Re: [time-nuts] HP-59309A Clock counts only seconds

[Tom Miller]

4000 series logic is just fine. Just replace the original parts with like.

http://www.ebay.com/itm/CD4011CN-CMOS-Quad-2-Input-NAND-Gate-5-pcs-/272579598216?hash=item3f77029788:m:mSmV7o8qfGTPl04DaSu3DSw

Regards


- Original Message - 
From: "Jeremy Nichols" 

To: 
Sent: Monday, April 03, 2017 7:43 PM
Subject: [time-nuts] HP-59309A Clock counts only seconds


I have a new-to-me HP-59309A HP-IB Digital Clock. The clock works on both 
the internal crystal oscillator and on an external 10 MHz standard (GPSDO). 
However, it counts only to 60 seconds and then repeats without updating the 
minutes digit. The TIME SET (FAST and SLOW) push-buttons work but again, 
the count will not update minutes and hours, only seconds. The DAY SET 
procedure works correctly for days and months. All the other switches and 
buttons do what they are supposed to do. Using my 10526T logic pulser I can 
force the minutes and hours counters to work. The power supply is in good 
condition (after replacement of a few components) and I see no other 
problems (yet).


Tracing the clock signal through the logic circuitry brought me to U3 on 
the A4 board. This (U3) is a 4011 quad 2-input NAND gate in a 14-pin DIP 
package. It "connects" the seconds counter to the minutes counter and 
appears to have failed. One of the people on the email list 
 commented that the 4000 series CMOS 
chips have a known limited lifetime. Since these parts are no longer in 
production, the writer expressed the concern that any parts I might find 
to buy may be DOA.


Before I go hunting for parts, I'd appreciate hearing from anyone in the 
group who has experience with the 59309A Clock and/or the 4000-series CMOS 
family. In particular, are there modern equivalents to my 4011 chip? If 
the 4000's really do have a limited lifetime I'd rather use a substitute.


Jeremy, N6WFO


___
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-59309A Clock counts only seconds

[Scott McGrath]

A 4011 is an inexpensive part,   Mouser still carries them,   I would NOT buy a 
chinese part.  Generally ESD mishandling is the cause of short component life

Content by Scott
Typos by Siri

> On Apr 3, 2017, at 7:43 PM, Jeremy Nichols  wrote:
> 
> I have a new-to-me HP-59309A HP-IB Digital Clock. The clock works on both the 
> internal crystal oscillator and on an external 10 MHz standard (GPSDO). 
> However, it counts only to 60 seconds and then repeats without updating the 
> minutes digit. The TIME SET (FAST and SLOW) push-buttons work but again, the 
> count will not update minutes and hours, only seconds. The DAY SET procedure 
> works correctly for days and months. All the other switches and buttons do 
> what they are supposed to do. Using my 10526T logic pulser I can force the 
> minutes and hours counters to work. The power supply is in good condition 
> (after replacement of a few components) and I see no other problems (yet).
> 
> Tracing the clock signal through the logic circuitry brought me to U3 on the 
> A4 board. This (U3) is a 4011 quad 2-input NAND gate in a 14-pin DIP package. 
> It "connects" the seconds counter to the minutes counter and appears to have 
> failed. One of the people on the email list 
>  commented that the 4000 series CMOS 
> chips have a known limited lifetime. Since these parts are no longer in 
> production, the writer expressed the concern that any parts I might find to 
> buy may be DOA.
> 
> Before I go hunting for parts, I'd appreciate hearing from anyone in the 
> group who has experience with the 59309A Clock and/or the 4000-series CMOS 
> family. In particular, are there modern equivalents to my 4011 chip? If the 
> 4000's really do have a limited lifetime I'd rather use a substitute.
> 
> Jeremy, N6WFO
> 
> 
> ___
> 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] HP-59309A Clock counts only seconds

[Mark Sims]

You can get 10 CD4011's shipped f rom China for $1 off of Ebay...
___
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-59309A Clock counts only seconds, HEF4011

[Alex Pummer]

NOt pin-compatible but functional compatible and available is the 74C00 
or74 HC00, which has a long life I am using one in my gate alarm 
since1989, it is on all the time and the chip is available the HEF4011 
is also available -- at Mouser -- it is pin-compatible and function  
with CD4011 ad it is made by Philips [NXP]


73

KJ6UHN

Alex


On 4/3/2017 4:43 PM, Jeremy Nichols wrote:
I have a new-to-me HP-59309A HP-IB Digital Clock. The clock works on 
both the internal crystal oscillator and on an external 10 MHz 
standard (GPSDO). However, it counts only to 60 seconds and then 
repeats without updating the minutes digit. The TIME SET (FAST and 
SLOW) push-buttons work but again, the count will not update minutes 
and hours, only seconds. The DAY SET procedure works correctly for 
days and months. All the other switches and buttons do what they are 
supposed to do. Using my 10526T logic pulser I can force the minutes 
and hours counters to work. The power supply is in good condition 
(after replacement of a few components) and I see no other problems 
(yet).


Tracing the clock signal through the logic circuitry brought me to U3 
on the A4 board. This (U3) is a 4011 quad 2-input NAND gate in a 
14-pin DIP package. It "connects" the seconds counter to the minutes 
counter and appears to have failed. One of the people on the email 
list  commented that the 4000 
series CMOS chips have a known limited lifetime. Since these parts are 
no longer in production, the writer expressed the concern that any 
parts I might find to buy may be DOA.


Before I go hunting for parts, I'd appreciate hearing from anyone in 
the group who has experience with the 59309A Clock and/or the 
4000-series CMOS family. In particular, are there modern equivalents 
to my 4011 chip? If the 4000's really do have a limited lifetime I'd 
rather use a substitute.


Jeremy, N6WFO


___
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.


-
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2016.0.8012 / Virus Database: 4769/14231 - Release Date: 
04/03/17


___
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-59309A Clock counts only seconds

[Tim Shoppa]

CMOS availability has held up pretty well because of its wide voltage
range. Mouser has over eleven thousand CD4011BE's in stock:
http://www.mouser.com/ProductDetail/Texas-Instruments/CD4011BE/

I can't rule out an old logic gate dying, but I would suspect this first:
Some clocks were made or modified to be tied to external minute and/or hour
synchronization pulses and these inputs can hold and jam zeroes into the
minute and hour counters.

Tim N3QE

On Mon, Apr 3, 2017 at 7:43 PM, Jeremy Nichols  wrote:

> I have a new-to-me HP-59309A HP-IB Digital Clock. The clock works on both
> the internal crystal oscillator and on an external 10 MHz standard (GPSDO).
> However, it counts only to 60 seconds and then repeats without updating the
> minutes digit. The TIME SET (FAST and SLOW) push-buttons work but again,
> the count will not update minutes and hours, only seconds. The DAY SET
> procedure works correctly for days and months. All the other switches and
> buttons do what they are supposed to do. Using my 10526T logic pulser I can
> force the minutes and hours counters to work. The power supply is in good
> condition (after replacement of a few components) and I see no other
> problems (yet).
>
> Tracing the clock signal through the logic circuitry brought me to U3 on
> the A4 board. This (U3) is a 4011 quad 2-input NAND gate in a 14-pin DIP
> package. It "connects" the seconds counter to the minutes counter and
> appears to have failed. One of the people on the email list <
> hp_agilent_equipm...@yahoogroups.com> commented that the 4000 series CMOS
> chips have a known limited lifetime. Since these parts are no longer in
> production, the writer expressed the concern that any parts I might find to
> buy may be DOA.
>
> Before I go hunting for parts, I'd appreciate hearing from anyone in the
> group who has experience with the 59309A Clock and/or the 4000-series CMOS
> family. In particular, are there modern equivalents to my 4011 chip? If the
> 4000's really do have a limited lifetime I'd rather use a substitute.
>
> Jeremy, N6WFO
>
>
> ___
> 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] HP-59309A Clock counts only seconds

[Bob kb8tq]

Hi

A lot of the early IC’s were packaged in the era of “learning by doing”. The
net result is that some of them do indeed have process related packaging 
“use by dates”. On top of that the early IC processes were not as fully 
worked out as the more modern ones. Things like metallization thickness 
turned out to be on the skimpy side in some cases. The other nail in the 
coffin is input and output protection. The hazards of ESD and device upset
were not handled as rigorously early on.  The net result is that depending on 
*any* IC that dates back ~30 years is a bit of a gamble. 

Bob



> On Apr 3, 2017, at 7:43 PM, Jeremy Nichols  wrote:
> 
> I have a new-to-me HP-59309A HP-IB Digital Clock. The clock works on both the 
> internal crystal oscillator and on an external 10 MHz standard (GPSDO). 
> However, it counts only to 60 seconds and then repeats without updating the 
> minutes digit. The TIME SET (FAST and SLOW) push-buttons work but again, the 
> count will not update minutes and hours, only seconds. The DAY SET procedure 
> works correctly for days and months. All the other switches and buttons do 
> what they are supposed to do. Using my 10526T logic pulser I can force the 
> minutes and hours counters to work. The power supply is in good condition 
> (after replacement of a few components) and I see no other problems (yet).
> 
> Tracing the clock signal through the logic circuitry brought me to U3 on the 
> A4 board. This (U3) is a 4011 quad 2-input NAND gate in a 14-pin DIP package. 
> It "connects" the seconds counter to the minutes counter and appears to have 
> failed. One of the people on the email list 
>  commented that the 4000 series CMOS 
> chips have a known limited lifetime. Since these parts are no longer in 
> production, the writer expressed the concern that any parts I might find to 
> buy may be DOA.
> 
> Before I go hunting for parts, I'd appreciate hearing from anyone in the 
> group who has experience with the 59309A Clock and/or the 4000-series CMOS 
> family. In particular, are there modern equivalents to my 4011 chip? If the 
> 4000's really do have a limited lifetime I'd rather use a substitute.
> 
> Jeremy, N6WFO
> 
> 
> ___
> 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-59309A Clock counts only seconds

[Jim Harman]

The 14SO package is still in production, HEF4011B $0.37 ea from Digikey.
You should be able to get a breakout adapter if you need the DIP.

On Mon, Apr 3, 2017 at 7:43 PM, Jeremy Nichols  wrote:

> I have a new-to-me HP-59309A HP-IB Digital Clock. The clock works on both
> the internal crystal oscillator and on an external 10 MHz standard (GPSDO).
> However, it counts only to 60 seconds and then repeats without updating the
> minutes digit. The TIME SET (FAST and SLOW) push-buttons work but again,
> the count will not update minutes and hours, only seconds. The DAY SET
> procedure works correctly for days and months. All the other switches and
> buttons do what they are supposed to do. Using my 10526T logic pulser I can
> force the minutes and hours counters to work. The power supply is in good
> condition (after replacement of a few components) and I see no other
> problems (yet).
>
> Tracing the clock signal through the logic circuitry brought me to U3 on
> the A4 board. This (U3) is a 4011 quad 2-input NAND gate in a 14-pin DIP
> package. It "connects" the seconds counter to the minutes counter and
> appears to have failed. One of the people on the email list <
> hp_agilent_equipm...@yahoogroups.com> commented that the 4000 series CMOS
> chips have a known limited lifetime. Since these parts are no longer in
> production, the writer expressed the concern that any parts I might find to
> buy may be DOA.
>
> Before I go hunting for parts, I'd appreciate hearing from anyone in the
> group who has experience with the 59309A Clock and/or the 4000-series CMOS
> family. In particular, are there modern equivalents to my 4011 chip? If the
> 4000's really do have a limited lifetime I'd rather use a substitute.
>
> Jeremy, N6WFO
>
>
> ___
> 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.
>



-- 

--Jim Harman
___
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-59309A Clock counts only seconds

[David C. Partridge]

You can get a brand new CD4011BE from pretty much most suppliers at about 15 
cents that won't exhibit the problems of the earlier ones.

Dave

-Original Message-
From: time-nuts [mailto:time-nuts-boun...@febo.com] On Behalf Of Jeremy Nichols
Sent: 04 April 2017 00:44
To: time-nuts@febo.com
Subject: [time-nuts] HP-59309A Clock counts only seconds

I have a new-to-me HP-59309A HP-IB Digital Clock. The clock works on both the 
internal crystal oscillator and on an external 10 MHz standard (GPSDO). 
However, it counts only to 60 seconds and then repeats without updating the 
minutes digit. The TIME SET (FAST and SLOW) push-buttons work but again, the 
count will not update minutes and hours, only seconds. The DAY SET procedure 
works correctly for days and months. All the other switches and buttons do what 
they are supposed to do. Using my 10526T logic pulser I can force the minutes 
and hours counters to work. 
The power supply is in good condition (after replacement of a few
components) and I see no other problems (yet).

Tracing the clock signal through the logic circuitry brought me to U3 on the A4 
board. This (U3) is a 4011 quad 2-input NAND gate in a 14-pin DIP package. It 
"connects" the seconds counter to the minutes counter and appears to have 
failed. One of the people on the email list 
 commented that the 4000 series CMOS 
chips have a known limited lifetime. Since these parts are no longer in 
production, the writer expressed the concern that any parts I might find to buy 
may be DOA.

Before I go hunting for parts, I'd appreciate hearing from anyone in the group 
who has experience with the 59309A Clock and/or the 4000-series CMOS family. In 
particular, are there modern equivalents to my 4011 chip? If the 4000's really 
do have a limited lifetime I'd rather use a substitute.

Jeremy, N6WFO


___
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] HP5061B Peak to Valley Ratios

[Donald E. Pauly]

https://www.febo.com/pipermail/time-nuts/2017-April/104600.html

We hooked up the oscilloscope directly to the electron multiplier in
the beam tube with a 1 Meg termination.  We were operating open loop
and varying the fine crystal oscillator adjust for the different
photos. Remember that higher beam current is downward in the photo.
Currents listed in nA translate to beam current numbers on the HP5061B
meter.

First photo at http://gonascent.com/papers/hp/hp5061/waveform/centered.jpg
is the frequency swept at a 137 cps rate around the peak of the cesium
resonance.  Ground is at the top of the screen and 5 mV (also 5 nA)
per division.  Horizonal is 4 ms per division.This corresponds to a
peak beam current of -22 nA and a valley current of -17 nA.  One
valley current is below resonance and the other is above resonance.
The waveform is 274 cps.  Sequence is low to high and then high to
low.  No 137 cps is present.

The fine frequency control was adjusted to produce a slight offset
from center for the photo at
http://gonascent.com/papers/hp/hp5061/waveform/smalloff.jpg .   Note
that the peak beam current is still -22 nA but two different valley
currents exist.  One is 17 nA and the other is -18 nA.  Most of the
signal is 274 cps but some asymmetry is present at 137 cps.  One
extreme of the sweep is slightly closer to center frequency than the
other.

Finally the frequency was adjusted to produce a severe offset with all
of the sweep below resonance in the photo of
http://gonascent.com/papers/hp/hp5061/waveform/largeoff.jpg .  The
peak is now slightly below resonance at -21 nA and the valley is far
below resonance at 5 nA.  The signal is all 137 cps with no 274 cps.
See figures 4-43 and 4-44 on page 4-31 of the manual for further
explanation.

We will be finding out the effect on beam current of tests shortly on
the Zeeman coil and LF coil.  We think that the 55.53 kc Zeeman
frequency introduces some AM into the beam frequency but no FM.  What
the LF coil has is unclear at this time.

πθ°μΩω±√·Γλ
WB0KVV
___
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] HP-59309A Clock counts only seconds

[Jeremy Nichols]

I have a new-to-me HP-59309A HP-IB Digital Clock. The clock works on 
both the internal crystal oscillator and on an external 10 MHz standard 
(GPSDO). However, it counts only to 60 seconds and then repeats without 
updating the minutes digit. The TIME SET (FAST and SLOW) push-buttons 
work but again, the count will not update minutes and hours, only 
seconds. The DAY SET procedure works correctly for days and months. All 
the other switches and buttons do what they are supposed to do. Using my 
10526T logic pulser I can force the minutes and hours counters to work. 
The power supply is in good condition (after replacement of a few 
components) and I see no other problems (yet).


Tracing the clock signal through the logic circuitry brought me to U3 on 
the A4 board. This (U3) is a 4011 quad 2-input NAND gate in a 14-pin DIP 
package. It "connects" the seconds counter to the minutes counter and 
appears to have failed. One of the people on the email list 
 commented that the 4000 series 
CMOS chips have a known limited lifetime. Since these parts are no 
longer in production, the writer expressed the concern that any parts I 
might find to buy may be DOA.


Before I go hunting for parts, I'd appreciate hearing from anyone in the 
group who has experience with the 59309A Clock and/or the 4000-series 
CMOS family. In particular, are there modern equivalents to my 4011 
chip? If the 4000's really do have a limited lifetime I'd rather use a 
substitute.


Jeremy, N6WFO


___
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] Conversion of Stanford PRS10 to RS-232 Communication

[Mike Cook]

The word jumper may be misleading. Soldering/de-solderig of components required.
In Appendix 19, "Configuration notes » there is an explanation of how to alter 
the function of pins 4,5,7.



> Le 3 avr. 2017 à 22:05, Philip Jackson  a écrit :
> 
> I understand that the PRS10 Rubidium units either talk to the host
> device via a RS-232 serial command language or via analog values using
> the same set of pins in both cases.
> 
> The manual says that internal jumpers can be used to switch between
> communication modes.
> 
> I have a PRS10 that is set for analog working and would like to switch
> it over to RS-232 mode.
> 
> A cursory peek inside the case with no disassembly doesn't reveal any
> simple jumpers/headers that might be used to flip formats.
> 
> Does anyone know how the format can be changed?
> 
> The manual doesn't say anything more than there are jumpers inside that
> can be changed.
> 
> Many thanks
> 
> Philip
> ___
> 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.

"The power of accurate observation is commonly called cynicism by those who 
have not got it. »
George Bernard Shaw

___
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] Conversion of Stanford PRS10 to RS-232 Communication

[Dan Rae]

On 4/3/2017 1:05 PM, Philip Jackson wrote:

I understand that the PRS10 Rubidium units either talk to the host
device via a RS-232 serial command language or via analog values using
the same set of pins in both cases.




Does anyone know how the format can be changed?

The manual doesn't say anything more than there are jumpers inside that
can be changed.

Page 20 of the manual has the details...

Dan
___
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] Conversion of Stanford PRS10 to RS-232 Communication

[Philip Jackson]

I understand that the PRS10 Rubidium units either talk to the host
device via a RS-232 serial command language or via analog values using
the same set of pins in both cases.

The manual says that internal jumpers can be used to switch between
communication modes.

I have a PRS10 that is set for analog working and would like to switch
it over to RS-232 mode.

A cursory peek inside the case with no disassembly doesn't reveal any
simple jumpers/headers that might be used to flip formats.

Does anyone know how the format can be changed?

The manual doesn't say anything more than there are jumpers inside that
can be changed.

Many thanks

Philip
___
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] Oscillator Maker is hiring

[David Tang]

Dear Time Enthusiasts,

SiTime, makers of MEMS-based oscillators, is hiring for multiple groups.  My 
group in particular is hiring Systems Characterization Engineers.  We get to 
characterize and integrate the guts (e.g. MEMS, regulators, pll, oscillators 
circuits, etc) to make products that push the envelope of size, power, and 
performance.  In my group, you will get to know how our clocks work and see 
everything (down to the physics) underneath the hood.  You will get paid to 
make great timing devices.

Other groups, such as circuit design, Marketing, and Operations, are also 
hiring.  Please contact me if you are interested - even in positions in other 
groups.  dt...@sitime.com

Best regards,
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.

Re: [time-nuts] TAPR TICC boxed

[Mark Sims]

I could use the 1PPS output of a crappy GPS receiver.  Should have at least 30 
ns of jitter. Or for finer, but still noisy, the PPS from the Venus timing 
receiver is around 6 ns.   I'd need at least 100 feet of PTFE coax to get it 
out the door and back inside with enough cable outside to make a difference.   
Alas, I have none.

Or I could do it all indoors and set the air conditioning to around 20C and let 
that cycle the temperature.  Or put the cable in the freezer and monitor it 
while it warms up...  hmmm... that sounds fun and easy to do.

I've been pretty amazed by what a TICC can do for $200...   I wonder how a GPX 
chip based one would perform?
I used the TICC to dial in my HP-5065 freq to around 1E-12.  I had not adjusted 
it for a couple of years and it was around 1-E11 off.

From: Bruce Griffiths 
Sent: Monday, April 3, 2017 3:05 AM
To: Mark Sims; Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] TAPR TICC boxed

For even more fun you could try to detect the PTFE phase change  at around 20C 
using a cable with PTFE dielectric.

A pulse source with somewhat more pulse to pulse jitter may be more useful in 
that averaging will occur over a wider range of fine interpolator codes.

Bruce

On 03 April 2017 at 05:34 Mark Sims  wrote:

I implemented the channel offset compensation feature specifically to make 
measuring cable delays more accurate. I wanted to measure my TDR calibration 
cable and another very precision delay line. I used Heather to null out the 
channel/connector delays and then replaced one of the "T" cables with the TDR 
cable.

My test setup / TICC was coming up with a -306 ps channel offset error. The 
test signal was the 1PPS output of a FTS4060 cesium. Connecting / reconnecting 
one of the test setup cables and re-doing the offset test (I was averaging for 
1800 seconds) could produce compensation values that varied from -300 ps to 
-325 ps. Just de-doing the offset test without messing with the cables produced 
values around -300 to -310 ps.

BNC connectors aren't the best for precision timing. I need to re-run the test 
with SMA cables / T adapter and the precision HP connector torque wrench and 
see what that looks like. It would also be fun to lay a coax outdoors and see 
how the delay changes over a day as it heats/cools.



Some “cables” have very long delay numbers.

___
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

[Bruce Griffiths]

Copper jacketed low density PTFE insulated coax cables typically exhibit a net 
phase change of over 1000ppm during the PTFE phase transition. See figure 2 p14 
of the Cables and connectors supplement to March 2017 microwave journal. Solid 
PTFE insulated cables exhibit an even greater (2 -3x) phase change. to achieve 
a phase shift tempco of 10ppm/C either hermetically selaed silicon dioxide 
powder insulated coax or "phase tracking" semirigid or better phase tracking 
flexible coax is required. 

Bruce

> 
> On 03 April 2017 at 21:22 Attila Kinali  wrote:
> 
> On Mon, 3 Apr 2017 15:05:55 +1200 (NZST)
> Bruce Griffiths  wrote:
> 
> > > 
> > For even more fun you could try to detect the PTFE phase change at 
> > around 20C using a cable with PTFE dielectric.
> > 
> > > 
> This will require several 100 meters of cable to be measureable with
> the TICC. Modern cables are all <500ppm/K, good cables <10ppm/K, phase
> stable cables reach even <1ppm/K. Measuring a change of 10ppm with
> a resolution of 60ps means that the delay has to be in the order of 6µs,
> which is close to 1000m of cable. Even if dithering gives another facor
> of 10, this still means 100m of cable.
> 
> For this level of comparison I would suggest to use a sinusoidal signal,
> instead of a pulse, and do phase comparison, which gives a resolution
> in the order of 1ps with very little effort, thus reducing the required
> cable length to 10-20 meters.
> 
> Attila Kinali
> 
> [1] "Temperature Stability of Coaxial Cables", Czuba and Sikora, 2011
> http://przyrbwn.icm.edu.pl/APP/PDF/119/a119z4p17.pdf
> 
> --
> 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] TAPR TICC boxed

[Bob Camp]

Hi

You could simply immerse all the cables in a swimming pool full of mercury … :)
(bonus points for a link to the prior discussion of that topic ..).

Bob

> On Apr 2, 2017, at 11:05 PM, Bruce Griffiths  
> wrote:
> 
> For even more fun you could try to detect the PTFE phase change  at around 
> 20C using a cable with PTFE dielectric.
> 
> A pulse source with somewhat more pulse to pulse jitter may be more useful in 
> that averaging will occur over a wider range of fine interpolator codes.
> 
> Bruce
> 
>> 
>>On 03 April 2017 at 05:34 Mark Sims  wrote:
>> 
>>I implemented the channel offset compensation feature specifically to 
>> make measuring cable delays more accurate. I wanted to measure my TDR 
>> calibration cable and another very precision delay line. I used Heather to 
>> null out the channel/connector delays and then replaced one of the "T" 
>> cables with the TDR cable.
>> 
>>My test setup / TICC was coming up with a -306 ps channel offset error. 
>> The test signal was the 1PPS output of a FTS4060 cesium. Connecting / 
>> reconnecting one of the test setup cables and re-doing the offset test (I 
>> was averaging for 1800 seconds) could produce compensation values that 
>> varied from -300 ps to -325 ps. Just de-doing the offset test without 
>> messing with the cables produced values around -300 to -310 ps.
>> 
>>BNC connectors aren't the best for precision timing. I need to re-run the 
>> test with SMA cables / T adapter and the precision HP connector torque 
>> wrench and see what that looks like. It would also be fun to lay a coax 
>> outdoors and see how the delay changes over a day as it heats/cools.
>> 
>>
>> 
 
>>>Some “cables” have very long delay numbers.
>>> 
>>>___
>>>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] TAPR TICC boxed

[Attila Kinali]

On Mon, 3 Apr 2017 15:05:55 +1200 (NZST)
Bruce Griffiths  wrote:

> For even more fun you could try to detect the PTFE phase change  at around 
> 20C using a cable with PTFE dielectric.

This will require several 100 meters of cable to be measureable with
the TICC. Modern cables are all <500ppm/K, good cables <10ppm/K, phase
stable cables reach even <1ppm/K. Measuring a change of 10ppm with
a resolution of 60ps means that the delay has to be in the order of 6µs,
which is close to 1000m of cable. Even if dithering gives another facor
of 10, this still means 100m of cable. 

For this level of comparison I would suggest to use a sinusoidal signal,
instead of a pulse, and do phase comparison, which gives a resolution
in the order of 1ps with very little effort, thus reducing the required
cable length to 10-20 meters.


Attila Kinali

[1] "Temperature Stability of Coaxial Cables", Czuba and Sikora, 2011
http://przyrbwn.icm.edu.pl/APP/PDF/119/a119z4p17.pdf

-- 
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.