correct a and b for a*ln(bt+1) with
stable32 or matlab for this data set:
> Days ppb
> 2   2
> 4   3.5
> 7   4.65
> 8   5.05
> 9   5.22
> 12 6.11
> 13 6.19
> 25 7.26
> 32 7.92
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

I couldn't agree more, that, once you add a correlated disturbance or
1/f^a power law noise, things get even messier. Gaussian is just the
easiest to toss in.

I once herd a story from once upon a time that, if you bought a 10%
resistor, what you ended up with is something like this in the figure
attached.

Of course 1% percent resistors (EIA96) are manufactured in high yield
today, but I would guess some of this still applies to OCXOs, you
aren't likely to find a gem in the D grade parts. After pre-aging for
a couple of weeks they are either binned, labeled D, or the ones that
show promise are left to age some more before being tested to C grade,
etc, etc.

On Wed, Nov 16, 2016 at 8:06 PM, Bob Camp <kb...@n1k.org> wrote:
> Hi
>
> The issue in fitting over short time periods is that the noise is very much
> *not* gaussian. You have effects from things like temperature and warmup
> that *do* have trends to them. They will lead you off into all sorts of dark
> holes fit wise.
>
> Bob
>
>> On Nov 16, 2016, at 6:48 PM, Scott Stobbe <scott.j.sto...@gmail.com> wrote:
>>
>> A few different plots. I didn't have an intuitive feel for what the B
>> coefficient in log term looks like on a plot, so that is the first
>> plot. The same aging curve is plotted three times, with the exception
>> of the B coefficient being scaled by 1/10, 1, 10 respectively. In hand
>> waving terms, it does have an enormous impact during the first 30 days
>> (or until Bt >>1), but from then on, it is just an additive offset.
>>
>> The next 4 plots are just sample fits with noise added.
>>
>> Finally the 6th plot is of just the first 30 days, the data would seem
>> to be cleaner than what was shown as a sample in the paper, but the
>> stability of the B coefficient in 10 monte-carlo runs is not great.
>> But when plotted over a year the results are minimal.
>>
>>  A1  A2A3
>> 0.022914   6.8459   0.00016743
>> 0.022932   6.6702   0.00058768
>> 0.023206   5.79690.0026103
>> 0.023219   4.31270.0093793
>>  0.02374   2.8309 0.016838
>> 0.023119   5.02140.0061557
>> 0.023054   5.83990.0031886
>> 0.022782   9.8582   -0.0074089
>> 0.023279   3.7392 0.012161
>>  0.02345   4.10620.0095448
>>
>> The only other thing to point out from this, is that the A2 and A3
>> coefficients are highly non-orthogonal, as A2 increases, A3 drops to
>> make up the difference.
>>
>> On Wed, Nov 16, 2016 at 7:38 AM, Bob Camp <kb...@n1k.org> wrote:
>>> Hi
>>>
>>> The original introduction of 55310 written by a couple of *very* good guys:
>>>
>>> http://tycho.usno.navy.mil/ptti/1987papers/Vol%2019_16.pdf
>>>
>>> A fairly current copy of 55310:
>>>
>>> https://nepp.nasa.gov/DocUploads/1F3275A6-9140-4C0C-864542DBF16EB1CC/MIL-PRF-55310.pdf
>>>
>>> The “right” equation is on page 47. It’s the “Bt+1” in the log that messes 
>>> up the fit. If you fit it without
>>> the +1, the fit is *much* easier to do. The result isn’t quite right.
>>>
>>> Bob
>>>
>>>
>>>> On Nov 15, 2016, at 11:58 PM, Scott Stobbe <scott.j.sto...@gmail.com> 
>>>> wrote:
>>>>
>>>> Hi Bob,
>>>>
>>>> Do you recall if you fitted with true ordinary least squares, or fit with a
>>>> recursive/iterative approach in a least squares sense. If the aging curve
>>>> is linearizable, it isn't jumping out at me.
>>>>
>>>> If the model was hypothetically:
>>>>F = A ln( B*t )
>>>>
>>>>F = A ln(t) + Aln(B)
>>>>
>>>> which could easily be fit as
>>>>F  = A' X + B', where X = ln(t)
>>>>
>>>> It would appear stable32 uses an iterative approach for the non-linear
>>>> problem
>>>>
>>>> "y(t) = a·ln(bt+1), where slope = y'(t) = ab/(bt+1) Determining the
>>>> nonlinear log fit coefficients requires an iterative procedure. This
>>>> involves setting b to an in initial value, linearizing the equation,
>>>> solving for the other coefficients and the sum of the squared error,
>>>> comparing that with an error criterion, and iterating until a satisfactory
>>>> result is found. The key aspects to this numerical analysis process are
>>>> establishing a satisfactory iteration factor and error criterion to assure
>>>> both convergence and small res

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

It sounds like you knew what I meant by linearization, but I really
should have wrote linearize in parameters. Of course functions like

F = Ax^2 + Bx + C
F = Asin(omega t) + Bcos(omega t)

fit extremely well with ordinary least squares.

Well there is no free lunch, nlsq has its own challenges, the results
you get depends on the initial conditions you provide, and can also
completely diverge, or converge on a local minima which isn't the best
fit. There don't seem to be a lot of worked examples out there so for
someone one day in the future here is some sample octave/matlab code
for nlsq.

A1 = 0.0233;
A2 = 4.4583;
A3 = 0.0082;

A = [ A1 A2 A3 ];

ts = (1:(30))';

Fage = @(A,x) A(1).*log(A(2).*x + 1) + A(3)

F = Fage(A,ts);

% Add Gaussian measurement noise
F_pn = F + 1E6*0.5e-9*randn(size(ts));

% Fit F_pn to non-linear function Fage
Ainit = [ 1 1 1 ];

options = optimset('Display','on','TolFun',1e-12, 'TolX', 1e-12);
[Aest,resnorm, e, o] = lsqcurvefit(Fage,Ainit,ts,F_pn,[],[],options);


On Thu, Nov 17, 2016 at 8:24 AM, Attila Kinali <att...@kinali.ch> wrote:
> On Tue, 15 Nov 2016 23:58:31 -0500
> Scott Stobbe <scott.j.sto...@gmail.com> wrote:
>
>> Do you recall if you fitted with true ordinary least squares, or fit with a
>> recursive/iterative approach in a least squares sense. If the aging curve
>> is linearizable, it isn't jumping out at me.
>
> Least square fits to non-linear functions are almost always iterative.
> There are only very few functions for which closed formulas are known.
> Quite a few people do "linearization", but in general this does not
> work well or leads to sub-optimal solutions (aka not an least squares fit)
> without people realizing it.
>
> An alternative method that works for some functions are state-space
> methods. But I barely know about them, so I cannot really comment on them.
>
>
> Attila Kinali
> --
> It is upon moral qualities that a society is ultimately founded. All
> the prosperity and technological sophistication in the world is of no
> use without that foundation.
>  -- Miss Matheson, The Diamond Age, Neil Stephenson
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

A few different plots. I didn't have an intuitive feel for what the B
coefficient in log term looks like on a plot, so that is the first
plot. The same aging curve is plotted three times, with the exception
of the B coefficient being scaled by 1/10, 1, 10 respectively. In hand
waving terms, it does have an enormous impact during the first 30 days
(or until Bt >>1), but from then on, it is just an additive offset.

The next 4 plots are just sample fits with noise added.

Finally the 6th plot is of just the first 30 days, the data would seem
to be cleaner than what was shown as a sample in the paper, but the
stability of the B coefficient in 10 monte-carlo runs is not great.
But when plotted over a year the results are minimal.

  A1  A2A3
 0.022914   6.8459   0.00016743
 0.022932   6.6702   0.00058768
 0.023206   5.79690.0026103
 0.023219   4.31270.0093793
  0.02374   2.8309 0.016838
 0.023119   5.02140.0061557
 0.023054   5.83990.0031886
 0.022782   9.8582   -0.0074089
 0.023279   3.7392 0.012161
  0.02345   4.10620.0095448

The only other thing to point out from this, is that the A2 and A3
coefficients are highly non-orthogonal, as A2 increases, A3 drops to
make up the difference.

On Wed, Nov 16, 2016 at 7:38 AM, Bob Camp <kb...@n1k.org> wrote:
> Hi
>
> The original introduction of 55310 written by a couple of *very* good guys:
>
> http://tycho.usno.navy.mil/ptti/1987papers/Vol%2019_16.pdf
>
> A fairly current copy of 55310:
>
> https://nepp.nasa.gov/DocUploads/1F3275A6-9140-4C0C-864542DBF16EB1CC/MIL-PRF-55310.pdf
>
> The “right” equation is on page 47. It’s the “Bt+1” in the log that messes up 
> the fit. If you fit it without
> the +1, the fit is *much* easier to do. The result isn’t quite right.
>
> Bob
>
>
>> On Nov 15, 2016, at 11:58 PM, Scott Stobbe <scott.j.sto...@gmail.com> wrote:
>>
>> Hi Bob,
>>
>> Do you recall if you fitted with true ordinary least squares, or fit with a
>> recursive/iterative approach in a least squares sense. If the aging curve
>> is linearizable, it isn't jumping out at me.
>>
>> If the model was hypothetically:
>> F = A ln( B*t )
>>
>> F = A ln(t) + Aln(B)
>>
>> which could easily be fit as
>> F  = A' X + B', where X = ln(t)
>>
>> It would appear stable32 uses an iterative approach for the non-linear
>> problem
>>
>> "y(t) = a·ln(bt+1), where slope = y'(t) = ab/(bt+1) Determining the
>> nonlinear log fit coefficients requires an iterative procedure. This
>> involves setting b to an in initial value, linearizing the equation,
>> solving for the other coefficients and the sum of the squared error,
>> comparing that with an error criterion, and iterating until a satisfactory
>> result is found. The key aspects to this numerical analysis process are
>> establishing a satisfactory iteration factor and error criterion to assure
>> both convergence and small residuals."
>>
>> http://www.stable32.com/Curve%20Fitting%20Features%20in%20Stable32.pdf
>>
>> Not sure what others do.
>>
>>
>> On Mon, Nov 14, 2016 at 7:15 AM, Bob Camp <kb...@n1k.org> wrote:
>>
>>> Hi
>>>
>>> If you already *have* data over a year (or multiple years) the fit is
>>> fairly easy.
>>> If you try to do this with data from a few days or less, the whole fit
>>> process is
>>> a bit crazy. You also have *multiple* time constants involved on most
>>> OCXO’s.
>>> The result is that an earlier fit will have a shorter time constant (and
>>> will ultimately
>>> die out). You may not be able to separate the 25 year curve from the 3
>>> month
>>> curve with only 3 months of data.
>>>
>>> Bob
>>>
>>>> On Nov 13, 2016, at 10:59 PM, Scott Stobbe <scott.j.sto...@gmail.com>
>>> wrote:
>>>>
>>>> On Mon, Nov 7, 2016 at 10:34 AM, Scott Stobbe <scott.j.sto...@gmail.com>
>>>> wrote:
>>>>
>>>>> Here is a sample data point taken from http://tycho.usno.navy.mil/ptt
>>>>> i/1987papers/Vol%2019_16.pdf; the first that showed up on a google
>>> search.
>>>>>
>>>>>Year   Aging [PPB]  dF/dt [PPT/Day]
>>>>>   1   180.51   63.884
>>>>>   2   196.6531.93
>>>>>   5  218   12.769
>>>>>   9   231.69   7.0934
>>>>>  10   234.156.384
>>&g

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

Hi Bob,

Do you recall if you fitted with true ordinary least squares, or fit with a
recursive/iterative approach in a least squares sense. If the aging curve
is linearizable, it isn't jumping out at me.

If the model was hypothetically:
 F = A ln( B*t )

 F = A ln(t) + Aln(B)

which could easily be fit as
 F  = A' X + B', where X = ln(t)

It would appear stable32 uses an iterative approach for the non-linear
problem

"y(t) = a·ln(bt+1), where slope = y'(t) = ab/(bt+1) Determining the
nonlinear log fit coefficients requires an iterative procedure. This
involves setting b to an in initial value, linearizing the equation,
solving for the other coefficients and the sum of the squared error,
comparing that with an error criterion, and iterating until a satisfactory
result is found. The key aspects to this numerical analysis process are
establishing a satisfactory iteration factor and error criterion to assure
both convergence and small residuals."

http://www.stable32.com/Curve%20Fitting%20Features%20in%20Stable32.pdf

Not sure what others do.


On Mon, Nov 14, 2016 at 7:15 AM, Bob Camp <kb...@n1k.org> wrote:

> Hi
>
> If you already *have* data over a year (or multiple years) the fit is
> fairly easy.
> If you try to do this with data from a few days or less, the whole fit
> process is
> a bit crazy. You also have *multiple* time constants involved on most
> OCXO’s.
> The result is that an earlier fit will have a shorter time constant (and
> will ultimately
> die out). You may not be able to separate the 25 year curve from the 3
> month
> curve with only 3 months of data.
>
> Bob
>
> > On Nov 13, 2016, at 10:59 PM, Scott Stobbe <scott.j.sto...@gmail.com>
> wrote:
> >
> > On Mon, Nov 7, 2016 at 10:34 AM, Scott Stobbe <scott.j.sto...@gmail.com>
> > wrote:
> >
> >> Here is a sample data point taken from http://tycho.usno.navy.mil/ptt
> >> i/1987papers/Vol%2019_16.pdf; the first that showed up on a google
> search.
> >>
> >> Year   Aging [PPB]  dF/dt [PPT/Day]
> >>1   180.51   63.884
> >>2   196.6531.93
> >>5  218   12.769
> >>9   231.69   7.0934
> >>   10   234.156.384
> >>   25255.5   2.5535
> >>
> >> If you have a set of coefficients you believe to be representative of
> your
> >> OCXO, we can give those a go.
> >>
> >>
> > I thought I would come back to this sample data point and see what the
> > impact of using a 1st order estimate for the log function would entail.
> >
> > The coefficients supplied in the paper are the following:
> > A1 = 0.0233;
> > A2 = 4.4583;
> > A3 = 0.0082;
> >
> > F =  A1*ln( A2*x +1 ) + A3;  where x is time in days
> >
> > Fdot = (A1*A2)/(A2*x +1)
> >
> > Fdotdot = -(A1*A2^2)/(A2*x +1)^2
> >
> > When x is large, the derivatives are approximately:
> >
> > Fdot ~= A1/x
> >
> > Fdotdot ~= -A1/x^2
> >
> > It's worth noting that, just as it is visually apparent from the graph,
> the
> > aging becomes more linear as time progresses, the second, third, ...,
> > derivatives drop off faster than the first.
> >
> > A first order taylor series of the aging would be,
> >
> > T1(x, xo) = A3 + A1*ln(A2*xo + 1) +  (A1*A2)(x - xo)/(A2*xo +1) + O(
> > (x-xo)^2 )
> >
> > The remainder (error) term for a 1st order taylor series of F would be:
> >  R(x) = Fdotdot(c) * ((x-xo)^2)/(2!);  where c is some value between
> x
> > and xo.
> >
> > So, take for example, forward projecting the drift one day after the
> 365th
> > day using a first order model,
> > xo = 365
> >
> > Fdot(365) =  63.796 PPT/day, alternatively the approximate derivative
> > is: 63.836 PPT/day
> >
> > |R(366)| =  0.087339 PPT (more than likely, this is no where near 1
> > DAC LSB on the EFC line)
> >
> > More than likely you wouldn't try to project 7 days out, but considering
> > only the generalized effects of aging, the error would be:
> >
> > |R(372)| = 4.282 PPT (So on the 7th day, a 1st order model starts to
> > degrade into a few DAC LSB)
> >
> > In the case of forward projecting aging for one day, using a 1st order
> > model versus the full logarithmic model, would likely be a discrepancy of
> > less than one dac LSB.
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] precision timing pulse

[Scott Stobbe]

If you are trying to avoid writing code, the 74hc might be worth taking
a look at.

On Tue, Nov 15, 2016 at 3:00 PM Chuck / Judy Burch 
wrote:

>
> I'm building a laboratory scintillation counter that uses my HP 5335a
> counter as a read-out.  The FREQ mode gives only approximate results
> (maximum gate time is about 5 seconds).  The TOT mode counts pulses for
> an arbitrary time that can be set using the "external arm input" on the
> rear panel.  So I need a timing pulse (of either polarity) of known and
> adjustable width (time).  One way to do this is with a PIC frequency
> divider taking the counter 10MHz output down to 1 PPS following that
> with two or three ripple counters to get a 1 - 5 - 10 - 50 ...
> sequence.  That I understand.
>
> How do I get a pulse to start with a push button and then stop for
> example 500 or 10,000 seconds later?
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

On Mon, Nov 7, 2016 at 10:34 AM, Scott Stobbe <scott.j.sto...@gmail.com>
wrote:

> Here is a sample data point taken from http://tycho.usno.navy.mil/ptt
> i/1987papers/Vol%2019_16.pdf; the first that showed up on a google search.
>
>  Year   Aging [PPB]  dF/dt [PPT/Day]
> 1   180.51   63.884
> 2   196.6531.93
> 5  218   12.769
> 9   231.69   7.0934
>10   234.156.384
>25255.5   2.5535
>
> If you have a set of coefficients you believe to be representative of your
> OCXO, we can give those a go.
>
>
I thought I would come back to this sample data point and see what the
impact of using a 1st order estimate for the log function would entail.

The coefficients supplied in the paper are the following:
 A1 = 0.0233;
 A2 = 4.4583;
 A3 = 0.0082;

F =  A1*ln( A2*x +1 ) + A3;  where x is time in days

 Fdot = (A1*A2)/(A2*x +1)

 Fdotdot = -(A1*A2^2)/(A2*x +1)^2

When x is large, the derivatives are approximately:

 Fdot ~= A1/x

 Fdotdot ~= -A1/x^2

It's worth noting that, just as it is visually apparent from the graph, the
aging becomes more linear as time progresses, the second, third, ...,
derivatives drop off faster than the first.

A first order taylor series of the aging would be,

 T1(x, xo) = A3 + A1*ln(A2*xo + 1) +  (A1*A2)(x - xo)/(A2*xo +1) + O(
(x-xo)^2 )

The remainder (error) term for a 1st order taylor series of F would be:
  R(x) = Fdotdot(c) * ((x-xo)^2)/(2!);  where c is some value between x
and xo.

So, take for example, forward projecting the drift one day after the 365th
day using a first order model,
 xo = 365

 Fdot(365) =  63.796 PPT/day, alternatively the approximate derivative
is: 63.836 PPT/day

 |R(366)| =  0.087339 PPT (more than likely, this is no where near 1
DAC LSB on the EFC line)

More than likely you wouldn't try to project 7 days out, but considering
only the generalized effects of aging, the error would be:

 |R(372)| = 4.282 PPT (So on the 7th day, a 1st order model starts to
degrade into a few DAC LSB)

In the case of forward projecting aging for one day, using a 1st order
model versus the full logarithmic model, would likely be a discrepancy of
less than one dac LSB.
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] quartz drift rates, linear or log

[Scott Stobbe]

Those are wonderful plots :)

I vaguely recall that a 1ppm frequency shift is approximately equivalent to
the mass transfer of one molecular layer of a crystal. So at some point
your counting atoms if there was no noise, thermal disturbance, mechanical
disturbance...

On Sat, Nov 12, 2016 at 5:00 PM Tom Van Baak  wrote:

> There were postings recently about OCXO ageing, or drift rates.
>
> I've been testing a batch of TBolts for a couple of months and it provides
> an interesting set of data from which to make visual answers to recent
> questions. Here are three plots.
>
>
> 1) attached plot: TBolt-10day-fit0-e09.gif (
> http://leapsecond.com/pages/tbolt/TBolt-10day-fit0-e09.gif )
>
> A bunch of oscillators are measured with a 20-channel system. Each
> frequency plot is a free-running TBolt (no GPS, no disciplining). The
> X-scale is 10 days and the Y-scale is 1 ppb, or 1e-9 per Y-division. What
> you see at this scale is that all the OCXO are quite stable. Also, some of
> them show drift.
>
> For example, the OCXO frequency in channel 14 changes by 2e-9 in 10 days
> for a drift rate of 2e-10/day. It looks large in this plot but its well
> under the typical spec, such as 5e-10/day for a 10811A. We see a variety of
> drift rates, including some that appear to be zero: flat line. At this
> scale, CH13, for example, seems to have no drift.
>
> But the drift, when present, appears quite linear. So there are two things
> to do. Zoom in and zoom out.
>
>
> 2) attached plot: TBolt-10day-fit0-e10.gif (
> http://leapsecond.com/pages/tbolt/TBolt-10day-fit0-e10.gif )
>
> Here we zoom in by changing the Y-scale to 1e-10 per division. The X-scale
> is still 10 days. Now we can see the drift much better. Also at this level
> we can see instability of each OCXO (or the lab environment). At this
> scale, channels CH10 and CH14 are "off the chart". An OCXO like the one in
> CH01 climbs by 2e-10 over 10 days for a drift rate of 2e-11/day. This is
> 25x better than the 10811A spec. CH13, mentioned above, is not zero drift
> after all, but its drift rate is even lower, close to 1e-11/day.
>
> For some oscillators the wiggles in the data (frequency instability) are
> large enough that the drift rate is not clearly measurable.
>
> The 10-day plots suggests you would not want to try to measure drift rate
> based on just one day of data.
>
> The plots also suggest that drift rate is not a hard constant. Look at any
> of the 20 10-day plots. Your eye will tell you that the daily drift rate
> can change significantly from day to day to day.
>
> The plots show that an OCXO doesn't necessarily follow strict rules. In a
> sense they each have their own personality. So one needs to be very careful
> about algorithms that assume any sort of constant or consistent behavior.
>
>
> 3) attached plot: TBolt-100day-fit0-e08.gif (
> http://leapsecond.com/pages/tbolt/TBolt-100day-fit0-e08.gif )
>
> Here we look at 100 days of data instead of just 10 days. To fit, the
> Y-scale is now 1e-8 per division. Once a month I created a temporary
> thermal event in the lab (the little "speed bumps") which we will ignore
> for now.
>
> At this long-term scale, OCXO in CH09 has textbook logarithmic drift. Also
> CH14 and CH16. In fact over 100 days most of them are logarithmic but the
> coefficients vary considerably so it's hard to see this at a common scale.
> Note also the logarithmic curve is vastly more apparent in the first few
> days or weeks of operation, but I don't have that data.
>
> In general, any exponential or log or parabolic or circular curve looks
> linear if you're looking close enough. A straight highway may look linear
> but the equator is circular. So most OCXO drift (age) with a logarithmic
> curve and this is visible over long enough measurements. But for shorter
> time spans it will appear linear. Or, more likely, internal and external
> stability issues will dominate and this spoils any linear vs. log
> discussion.
>
> So is it linear or log? The answer is it depends. Now I sound like Bob ;-)
>
> /tvb
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Need some wisdom from the cesium beam tube gurus out there

[Scott Stobbe]

If you want sub degree precision, you will need to make your connections to
dissimilar metals on an isothermal boundary, a terminal block is better
than clips in free air.

On Fri, Nov 11, 2016 at 8:28 AM, Bob Camp  wrote:

> Hi
>
> > On Nov 11, 2016, at 8:02 AM, jimlux  wrote:
> >
> > On 11/10/16 10:28 PM, Mike Millen wrote:
> >> It would work as well if you used a pair of regular copper wires to
> >> connect the meter to the thermocouple...
> >>
> >> The junctions created by all the new connections will cancel out.
> >>
> >
> > as long as the temperatures are "exactly" the same,
> > (Seebeck coefficient varies with temperature)
> > and the two metals at the junctions are the same,
> > (ditto, but the curves are different for different materials)
> > and the mechanical configuration is the same
> > (current density also affects it)
>
> The gotcha is that few of us weld copper directly to the thermocouple
> leads. The far more
> common approach is to grab clip leads. At least around here, the clips on
> the leads are
> not made of copper. They are some sort of (badly worn) plating over
> (oxidized) base
> material.
>
> I grab a “copper wire” clip lead and hook up to the thermocouple. There
> isn’t a lot of
> delta T in most bench situations. In this case you have a heated gizmo
> warming things up ….
> Who knows what the delta T may be or how small the contact area actually
> is.
>
> Simple answer: Don’t trust the first number you get. Try it a couple of
> times with *different*
> leads. Make sure you do indeed get within a degree or three on each of
> them. Depending on
> how you have your cold junction set up, that may also need the same
> treatment.
>
> Bob
>
>
> >
> >
> > For run of the mill "measure to 1 degree at room temperature" you can
> probably make that assumption.
> >
> > But if you're looking for precision, you need to take this stuff into
> account (that's what "cold junction compensation" is all about.. )
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] How to get PPS from ublox mini-PCI GPS to APU2 SoC serial port for ntpd

[Scott Stobbe]

When 1PPS is implemented as a time mark signal narrow width (versus 1 Hz
50% duty cycle) and you time stamp both edges, you can always tell which is
supposed to be the leading edge. Either the assert and deassert are close
together in the correct case or far apart when using incorrect polarity.

On Thu, Nov 10, 2016 at 10:22 AM Chris Albertson 
wrote:

> The problem here is "real world".  Yes in theory you can do it perfectly
> but in the real world do all makes and models of GPS receivers get it
> right?   I would not bet on it.
>
> This is just like the argument over using NMEA only for timing.  Some GPS
> receivers push out NEMA sentences with very little jitter and others only
> follow the NMEA spec with only requires plus or minus one second accuracy
> and then there is every shade of grey between.
>
> My conservative engineering background tells me that unless the variation
> of the pulse width is species you should assume the width is not controlled
> at all.  It may even look good in testing but a firmware update would kill
> that.  Likely in there is no regression test for not specified features.
> People now days who write software (or firmware especially) will build an
> automated test suit that lets them checks f the software still works after
> making changes.  They are careful to test each written requirement.
>
> The UART on the PC can interrupt on either raising or facing edges so if
> the PPS is inverted you interrupt on the falling edge.   But you have to
> get this right too
>
> one IMPORTENT thing to  get right is to remember that on modern control
> line under RS232 the logic is "active low".  TTL logic is "active high"
> where a higher voltage means "one" and allow voltage means "zero" but this
> is backwards for modern control under RS232   So, it is WAY-EASY to get it
> wrong when connecting a TTL PPS to a modern control line. You might
> need an inverter because many level shifters invert the signal
>
> The one thing that helps is the at 1Hz the signal is slow enough to see on
> a volt meter, Even an old VTVM reacts fast enough
>
> On Wed, Nov 9, 2016 at 2:44 PM, Tom Van Baak  wrote:
>
> > Bob,
> >
> > The PIC's I use have essentially no jitter. If they generate a 1PPS the
> > edge and the pulse width are perfect, down to picoseconds. The talk about
> > "other stuff" and "priority" and "number of compares" and "ambiguity" is
> > worrisome. It sounds like a design or coding flaw to me, like what
> happens
> > when people try to do precise time with a high level language.
> >
> > /tvb
> >
> > - Original Message -
> > From: "Bob Stewart" 
> > To: "Discussion of precise time and frequency measurement" <
> > time-nuts@febo.com>
> > Sent: Wednesday, November 09, 2016 9:30 AM
> > Subject: Re: [time-nuts] How to get PPS from ublox mini-PCI GPS to APU2
> > SoC serial port for ntpd
> >
> >
> > I'd like to comment on the idea of measuring the width of the pulse. My
> > experience with creating the 1PPS from an interrupt is that it's fairly
> > straightforward how to do the set interrupt: The interrupt happens, you
> > execute one, maybe two instructions to raise the output pin, and you
> leave
> > the interrupt routine. But, resetting the pin is a different story.
> Unless
> > you've got a lot of interrupt vectors to play with, the reset part of the
> > 1PPS signal is delegated to a general purpose timing interrupt where
> you're
> > doing a lot of other stuff - and it has a somewhat lower priority. So,
> you
> > wind up doing a number of compares to see if you should reset the pin,
> > which adds some ambiguity to the set/reset times.
> > Granted, I'm doing this on a general purpose PIC, but I have read
> comments
> > about various receivers having some jitter on the reset side of their
> 1PPS
> > pulse.
> > Bob
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> > mailman/listinfo/time-nuts
> > and follow the instructions there.
> >
>
>
>
> --
>
> Chris Albertson
> Redondo Beach, California
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] How to get PPS from ublox mini-PCI GPS to APU2 SoC serial port for ntpd

[Scott Stobbe]

FWIW, the 16450 uart can interrupt on any modem line change, whether it be
rising or falling.

On Wed, Nov 9, 2016 at 11:41 AM, Chris Albertson 
wrote:

> On Tue, Nov 8, 2016 at 11:08 PM, Hal Murray 
> wrote:
>
> >
> > g...@rellim.com said:
> > > Who cares if the PPS is inverted?
> >
>
> If the signal is NOT inverted then the raising edge will be within
> nanoseconds of the UTC seconds "tick".The inverted PPS will have the
> edge delayed somewhat.   It is not even clear if the length of the delay is
> constant.
>
> If using the PPS for timing and the PPS is inverted you will have to
> measure the delay length an account for in the config file.
>
>
>
>
> --
>
> Chris Albertson
> Redondo Beach, California
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] How to get PPS from ublox mini-PCI GPS to APU2 SoC serial port for ntpd

[Scott Stobbe]

Just make sure you get a board with more than one Rx/Tx pair, unless you
are happy with two boards; one for Rx/Tx and one for PPS.

On Wed, Nov 9, 2016 at 2:08 AM, Hal Murray  wrote:

>
> g...@rellim.com said:
> > Who cares if the PPS is inverted?
>
> Good point.  I wasn't paying attention to the PPS only context in the
> subject.  I keep forgetting which way is best for PPS.  Fortunately, all
> the
> software I've used knows how to handle inverted PPS.
>
> The original message in this thread also mentioned RX and TX.
>
> I'd probably get the RS-232 version since I'm likely to want the RX/TX too
> and it's more likely to be useful on other projects.
>
>
> --
> These are my opinions.  I hate spam.
>
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

Deeper into the rabbit hole :)

I'm not sure what specific pic you are you using but most of them have at
least one timer that will run fully asynchronously and its timer input is
usually shared with one of the 32k osc pins. Which you can then use to wake
the prossesor from one of its sleep states. Do you end up hitting a clock
sync with the pwm block?

Good old 74 series to the rescue.

On Monday, 7 November 2016, Bob Stewart <b...@evoria.net> wrote:

> Hi Scott,
>
> I wish I had some long term data, but I don't.  I had initially set out to
> build an accurate GPS frequency reference type of GPSDO.  So, aging wasn't
> an issue.  It's either on and locked or it's not.  So, I didn't worry about
> leaving a unit running for months and collecting aging data.
>
> But then I had this crackpot idea of using the latches in a 7474 to
> stabilize the 1PPS from the receiver to the OCXO.  Tom had a bit of a
> misunderstanding about what I was doing and we had a bit of a discussion
> until he caught on to what I was doing and its limitations.
>
> But, somewhere along the line, I realized that the idea was sound but my
> implementation was poor.  The basic problem with using a timer in the
> dsPIC33 is that they use a PLL to generate the internal clock - even if you
> supply a clock.  That gives you a 1-count jitter in any output pulse you
> try to create.  So, I realized that I could use the latches in a 7474 to
> latch the OCXO to the output of a timer on the PIC.  With the PIC running
> at 40MHz, I have plenty of room for the jitter without the worry of a phase
> slip.  I think I've proved that that works, so now I have the possibility
> of using my system as a time server.  And that means I now have to deal
> with such arcane matters as holdover, aging, and generating the time from
> the OCXO.  The learning curve has been a bit steep.
>
> Bob
>
> -
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>
>
> --
> *From:* Scott Stobbe <scott.j.sto...@gmail.com
> <javascript:_e(%7B%7D,'cvml','scott.j.sto...@gmail.com');>>
> *To:* Bob Camp <kb...@n1k.org
> <javascript:_e(%7B%7D,'cvml','kb...@n1k.org');>>
> *Cc:* Bob Stewart <b...@evoria.net
> <javascript:_e(%7B%7D,'cvml','b...@evoria.net');>>; Discussion of precise
> time and frequency measurement <time-nuts@febo.com
> <javascript:_e(%7B%7D,'cvml','time-nuts@febo.com');>>
> *Sent:* Monday, November 7, 2016 9:34 AM
> *Subject:* Re: [time-nuts] Thermal impact on OCXO
>
> Here is a sample data point taken from http://tycho.usno.navy.mil/
> ptti/1987papers/Vol%2019_16.pdf; the first that showed up on a google
> search.
>
>  Year   Aging [PPM]  dF/dt [PPT/Day]
> 1   180.51   63.884
> 2   196.6531.93
> 5  218   12.769
> 9   231.69   7.0934
>10   234.156.384
>25255.5   2.5535
>
> If you have a set of coefficients you believe to be representative of your
> OCXO, we can give those a go.
>
> On Mon, Nov 7, 2016 at 7:08 AM, Bob Camp <kb...@n1k.org
> <javascript:_e(%7B%7D,'cvml','kb...@n1k.org');>> wrote:
>
> Hi
>
>
> > On Nov 5, 2016, at 10:43 PM, Bob Stewart <b...@evoria.net
> <javascript:_e(%7B%7D,'cvml','b...@evoria.net');>> wrote:
> >
> > Hi Scott,
> > D'oh.  Thanks for the correction!  Like I said, I don't do these
> calculations often.
> >
> > If as Bob Camp implies, the aging isn't from the OXCO, then I'm a bit
> stumped.  I do have an op-amp in the EFC string with  a voltage divider for
> gain.  The resistors are Panasonic ERA-6AEDxxxV resistors.  Mouser says
> they're temperature stable to 25PPM/C, but of course they don't mention an
> aging rate.  I don't really see anything else, other than the OCXO, that is
> likely to be prone to a linear type of aging.
>
> OCXO’s don’t age in a linear fashion. At least 90% of them don’t. If you
> dig into the FCS papers there are various
> curves proposed as models. Mil-O-55310 has one of them as the “official”
> approach. All of them have the basic
> issue of mistakenly fitting to to short a time constraint.
>
> Bob
>
> >  The aging rate appears to be stable from unit to unit, so naturally I
> considered the OCXO first.
> >
> > There is one other bit in the EFC string that might be controversial,
> but I don't see that it would be a candidate for the symptoms of aging.
> >
> > Bob
> >  -- -- -
> > AE6RV.co

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

Another nice plot! It looks like after 2am you see temperature swings of
1.5 degF roughly every 30 minutes? Correspondingly, the EFC line which is
nominally ~2.8vdc sees swings of +-50 uV?

On Monday, 7 November 2016, Bob Stewart  wrote:

> Hi guys,
> First of all, thanks for the additional responses.  I was a bit angry and
> rude yesterday, and I figured this thread was over.  Thanks for staying
> with me.
> I haven't had time to look over the data etc in your responses.  I'll do
> that and get back to the list if appropriate.
>
> I spoke to Attila and Azelio offline last nite and from their input, I
> decided to hook up the 3456A and collect some data, which is in the plot
> attached.  As usual, I've modified one of my standard plot scripts, so
> there is some extraneous data that wasn't removed.
>
> First for our purposes is the thin red line, which is the DAC value locked
> at 0x734B0.  The orange trace is the temperature adjusted so that each step
> of 10 on the right hand Y tics is one degree F.  The dark blue trace is the
> EFC value read by the 3456A.  It has been multiplied by 100,000 and then
> had 282600 subtracted.  This leaves just the LSD scaled at 1:1 on the right
> hand Y tics.
> There's something interesting on the far right hand side where the
> temperature goes low and stays there.  The DVM value follows it down, but
> then recovers while the temperature stays down.  I'm not sure what to make
> of this.  Either then OCXO is making up for the temperature change by
> increasing the temperature, or the 3456 is compensating for it after the
> fact.  In either case, the EFC seems to only follow the transient
> temperature changes, and doesn't actually track the temperature on the
> board.
>
> So, to my eye, after 14 hours, there is only a dependency on thermal
> transients.  I'll leave it running for some time yet, but the EFC doesn't
> seem to be drifting in any meaningful way at this point, other than in
> relation to temperature changes.
>
> Note:  I used a shielded twisted-pair with the usual clips attached to
> ground and EFC in my GPSDO.  On the 3456, the two leads go to the
> appropriate volts inputs, and the shield goes to the ground input.  The
> "guard" switch is out, which is the off position.  There is no shield
> connector on the DUT side.
>
> Bob
>
>  -
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>
>
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

Typo, "Aging [PPM]" should read "Aging [PPB]".

On Mon, Nov 7, 2016 at 10:34 AM, Scott Stobbe <scott.j.sto...@gmail.com>
wrote:

> Here is a sample data point taken from http://tycho.usno.navy.mil/
> ptti/1987papers/Vol%2019_16.pdf; the first that showed up on a google
> search.
>
>  Year   Aging [PPM]  dF/dt [PPT/Day]
> 1   180.51   63.884
> 2   196.6531.93
> 5  218   12.769
> 9   231.69   7.0934
>10   234.156.384
>25255.5   2.5535
>
> If you have a set of coefficients you believe to be representative of your
> OCXO, we can give those a go.
>
> On Mon, Nov 7, 2016 at 7:08 AM, Bob Camp <kb...@n1k.org> wrote:
>
>> Hi
>>
>>
>> > On Nov 5, 2016, at 10:43 PM, Bob Stewart <b...@evoria.net> wrote:
>> >
>> > Hi Scott,
>> > D'oh.  Thanks for the correction!  Like I said, I don't do these
>> calculations often.
>> >
>> > If as Bob Camp implies, the aging isn't from the OXCO, then I'm a bit
>> stumped.  I do have an op-amp in the EFC string with  a voltage divider for
>> gain.  The resistors are Panasonic ERA-6AEDxxxV resistors.  Mouser says
>> they're temperature stable to 25PPM/C, but of course they don't mention an
>> aging rate.  I don't really see anything else, other than the OCXO, that is
>> likely to be prone to a linear type of aging.
>>
>> OCXO’s don’t age in a linear fashion. At least 90% of them don’t. If you
>> dig into the FCS papers there are various
>> curves proposed as models. Mil-O-55310 has one of them as the “official”
>> approach. All of them have the basic
>> issue of mistakenly fitting to to short a time constraint.
>>
>> Bob
>>
>> >  The aging rate appears to be stable from unit to unit, so naturally I
>> considered the OCXO first.
>> >
>> > There is one other bit in the EFC string that might be controversial,
>> but I don't see that it would be a candidate for the symptoms of aging.
>> >
>> > Bob
>> >  -
>> > AE6RV.com
>> >
>> > GFS GPSDO list:
>> > groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>> >
>> >  From: Scott Stobbe <scott.j.sto...@gmail.com>
>> > To: Bob Stewart <b...@evoria.net>
>> > Cc: Discussion of Precise Time and Frequency Measurement <
>> time-nuts@febo.com>
>> > Sent: Saturday, November 5, 2016 9:19 PM
>> > Subject: Re: [time-nuts] Thermal impact on OCXO
>> >
>> > If your DAC spans the full EFC range than 1LSB is 1/2^20 ~ 1 PPM of the
>> EFC range, and the EFC tuning range is 8/10E6 ~ 1 PPM full scale, so 1 LSB
>> is ~1PPT. So, if everything else is stable the DAC code reflects changes
>> solely due to the OCXO, which would be an aging of 24 PPT/day.
>> > On Sat, Nov 5, 2016 at 9:57 PM, Bob Stewart <b...@evoria.net> wrote:
>> >
>> > Hi Scott,
>> > The 20 bits span about 6 volts.  The EFC range spans about 8Hz
>> (+/-4Hz).  I don't do these calculations every day, but that's about 4.5PPT?
>> > Bob   -- --
>> -
>> > AE6RV.com
>> >
>> > GFS GPSDO list:
>> > groups.yahoo.com/neo/groups/ GFS-GPSDOs/info
>> >
>> >
>> >
>> > From: Scott Stobbe <scott.j.sto...@gmail.com>
>> > To: Bob Stewart <b...@evoria.net>; Discussion of precise time and
>> frequency measurement <time-nuts@febo.com>
>> > Sent: Saturday, November 5, 2016 8:38 PM
>> > Subject: Re: [time-nuts] Thermal impact on OCXO
>> >
>> > I think that's a nice plot, it looks like you have stepped 160 LSB over
>> 7 days or roughly 1 LSB per hour. With a 20bit dac you are trimming maybe 1
>> ppt/LSB to 4 ppt/LSB? In allan devation terms, the case of 1ppt/LSB, solely
>> due to drift, you're at 1E-12 at 3600*sqrt(2) = 5000 s, in the case of
>> 4ppt/hour your at 1E-12 at 1280 s. Seems reasonable.
>> > On Sat, Nov 5, 2016 at 2:47 PM, Bob Stewart <b...@evoria.net> wrote:
>> >
>> > Oh dear.  I attached the wrong file.  Here's the correct one.
>> >  - -- --
>> > AE6RV.com
>> >
>> > GFS GPSDO list:
>> > groups.yahoo.com/neo/groups/ GFS-GPSDOs/info
>> >
>> >
>> >
>> > __ _
>> > time-nuts mailing list -- time-nuts@febo.com
>> > To unsubscribe, go to https://www.febo.com/cgi-bin/
>> mailman/listinfo/time-nuts
>> > and follow the instructions there.
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> > ___
>> > time-nuts mailing list -- time-nuts@febo.com
>> > To unsubscribe, go to https://www.febo.com/cgi-bin/m
>> ailman/listinfo/time-nuts
>> > and follow the instructions there.
>>
>>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

Here is a sample data point taken from
http://tycho.usno.navy.mil/ptti/1987papers/Vol%2019_16.pdf; the first that
showed up on a google search.

 Year   Aging [PPM]  dF/dt [PPT/Day]
1   180.51   63.884
2   196.6531.93
5  218   12.769
9   231.69   7.0934
   10   234.156.384
   25255.5   2.5535

If you have a set of coefficients you believe to be representative of your
OCXO, we can give those a go.

On Mon, Nov 7, 2016 at 7:08 AM, Bob Camp <kb...@n1k.org> wrote:

> Hi
>
>
> > On Nov 5, 2016, at 10:43 PM, Bob Stewart <b...@evoria.net> wrote:
> >
> > Hi Scott,
> > D'oh.  Thanks for the correction!  Like I said, I don't do these
> calculations often.
> >
> > If as Bob Camp implies, the aging isn't from the OXCO, then I'm a bit
> stumped.  I do have an op-amp in the EFC string with  a voltage divider for
> gain.  The resistors are Panasonic ERA-6AEDxxxV resistors.  Mouser says
> they're temperature stable to 25PPM/C, but of course they don't mention an
> aging rate.  I don't really see anything else, other than the OCXO, that is
> likely to be prone to a linear type of aging.
>
> OCXO’s don’t age in a linear fashion. At least 90% of them don’t. If you
> dig into the FCS papers there are various
> curves proposed as models. Mil-O-55310 has one of them as the “official”
> approach. All of them have the basic
> issue of mistakenly fitting to to short a time constraint.
>
> Bob
>
> >  The aging rate appears to be stable from unit to unit, so naturally I
> considered the OCXO first.
> >
> > There is one other bit in the EFC string that might be controversial,
> but I don't see that it would be a candidate for the symptoms of aging.
> >
> > Bob
> >  ---------
> > AE6RV.com
> >
> > GFS GPSDO list:
> > groups.yahoo.com/neo/groups/GFS-GPSDOs/info
> >
> >  From: Scott Stobbe <scott.j.sto...@gmail.com>
> > To: Bob Stewart <b...@evoria.net>
> > Cc: Discussion of Precise Time and Frequency Measurement <
> time-nuts@febo.com>
> > Sent: Saturday, November 5, 2016 9:19 PM
> > Subject: Re: [time-nuts] Thermal impact on OCXO
> >
> > If your DAC spans the full EFC range than 1LSB is 1/2^20 ~ 1 PPM of the
> EFC range, and the EFC tuning range is 8/10E6 ~ 1 PPM full scale, so 1 LSB
> is ~1PPT. So, if everything else is stable the DAC code reflects changes
> solely due to the OCXO, which would be an aging of 24 PPT/day.
> > On Sat, Nov 5, 2016 at 9:57 PM, Bob Stewart <b...@evoria.net> wrote:
> >
> > Hi Scott,
> > The 20 bits span about 6 volts.  The EFC range spans about 8Hz
> (+/-4Hz).  I don't do these calculations every day, but that's about 4.5PPT?
> > Bob   -- -- -
> > AE6RV.com
> >
> > GFS GPSDO list:
> > groups.yahoo.com/neo/groups/ GFS-GPSDOs/info
> >
> >
> >
> > From: Scott Stobbe <scott.j.sto...@gmail.com>
> > To: Bob Stewart <b...@evoria.net>; Discussion of precise time and
> frequency measurement <time-nuts@febo.com>
> > Sent: Saturday, November 5, 2016 8:38 PM
> > Subject: Re: [time-nuts] Thermal impact on OCXO
> >
> > I think that's a nice plot, it looks like you have stepped 160 LSB over
> 7 days or roughly 1 LSB per hour. With a 20bit dac you are trimming maybe 1
> ppt/LSB to 4 ppt/LSB? In allan devation terms, the case of 1ppt/LSB, solely
> due to drift, you're at 1E-12 at 3600*sqrt(2) = 5000 s, in the case of
> 4ppt/hour your at 1E-12 at 1280 s. Seems reasonable.
> > On Sat, Nov 5, 2016 at 2:47 PM, Bob Stewart <b...@evoria.net> wrote:
> >
> > Oh dear.  I attached the wrong file.  Here's the correct one.
> >  - -- --
> > AE6RV.com
> >
> > GFS GPSDO list:
> > groups.yahoo.com/neo/groups/ GFS-GPSDOs/info
> >
> >
> >
> > __ _
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP 5275A

[Scott Stobbe]

You could try heating the transistor and bushing then give the to5 a shot
of cold spray or supper duster hoping it will shrink enough to slide out.

On Sunday, 6 November 2016, Adrian Godwin  wrote:

> Slightly off-topic, as this is a general repair question. But it's a TIC.
>
> I'm repairing a 5275A timer (all-discreet count logic to 100MHz, neon bulb
> display, a most amazing bcd to decimal decoder made from neons and LDRs,
> 1-2-2-4 decade counters ..) and the current problem is a 2n1038-2 germanium
> T05 transistor in the power supply.
>
> It's mounted in an aluminium bush which is then isolated from the chassis.
> I don't think the bush is also a collet but I can't see how to remove the
> transistor. It resists ungentle pushing .. should I push it with a hammer ?
> Or is there a kinder way ?
>
> (I'm hoping to eventually put this into a system with a 101A oscillator and
> a 9815A calculator to measure the ADEV of a Boule electric pendulum clock).
> ___
> time-nuts mailing list -- time-nuts@febo.com 
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] I love the smell of tantalum in the morning

[Scott Stobbe]

In the case of the rt7 (or knife soldering tips for other brands) you get a
fine point that easily handles 30-32 awg wire, and the edge of the tip is a
little over 100 thou long, so for parts 0805 and smaller you just lay the
edge along the side of the component and heat both pads and swipe it to the
side.

On Sunday, 6 November 2016, Bryan _  wrote:

> If the SMD is small enough I have found it easy to remove by just applying
> a blob of solder to one end, this will quickly gap over to the other lead,
> and the SMD component and solder ball just slides off the board. Only works
> on the very small components though.
>
>
> -=Bryan=-
>
>
> 
> From: time-nuts > on behalf of
> Robert LaJeunesse >
> Sent: November 6, 2016 8:42 AM
> To: time-nuts@febo.com 
> Subject: Re: [time-nuts] I love the smell of tantalum in the morning
>
> Another admittedly low-budget way of heating both ends is to use a heat
> spreader. Solder a short piece of braid to one component end, fold it over
> the top of the part, and solder it to the other end. Heat the braid in the
> center, add solder until both ends are melted, and lift the combination off
> with tweezers. Sometimes a heavy enough copper wire bent around the part
> will also work as the heat spreader..
>
> Bob LaJeunesse
>
> > Sent: Saturday, November 05, 2016 at 9:05 PM
> > From: "Andy ZL3AG via time-nuts" >
> > To: "Discussion of precise time and frequency measurement" <
> time-nuts@febo.com >
> > Subject: Re: [time-nuts] I love the smell of tantalum in the morning
> >
> >
> > Yes. Unless you're grinding it away with a dremel (which I wouldn't
> recommend as far as chemical dust is concerned), nibbling away with
> sidecutters would be trying to force the 2 ends of the component apart.
> That may be stressing the pads they're soldered to, leading to a possible
> pad lifting at some stage.
> >
> > Any of the methods mentioned that heat both ends at the same time -
> allowing the component to be wiped off the board - would have to be the
> best, stress-wise.
> ___
> time-nuts mailing list -- time-nuts@febo.com 
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> time-nuts Info Page - American Febo Enterprises com/cgi-bin/mailman/listinfo/time-nuts>
> www.febo.com
> time-nuts is a low volume, high SNR list for the discussion of precise
> time and frequency measurement and related topics. To see the collection of
> prior postings to ...
>
>
>
> and follow the instructions there.
> ___
> time-nuts mailing list -- time-nuts@febo.com 
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] I love the smell of tantalum in the morning

[Scott Stobbe]

I would not recommend purchasing soldering tweezers without trying them
first. They are not easy to control solder application when mounting a
component.

 I do really like the Weller rt7 knife tip.

On Sunday, 6 November 2016, Adrian Godwin  wrote:

> SMD parts aren't going to go away. It's worth investing in good tools to
> deal with them and learning the best way to use them.
>
> I'd put tweezers quite high on that list, and although the metcals are
> worthwhile if you can avoid full price, the chinese tweezers are
> surprisingly useful at very low cost. Buy them to explore their deficiences
> and buy better when you know what you really need.
>
> On Sun, Nov 6, 2016 at 4:53 PM, Clint Jay  > wrote:
>
> > Heating one end and adding a little solder to the joint will allow you to
> > lift the cap,  the leads are folded over tabs so they'll bend nicely and
> > allow the cap to lift,  once you've got one end lifted,  heat the other
> and
> > it will come away easily.
> >
> > Clean up the pads with solder wick then you're good to replace the part.
> >
> > While I'd like an excuse (and the funds)  to buy a pair of tweezers for
> > SMD,  I'd find it hard to use that job as an excuse.
> >
> > On 5 Nov 2016 19:12, "Tom Van Baak" >
> wrote:
> >
> > > See C13 in the attached photo. I need to replace some blown caps on a
> few
> > > boards [1]. In one instance the cap got so hot it melted itself off the
> > > board. Quiet convenient, actually -- it acts like its own fuse -- but I
> > > don't think the 5071 designers had that clever feature in mind.
> > >
> > > Having not done SMT before, how should I do it with minimal risk to the
> > > very precious PCB. Or, what equipment should I use this as a good
> excuse
> > to
> > > buy?
> > >
> > > Thanks,
> > > /tvb
> > >
> > > [0] http://www.imdb.com/title/tt0078788/quotes
> > > [1] http://leapsecond.com/museum/hp5071a/A1-mother.htm
> > >
> > > ___
> > > time-nuts mailing list -- time-nuts@febo.com 
> > > To unsubscribe, go to https://www.febo.com/cgi-bin/
> > > mailman/listinfo/time-nuts
> > > and follow the instructions there.
> > >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com 
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> > mailman/listinfo/time-nuts
> > and follow the instructions there.
> >
> ___
> time-nuts mailing list -- time-nuts@febo.com 
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

Well bob's comments & caution are accurate, everything drifts. In your
case, if the OCXO is rock solid then you would see a 160 PPM change on the
EFC line over 7 days which is a 1mV change on your 6 V full scale, which is
fairly easy to measure if you have a 6 1/2 digit DMM.

On Sat, Nov 5, 2016 at 10:43 PM, Bob Stewart <b...@evoria.net> wrote:

> Hi Scott,
>
> D'oh.  Thanks for the correction!  Like I said, I don't do these
> calculations often.
>
> If as Bob Camp implies, the aging isn't from the OXCO, then I'm a bit
> stumped.  I do have an op-amp in the EFC string with  a voltage divider for
> gain.  The resistors are Panasonic ERA-6AEDxxxV resistors.  Mouser says
> they're temperature stable to 25PPM/C, but of course they don't mention an
> aging rate.  I don't really see anything else, other than the OCXO, that is
> likely to be prone to a linear type of aging.  The aging rate appears to be
> stable from unit to unit, so naturally I considered the OCXO first.
>
> There is one other bit in the EFC string that might be controversial, but
> I don't see that it would be a candidate for the symptoms of aging.
>
> Bob
>
>
> -
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>
>
> --
> *From:* Scott Stobbe <scott.j.sto...@gmail.com>
> *To:* Bob Stewart <b...@evoria.net>
> *Cc:* Discussion of Precise Time and Frequency Measurement <
> time-nuts@febo.com>
> *Sent:* Saturday, November 5, 2016 9:19 PM
> *Subject:* Re: [time-nuts] Thermal impact on OCXO
>
> If your DAC spans the full EFC range than 1LSB is 1/2^20 ~ 1 PPM of the
> EFC range, and the EFC tuning range is 8/10E6 ~ 1 PPM full scale, so 1 LSB
> is ~1PPT. So, if everything else is stable the DAC code reflects
> changes solely due to the OCXO, which would be an aging of 24 PPT/day.
>
> On Sat, Nov 5, 2016 at 9:57 PM, Bob Stewart <b...@evoria.net> wrote:
>
> Hi Scott,
>
> The 20 bits span about 6 volts.  The EFC range spans about 8Hz (+/-4Hz).
> I don't do these calculations every day, but that's about 4.5PPT?
>
> Bob
>
> -- -- -
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/ GFS-GPSDOs/info
> <http://groups.yahoo.com/neo/groups/GFS-GPSDOs/info?soc_src=mail_trk=ma>
>
>
>
>
> --
> *From:* Scott Stobbe <scott.j.sto...@gmail.com>
> *To:* Bob Stewart <b...@evoria.net>; Discussion of precise time and
> frequency measurement <time-nuts@febo.com>
> *Sent:* Saturday, November 5, 2016 8:38 PM
> *Subject:* Re: [time-nuts] Thermal impact on OCXO
>
> I think that's a nice plot, it looks like you have stepped 160 LSB over 7
> days or roughly 1 LSB per hour. With a 20bit dac you are trimming maybe 1
> ppt/LSB to 4 ppt/LSB? In allan devation terms, the case of 1ppt/LSB, solely
> due to drift, you're at 1E-12 at 3600*sqrt(2) = 5000 s, in the case of
> 4ppt/hour your at 1E-12 at 1280 s. Seems reasonable.
>
> On Sat, Nov 5, 2016 at 2:47 PM, Bob Stewart <b...@evoria.net> wrote:
>
> Oh dear.  I attached the wrong file.  Here's the correct one.
>  - -- --
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/ GFS-GPSDOs/info
> <http://groups.yahoo.com/neo/groups/GFS-GPSDOs/info?soc_src=mail_trk=ma>
>
>
>
>
> __ _
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> <https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts>
> and follow the instructions there.
>
>
>
>
>
>
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

If your DAC spans the full EFC range than 1LSB is 1/2^20 ~ 1 PPM of the EFC
range, and the EFC tuning range is 8/10E6 ~ 1 PPM full scale, so 1 LSB is
~1PPT. So, if everything else is stable the DAC code reflects
changes solely due to the OCXO, which would be an aging of 24 PPT/day.

On Sat, Nov 5, 2016 at 9:57 PM, Bob Stewart <b...@evoria.net> wrote:

> Hi Scott,
>
> The 20 bits span about 6 volts.  The EFC range spans about 8Hz (+/-4Hz).
> I don't do these calculations every day, but that's about 4.5PPT?
>
> Bob
>
> -
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>
>
>
>
> --
> *From:* Scott Stobbe <scott.j.sto...@gmail.com>
> *To:* Bob Stewart <b...@evoria.net>; Discussion of precise time and
> frequency measurement <time-nuts@febo.com>
> *Sent:* Saturday, November 5, 2016 8:38 PM
> *Subject:* Re: [time-nuts] Thermal impact on OCXO
>
> I think that's a nice plot, it looks like you have stepped 160 LSB over 7
> days or roughly 1 LSB per hour. With a 20bit dac you are trimming maybe 1
> ppt/LSB to 4 ppt/LSB? In allan devation terms, the case of 1ppt/LSB, solely
> due to drift, you're at 1E-12 at 3600*sqrt(2) = 5000 s, in the case of
> 4ppt/hour your at 1E-12 at 1280 s. Seems reasonable.
>
> On Sat, Nov 5, 2016 at 2:47 PM, Bob Stewart <b...@evoria.net> wrote:
>
> Oh dear.  I attached the wrong file.  Here's the correct one.
>  - -- --
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/ GFS-GPSDOs/info
> <http://groups.yahoo.com/neo/groups/GFS-GPSDOs/info?soc_src=mail_trk=ma>
>
>
>
>
> __ _
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> <https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts>
> and follow the instructions there.
>
>
>
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] I love the smell of tantalum in the morning

[Scott Stobbe]

For through hole parts sure, but I would not recommend that on SMD parts,
the copper foil of a little pad is pretty easy to tear off and it's a royal
pain if you have to mount a device missing some of its landing pads.

On Sat, Nov 5, 2016 at 4:37 PM, Tom Miller  wrote:

> I usually nibble away at the center of the part until it is two separate
> pieces. Then unsolder each piece. Clean the pads off with wick then install
> the new part.
>
> Use a good sharp pair of flush cut side cutters.
>
> Tom
>
>
> - Original Message - From: "Bob Camp" 
> To: "Tom Van Baak" ; "Discussion of precise time and
> frequency measurement" 
> Sent: Saturday, November 05, 2016 4:24 PM
> Subject: Re: [time-nuts] I love the smell of tantalum in the morning
>
>
> Hi
>>
>> A *lot* depends on how many planes there are in that board. The weight of
>> he copper
>> also maters a bit. If there is enough thermal mass, you will need a
>> pre-heat process.
>> There are lots of ways to do it ranging from the kitchen oven to various
>> “frame and
>> lightbulb” setups and on into ever more complex heating approaches.
>>
>> If the hot tweezers / soldering iron / hot air tool does not reflow the
>> solder quickly (10 seconds
>> or less) stop. Get a pre-heat setup and try again. With proper heat you
>> should have the part
>> off in under 4 seconds. People don’t tend to use stopwatches when
>> soldering. 4 seconds is quite
>> a while on a joint. Ten seconds is pretty much forever ….
>>
>> Bob
>>
>> On Nov 5, 2016, at 3:12 PM, Tom Van Baak  wrote:
>>>
>>> See C13 in the attached photo. I need to replace some blown caps on a
>>> few boards [1]. In one instance the cap got so hot it melted itself off the
>>> board. Quiet convenient, actually -- it acts like its own fuse -- but I
>>> don't think the 5071 designers had that clever feature in mind.
>>>
>>> Having not done SMT before, how should I do it with minimal risk to the
>>> very precious PCB. Or, what equipment should I use this as a good excuse to
>>> buy?
>>>
>>> Thanks,
>>> /tvb
>>>
>>> [0] http://www.imdb.com/title/tt0078788/quotes
>>> [1] http://leapsecond.com/museum/hp5071a/A1-mother.htm
>>> ___
>>> time-nuts mailing list -- time-nuts@febo.com
>>> To unsubscribe, go to https://www.febo.com/cgi-bin/m
>>> ailman/listinfo/time-nuts
>>> and follow the instructions there.
>>>
>>
>> ___
>> time-nuts mailing list -- time-nuts@febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/m
>> ailman/listinfo/time-nuts
>> and follow the instructions there.
>>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/m
> ailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

I think that's a nice plot, it looks like you have stepped 160 LSB over 7
days or roughly 1 LSB per hour. With a 20bit dac you are trimming maybe 1
ppt/LSB to 4 ppt/LSB? In allan devation terms, the case of 1ppt/LSB, solely
due to drift, you're at 1E-12 at 3600*sqrt(2) = 5000 s, in the case of
4ppt/hour your at 1E-12 at 1280 s. Seems reasonable.

On Sat, Nov 5, 2016 at 2:47 PM, Bob Stewart  wrote:

> Oh dear.  I attached the wrong file.  Here's the correct one.
>  -
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>
>
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thermal impact on OCXO

[Scott Stobbe]

Sounds like you already realized this. Phase is the integral of frequency
and the derivative of phase (phase rate) is frequency. So if you go from
nominal frequency - slow - nominal or equivalently nominal frequency - fast
- nominal the phase integrates up/down.

It would be a little more complicated for an ocxo since it is servoing the
xo temperature, you would need to know the disturbance rejection (gain,
time constant for a simple Pi controller) to try and feedfoward correct the
phase error.

On Friday, 4 November 2016, Bob Stewart  wrote:

> OK, never mind.  I see the obvious.  Phase changes faster at a higher
> frequency than it does at a lower frequency.
>
> Bob
>  -
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>
>   From: Bob Stewart >
>  To: Discussion of Precise Time and Frequency Measurement <
> time-nuts@febo.com >
>  Sent: Friday, November 4, 2016 8:56 PM
>  Subject: [time-nuts] Thermal impact on OCXO
>
> In the general case, is the impact of changing the ambient temperature
> around an OCXO from, say, 40C to 41C the same as changing it from 41C to
> 40C all else being equal?  IOW, if I somehow have the same temperature ramp
> over the same time period in both directions, will I wind up with the same
> frequency and phase, or will the frequency revert but at some phase
> difference?
>
> Bob - AE6RV
>  -
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
> ___
> time-nuts mailing list -- time-nuts@febo.com 
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com 
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Man with too many clocks.

[Scott Stobbe]

I'm not sure if there is a reason counters don't let you digitally
calibrate beyond that, the 10 MHz ref out on the rear panel would still be
out of cal.

On Fri, Nov 4, 2016 at 1:48 PM, Bob Camp  wrote:

> Hi
>
> The only practical way to set the 10811 or 10544 is with a >= 10 turn pot
> on the EFC. I
> never have worked out just why there are so many instruments that don’t
> have a pot on
> the EFC.
>
> Bob
>
> > On Nov 4, 2016, at 11:35 AM, Peter Reilley 
> wrote:
> >
> > I gave up on trying to use the GPS 1 PPS signal to calibrate the 10 MHz
> OCXO's that
> > I have.   The reason that others have pointed out is that the
> uncorrected 1 PPS
> > signal from the GPS is has just a little too much a jitter to use it for
> calibration
> > with your eye using a scope.   If it were sawtooth corrected then it
> would be better
> > but you really need a GPS disciplined oscillator.
> >
> > Not to be outdone, I brought out a rubidium oscillator that I had put
> away because
> > it did not appear to work properly.   It only put out a 1 PPS signal and
> nothing else.
> > I compared that with the GPS PPS and could get a good comparison on the
> scope.
> > The rubidium drifted about 40 nS over 12 hours.   So it seemed to be
> good.
> >
> > With that I could adjust the OCXO's in my 5370's.   The spec for the HP
> 5370B with
> > a HP 10811 OCXO is better than 1 X 10^-10 RMS for 1 sec average. That
> is, it should
> > take more than 1,000 seconds for one 10 MHz wave to shift by 360
> degrees.   That
> > is very hard to do using the screw adjustment in the OCXO.   Even the
> slightest
> > movement possible will cause a frequency change greater that is
> spec'ed.   How
> > do cal labs do it?
> >
> > My HP 5370A has a 10544 OCXO which is spec'ed for short term stability of
> > better than 1 X 10^11 for 1 second.   Even better than the 5370B! The
> adjustment
> > screw is much coarser and it is not possible to get any better than a
> few seconds for
> > one cycle phase shift of the 10 MHz OCXO against the standard.   It
> seems that I can't
> > get even close to the spec.
> >
> > These have been running for a few days.   It that enough?
> >
> > Pete.
> >
> >
> >
> > On 11/3/2016 8:20 AM, Peter Reilley wrote:
> >> I am the proverbial man with too many clocks and I don't know what time
> it is.
> >> To correct this situation I have decided to calibrate everything.
> >>
> >> I have a HP 5370B, a HP 6370A, and a HP 5328A all with the TCXO
> option.   I also
> >> have some TCXO modules.   I figured that I would calibrate them against
> my Trimble
> >> Resolution T GPS receiver.
> >>
> >> I put the 1 PPS signal in one channel of my scope and one of the 10 MHz
> TCXO
> >> signals in the other channel and look at the phase relationship. The
> TCXO's are
> >> already close enough that I should not be out by more than a fraction
> of a waveform.
> >> I understand that I have to deal with the 1 PPS without sawtooth
> correction.
> >>
> >> I expected to see the 10 MHz signal bounce around but not move more
> than 1/2
> >> of a wave length.   Instead I see the 10 MHz waveform appear steady for
> a few seconds
> >> then jump a significant portion of the wave.   The jump is too much to
> be confident
> >> that I have not slipped one cycle.
> >>
> >> Can I do what I am trying to do or am I missing something?
> >>
> >> Pete.
> >>
> >> ___
> >> time-nuts mailing list -- time-nuts@febo.com
> >> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> >> and follow the instructions there.
> >>
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Thinking outside the box a super reference

[Scott Stobbe]

You will also share the same challenges as Touchstone semi did, no one
wanted to stick their neck out to design in a little startup.

On Thu, Nov 3, 2016 at 7:49 PM, Bob Camp  wrote:

> Hi
>
> Not many people have had exposure to Rb’s or Cs standards actually being
> built. That leaves a major gap in who you can call when you run into a
> problem.
>
> Until you have tried to build one it’s not at all clear just how much
> “missing information” there
> is in all those papers. It’s very much like the semiconductor business.
> Lots of
> information is published. There are indeed lots of gaps. At some point you
> must
> build tooling and get it all working.
>
> Again, we are talking about a device that is at least as good as a 5065
> and not
> something that just barely works. If you *could* build something better
> than a 5065
> for a thousand or two dollars, it would be on the market today.
>
> Bob
>
> > On Nov 3, 2016, at 6:34 PM, Attila Kinali  wrote:
> >
> > On Thu, 3 Nov 2016 16:54:24 -0400
> > Bob Camp  wrote:
> >
> >> If you look at a modern CPU as “just a handful of sand and some stuff”,
> it seems
> >> pretty easy to build one in the kitchen after an hour or two of setup.
> When you dig
> >> into the nasty details the line costs rapidly spiral off into the
> stratosphere. Atomic
> >> standards are not quite as complex, but there still is more than just a
> little custom
> >> equipment involved. $1M sounds a bit on the low side of what it might
> take.
> >
> >
> > Not necessarily. There is a large corpus of knowledge available on
> > how to build vapor cells standards and what is a good idea and what
> > isn't. Most of it is documented in papers of the PTTI, EFTF and IFCS.
> > The former two are freely available (for PTTI until 2010, but that
> > should be good enough). Getting access to those papers behind a
> > paywall, you only need to know someone with access to a university.
> > (not for PTTI post 2010 though, ION has quite anal access rules)
> >
> > Additionally, the people in the time and frequeny community are very
> > open to discussion and exchange of knowledge. You can almost always
> > just walk up to someone and ask questions with a high chance of getting
> > not only answers but help in how to proceede.
> >
> > Tapping into this knowhow would avoid the need to try out the whole
> > solution space and concentrate on the few parts that are unkown or
> > not well enough understood and optimize those. And by doing so safe
> > a lot of money.
> >
> >   Attila Kinali
> >
> > --
> > Malek's Law:
> >Any simple idea will be worded in the most complicated way.
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Man with too many clocks.

[Scott Stobbe]

Bob has stated this, but perhaps not emphasized that, when you sample the
phase of a 10 MHz clock once a second, you are essentially folding the 20
millionth nyquist band down to baseband. So you can alias any integer
multiple of 1 Hz as if it were 10 MHz, i.e. 10 MHz + 1Hz will hold phase to
1 Hz (sampled once a second) just as well as 10 MHz would.

On Thu, Nov 3, 2016 at 1:00 PM, Bob Camp  wrote:

> Hi
>
> If your TCXO is off by 1 ppm, it will slip 10 cycles per second at 10 MHz.
> If it is off by 0.1 ppm it will slip a full cycle at 10 MHz.
> If it is off by 0.01 ppm *and* uses some sort of digital compensation, it
> will hop around.
> If the GPS is not sawtooth corrected it will hop by a good fraction of a
> cycle at 10 MHz
> If the GPS is not surveyed in and seeing many satellites, it may hop by
> more than a cycle at 10 MHz.
>
> Best bet:
>
> Divide the TCXO down to a much lower frequency (< 100 Hz). Use the 5370 to
> look at the
> delta between the GPS pulse and the TCXO output.
>
> Bob
>
>
> > On Nov 3, 2016, at 9:22 AM, Peter Reilley 
> wrote:
> >
> > I am using the 1 PPS for the trigger.
> >
> > Pete.
> >
> >
> > On 11/3/2016 8:59 AM, Antonio A. S. Magalhaes wrote:
> >>
> >> Pete,
> >>
> >> Tell us about your trigger: where is it?
> >>
> >> Regards,
> >>
> >> Antonio/CT1TE
> >>
> >> ---
> >>
> >> A 2016-11-03 12:20, Peter Reilley escreveu:
> >>
> >>> I am the proverbial man with too many clocks and I don't know what
> time it is.
> >>> To correct this situation I have decided to calibrate everything.
> >>>
> >>> I have a HP 5370B, a HP 6370A, and a HP 5328A all with the TCXO
> option.   I also
> >>> have some TCXO modules.   I figured that I would calibrate them
> against my Trimble
> >>> Resolution T GPS receiver.
> >>>
> >>> I put the 1 PPS signal in one channel of my scope and one of the 10
> MHz TCXO
> >>> signals in the other channel and look at the phase relationship. The
> TCXO's are
> >>> already close enough that I should not be out by more than a fraction
> of a waveform.
> >>> I understand that I have to deal with the 1 PPS without sawtooth
> correction.
> >>>
> >>> I expected to see the 10 MHz signal bounce around but not move more
> than 1/2
> >>> of a wave length.   Instead I see the 10 MHz waveform appear steady
> for a few seconds
> >>> then jump a significant portion of the wave.   The jump is too much to
> be confident
> >>> that I have not slipped one cycle.
> >>>
> >>> Can I do what I am trying to do or am I missing something?
> >>>
> >>> Pete.
> >>>
> >>> ___
> >>> time-nuts mailing list -- time-nuts@febo.com  time-nuts@febo.com>
> >>> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> >>> and follow the instructions there.
> >
> > ___
> > time-nuts mailing list -- time-nuts@febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] How to get PPS from ublox mini-PCI GPS to APU2 SoC serial port for ntpd

[Scott Stobbe]

The signals on the DB9 connector are at RS232 levels, your GPS module is
likely 3V3 CMOS, you will need to make a level translator if you wish to
use a standard PC RS232 serial port.

On Thu, Nov 3, 2016 at 12:58 PM, STR .  wrote:

> Hello again list and Paul,
>
> The USB and Mini-PCIe converters I ordered from Ebay China have not shown
> up
> yet.
> It appears PC Engines may take a while to release the custom BIOS that
> exposes UART3 and 4 on the APU2  :/
>
> In the interest of getting this working without waiting on a BIOS I will
> try
> connecting it to COM1, which is wired to the DB9 connector.
> https://pcengines.ch/pdf/apu2.pdf
>
> Before I fry a 3.3V board with 5V signals, I hoped to ask here and confirm
> my ideas.
>
>
> Do I cross RX and TX lines from the GPS pins to the RX/TX on the DB9/COM1
> connector?
> And straight connect the PPS line from the GPS to the DCD pin on the
> DB9/COM1 connector?
>
> GPS:
> http://pcengines.info/forums/?page=post=B4D54F22-8224-
> 4A67-9F9E-85F988A1F
> 888
>
>
>
> Thanks!
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Man with too many clocks.

[Scott Stobbe]

You can also use your counter to directly measure your GPS receiver's 1PPS,
which ends up being the error of your internal timebase. (plus the error in
your 1PPS)

On Thu, Nov 3, 2016 at 8:20 AM, Peter Reilley 
wrote:

> I am the proverbial man with too many clocks and I don't know what time it
> is.
> To correct this situation I have decided to calibrate everything.
>
> I have a HP 5370B, a HP 6370A, and a HP 5328A all with the TCXO option.
>  I also
> have some TCXO modules.   I figured that I would calibrate them against my
> Trimble
> Resolution T GPS receiver.
>
> I put the 1 PPS signal in one channel of my scope and one of the 10 MHz
> TCXO
> signals in the other channel and look at the phase relationship. The
> TCXO's are
> already close enough that I should not be out by more than a fraction of a
> waveform.
> I understand that I have to deal with the 1 PPS without sawtooth
> correction.
>
> I expected to see the 10 MHz signal bounce around but not move more than
> 1/2
> of a wave length.   Instead I see the 10 MHz waveform appear steady for a
> few seconds
> then jump a significant portion of the wave.   The jump is too much to be
> confident
> that I have not slipped one cycle.
>
> Can I do what I am trying to do or am I missing something?
>
> Pete.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/m
> ailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] our favorite topics

[Scott Stobbe]

So, the elephant *is* in the room, how majestic. Thanks for taking the time
to list some sample books/papers. A lot of general communications
systems/RFIC books certainly acknowledge phase noise and the results
because of its presence, but I don't know how many have an in depth
discussion of ways to tame the beast, so to speak. It sounds as if
improving phase noise has a large focus of many of your published works, I
will take a look :). Ultimately for board level design, phase noise tends
to be purchased versus designed.

Its difficult to appreciate just how much effort goes into a OCXO/TXCO
without having a go at it yourself. It's one thing to to build an
oscillator with 2n3904, I can only speculate just how much goes it into
getting to -120 dBc/Hz and below at a 1Hz offset on a 10 MHz oscillator.

If I recall correctly in Frerking text, there is somewhere between a
paragraph to a page or two, warning that existing small-signal circuit
theory (linearized about an operating point) doesn't actually apply here,
and with a bit of hand waving says have faith it still sort of works.

Does your text provide a good discussion of the implications of operating
with large signals?

On Sat, Oct 29, 2016 at 6:30 PM,  wrote:

> On what topic ? older book miss the phase noise issue, most useful things
> are in papers, look at the references
>
>
>- Ulrich L. Rohde, Ajay K. Poddar, Georg Böck "The Design of Modern
>Microwave Oscillators for Wireless Applications ", John Wiley & Sons, New
>York, NY, May, 2005, ISBN 0-471-72342-8
>.
>- George Vendelin, Anthony M. Pavio, Ulrich L. Rohde " Microwave
>Circuit Design Using Linear and Nonlinear Techniques ", John Wiley & Sons,
>New York, NY, May, 2005, ISBN 0-471-41479-4
>.
>- Ulrich L. Rohde, Anisha M. Apte http://ieeexplore.ieee.
>org/document/7506417/ Everything You Always Wanted to Know About
>Colpitts Oscillators [Applications Note]
>- Anisha M. Apte; Ajay K. Poddar; Ulrich L. Rohde; Enrico Rubiola,
>Colpitts oscillator: A new criterion of energy saving for high performance
>signal sources 2016 IEEE International Frequency Control Symposium (IFCS)
>
>
> Not because my name is there but we tried to list ALL relevant literature,
> also
>
> https://www.amazon.com/Design-Crystal-Other-Harmonic-
> Oscillators/dp/0471088196
>
> is useful and look at
>
> http://ieeexplore.ieee.org/Xplore/home.jsp  .
>
> In the last 10 years Dr. Poddar and I have done the most oscillator and
> phase noise work..Look us up at Xplore
>
> 73 de Ulrich
>
> In a message dated 10/29/2016 3:38:34 P.M. Eastern Daylight Time,
> scott.j.sto...@gmail.com writes:
>
> I found Frerking's "Crystal Oscillator Design and Temperature
> Compensation" to be a fruitful read. It's free on the archive,
> https://archive.org/details/CrystalOscillatorDesignTemperatureCompensation
> .
>
> Are there any recommendations for one or more book(s) that are definitely
> worth skimming through, or reading?
>
> On Sat, Oct 29, 2016 at 3:12 PM, KA2WEU--- via time-nuts <
> time-nuts@febo.com> wrote:
>
>> Some useful literature
>>
>> https://en.wikipedia.org/wiki/Phase_noise
>>
>>
>> https://en.wikipedia.org/wiki/Colpitts_oscillator
>>
>> Some links seem not to work   73 de Ulrich
>>
>> ___
>> time-nuts mailing list -- time-nuts@febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/m
>> ailman/listinfo/time-nuts
>> and follow the instructions there.
>>
>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] our favorite topics

[Scott Stobbe]

I found Frerking's "Crystal Oscillator Design and Temperature Compensation"
to be a fruitful read. It's free on the archive,
https://archive.org/details/CrystalOscillatorDesignTemperatureCompensation .

Are there any recommendations for one or more book(s) that are definitely
worth skimming through, or reading?

On Sat, Oct 29, 2016 at 3:12 PM, KA2WEU--- via time-nuts  wrote:

> Some useful literature
>
> https://en.wikipedia.org/wiki/Phase_noise
>
>
> https://en.wikipedia.org/wiki/Colpitts_oscillator
>
> Some links seem not to work   73 de Ulrich
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Opening an Isotemp OCXO

[Scott Stobbe]

I wounder if originally the designer was hoping to use all 8 wire or'd
inputs to lower the input referred noise during midscale transition. Then
backed out later for some reason.

On Fri, Oct 28, 2016 at 1:03 PM, Scott Stobbe <scott.j.sto...@gmail.com>
wrote:

> Could also be a quirk about the 74S30 that gives it better phase noise
> over a basic buffer.
>
>
> On Friday, 28 October 2016, jimlux <jim...@earthlink.net> wrote:
>
>> On 10/28/16 9:13 AM, Scott Stobbe wrote:
>>
>>> The OCXO82-59 datasheet lists 12V supply, 5V clock out, could also be a
>>> blown regulator in your ocxo, if it is indeed a 12v model.
>>>
>>> There you go..the design could use a 74S30 as a driver - it's fast,
>> fairly good drive, but runs off 5V.  If the regulator is shorted, and you
>> put 12V on it, it will cook.
>>
>> ___
>> time-nuts mailing list -- time-nuts@febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/m
>> ailman/listinfo/time-nuts
>> and follow the instructions there.
>>
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Opening an Isotemp OCXO

[Scott Stobbe]

Could also be a quirk about the 74S30 that gives it better phase noise over
a basic buffer.

On Friday, 28 October 2016, jimlux <jim...@earthlink.net> wrote:

> On 10/28/16 9:13 AM, Scott Stobbe wrote:
>
>> The OCXO82-59 datasheet lists 12V supply, 5V clock out, could also be a
>> blown regulator in your ocxo, if it is indeed a 12v model.
>>
>> There you go..the design could use a 74S30 as a driver - it's fast,
> fairly good drive, but runs off 5V.  If the regulator is shorted, and you
> put 12V on it, it will cook.
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/m
> ailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Opening an Isotemp OCXO

[Scott Stobbe]

The OCXO82-59 datasheet lists 12V supply, 5V clock out, could also be a
blown regulator in your ocxo, if it is indeed a 12v model.

On Friday, 28 October 2016, Poul-Henning Kamp  wrote:

> 
> In message <10a3ea7d-37f0-51bc-2470-35645d767...@comcast.net
> >, Peter Reilley writes:
>
> >The chip is run off 12 volts so it must be CMOS.
>
> Or the OCXO is not a 12V model ?
>
>
> --
> Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
> p...@freebsd.org | TCP/IP since RFC 956
> FreeBSD committer   | BSD since 4.3-tahoe
> Never attribute to malice what can adequately be explained by incompetence.
> ___
> time-nuts mailing list -- time-nuts@febo.com 
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.