[time-nuts] Truetime XLi sat signal levels (dBW to dBc/SNR conversion question)

[Mark Sims]

I have figured out an empirical formula for mapping between the TrueTime dBW 
values and the more traditionally reported signal level values.   

I ran a Ublox F9T and the TrueTime in parallel off the same antenna.  By 
comparing the TrueTime and Ublox sig levels a rather trivial mapping function 
popped out.   Basically you divide the dBW level by the (max dBW)(=145) -min 
dBW(=196) values) to get a value in the range of 0..1   Multiply that by 50 and 
add a 3dB fudge factor).   The two values match within 1 dB.

My XLi uses a Motorola M12 GPS receiver.  The XLi reports signal levels less 
that around 30 dBc as -196 dBW,
___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Re: [time-nuts] Truetime XLi sat signal levels (dBW to dBc/SNR conversion question)

[Azelio Boriani]

According to this reference

the minimum received power on Earth should be -158.50dBW.
According to this XLi manual

the expected antenna gain should be between 20 to 36dB,
so the reported received signal should have 16dB of uncertainty since
there is no command to set the actual antenna gain.

On Sat, Jun 15, 2019 at 1:01 AM David I. Emery  wrote:
>
> On Fri, Jun 14, 2019 at 01:05:21AM +, Mark Sims wrote:
>
> Since nobody else has stuck their neck out to speculate... I
> will.
>
> I do not pretend to know for sure exactly what TrueTime means.
>
> > Pretty much all GPS receivers report sig levels in dBc or SNR type
> > values in the range of around 20 .. 50.   The Truetime XLi reports them
> > in dBW with a range of around -196 to -145 dBW).
>
> That of course is -166 dBm to -115 dBm
>
> But you know that.
>
> >Does anybody know how to convert the dBW values into something more
> > in line with standard dBc or SNR values?  The conversion does not need
> > to be exact, it just needs to be reasonable looking...
>
> As I understand the dBc values,  they are the signal power after
> despreading of the 50 symbol per second underlying 50 baud BPSK almanac
> in the appropriate approximately 50 Hz bandwidth it occupies compared
> to the thermal (or also spread interference) energy in that same
> bandwidth. Thus 20 dBc would be despread 50 baud BPSK energy 20 db over
> the noise in the relevant approximately 50 Hz BW (depending on filter
> rolloff).
>
> This is probably actually calibrated as Eb/No of the despread 50
> baud almanac signal.
>
> This of course is a ratio of signal to noise and interference
> and not directly related to actual signal power coming out of the
> antenna which depends on antenna gain, and feed losses and cable losses
> and so forth.
>
> There is also absolute power per square meter (flux density),
> but any measurement of that would depend on antenna, LNA, filter, cable,
> GPS receiver front end  gains and losses and noise temp which would not
> in general be the same from installation to installation and a measure
> of this would require calibration of those things for that installation.
>
> One possible interpretation of what Truetime is using is the
> noise density per root Hz compared to that from a 50 ohm termination at
> room temperature.   Obviously a 50 ohm system terminated in a 50 ohm
> load has -174 dBm/Hz at 290K and the lower end of their scale at -166
> dBm is 8 db above this, which I understand is about the theoretical
> threshold (BER of 10**-5) for the 50 baud BPSK in Eb/No.   What this
> implies is that despread, the input signal power density/Hz  inside the
> despread signal bandwidth would be -166 dBm/Hz (8 db Eb/No with No being
> -174 dBm).
>
> Now in actual reality the effective noise temp of antenna and
> filter, and sky and environment and LNA is probably not 290K exactly...
> I'm not familiar enough with this as respects GPS to be sure what would
> be considered normal for a typical or even theoretical deployed antenna.
>
> I would however expect that the True time dBW numbers would track
> Eb/No measures on other receivers starting at some base offset...
> and it seems likely perhaps that the 8 db Eb/No point for other receivers
> is -166 dBm or -196 dBW on their scale.
>
> This would be related to actual power density/Hz across the signal
> at L1 (or whatever)   by the process gain.. 50/1023000 or 43.1 db.
>
> In other words at the point where the BER of the 50 baud almanac
> gets up to around 1 error in 10**5 bits the effective signal power/Hz from
> the spread GPS signal on a zero gain hemispheric antenna would be (-166
> - 43.1) dBm/Hz or -209.1 dBm/Hz
>
> But what do I know.   Perhaps someone who does will answer.
>
>
> --
>   Dave Emery N1PRE/AE, d...@dieconsulting.com  DIE Consulting, Weston, Mass 
> 02493
> "An empty zombie mind with a forlorn barely readable weatherbeaten
> 'For Rent' sign still vainly flapping outside on the weed encrusted pole - in
> celebration of what could have been, but wasn't and is not to be now either."
>
>
> ___
> time-nuts mailing list -- time-nuts@lists.febo.com
> To unsubscribe, go to 
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.

___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Re: [time-nuts] Truetime XLi sat signal levels (dBW to dBc/SNR conversion question)

[David I. Emery]

On Fri, Jun 14, 2019 at 01:05:21AM +, Mark Sims wrote:
 
Since nobody else has stuck their neck out to speculate... I
will.

I do not pretend to know for sure exactly what TrueTime means.

> Pretty much all GPS receivers report sig levels in dBc or SNR type
> values in the range of around 20 .. 50.   The Truetime XLi reports them
> in dBW with a range of around -196 to -145 dBW). 

That of course is -166 dBm to -115 dBm

But you know that.

>Does anybody know how to convert the dBW values into something more
> in line with standard dBc or SNR values?  The conversion does not need
> to be exact, it just needs to be reasonable looking...

As I understand the dBc values,  they are the signal power after
despreading of the 50 symbol per second underlying 50 baud BPSK almanac
in the appropriate approximately 50 Hz bandwidth it occupies compared
to the thermal (or also spread interference) energy in that same
bandwidth. Thus 20 dBc would be despread 50 baud BPSK energy 20 db over
the noise in the relevant approximately 50 Hz BW (depending on filter
rolloff).

This is probably actually calibrated as Eb/No of the despread 50
baud almanac signal.

This of course is a ratio of signal to noise and interference
and not directly related to actual signal power coming out of the
antenna which depends on antenna gain, and feed losses and cable losses
and so forth.

There is also absolute power per square meter (flux density),
but any measurement of that would depend on antenna, LNA, filter, cable,
GPS receiver front end  gains and losses and noise temp which would not
in general be the same from installation to installation and a measure
of this would require calibration of those things for that installation.

One possible interpretation of what Truetime is using is the
noise density per root Hz compared to that from a 50 ohm termination at
room temperature.   Obviously a 50 ohm system terminated in a 50 ohm
load has -174 dBm/Hz at 290K and the lower end of their scale at -166
dBm is 8 db above this, which I understand is about the theoretical
threshold (BER of 10**-5) for the 50 baud BPSK in Eb/No.   What this
implies is that despread, the input signal power density/Hz  inside the
despread signal bandwidth would be -166 dBm/Hz (8 db Eb/No with No being
-174 dBm).

Now in actual reality the effective noise temp of antenna and
filter, and sky and environment and LNA is probably not 290K exactly... 
I'm not familiar enough with this as respects GPS to be sure what would
be considered normal for a typical or even theoretical deployed antenna.

I would however expect that the True time dBW numbers would track
Eb/No measures on other receivers starting at some base offset...
and it seems likely perhaps that the 8 db Eb/No point for other receivers
is -166 dBm or -196 dBW on their scale.

This would be related to actual power density/Hz across the signal
at L1 (or whatever)   by the process gain.. 50/1023000 or 43.1 db.

In other words at the point where the BER of the 50 baud almanac
gets up to around 1 error in 10**5 bits the effective signal power/Hz from
the spread GPS signal on a zero gain hemispheric antenna would be (-166
- 43.1) dBm/Hz or -209.1 dBm/Hz 

But what do I know.   Perhaps someone who does will answer.


-- 
  Dave Emery N1PRE/AE, d...@dieconsulting.com  DIE Consulting, Weston, Mass 
02493
"An empty zombie mind with a forlorn barely readable weatherbeaten
'For Rent' sign still vainly flapping outside on the weed encrusted pole - in 
celebration of what could have been, but wasn't and is not to be now either."


___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

[time-nuts] TICC reference source

[Mark Sims]

If you are using the TICC to measure the time interval between two signals then 
the reference clock does not need to be all that good (the gooder the better, 
though).  

But if you are using the TICC to measure one or two independent sources, then 
your reference clock should be as good as you can supply.
___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Re: [time-nuts] TICC reference source

[Taka Kamiya via time-nuts]

Thank you!  This is a great manual for time interval measurement.  I'll go with 
10811.

--- 
(Mr.) Taka Kamiya
I'm stuck in a wormhole  Hello, worms! 

On Friday, June 14, 2019, 12:28:26 AM EDT, Anders Wallin 
 wrote:  
 
 page 48 here has some notes on time-base 
error:http://leapsecond.com/hpan/an200-3.pdf

for time-interval measurement the number of digits matter. if you keep 
time-intervals 'small', say 123.45 ns (probably can't resolve much below 10ps 
anyway with a TICC) then a time-base with only about 5-6 digits of 
stability is sufficient - even if you are measuring two good H-masers 
against each other.The OCXO may have better short-term stability compared to 
the Rb.
Anders
On Fri, Jun 14, 2019 at 5:01 AM Taka Kamiya via time-nuts 
 wrote:

I am building a TICC-IN-A-BOX.  Basically TICC + TADD-2 mini + reference 
oscillator.
Initially, I was going to use HP11811-6011 but I am also considering PRS10.  (I 
already own both)  While I believe 11811 is sufficient, PRS10 can be kept off 
and reaches usable state (physics lock) faster, while 11811 has to be kept on 
or otherwise usable stability will be few hours.

I do have a house clock but since it can be a subject of measurement, I want my 
TICC-IN-A-BOX to have an independent clock.
Am I thinking correctly?
--- 
(Mr.) Taka Kamiya
I'm stuck in a wormhole  Hello, worms!
___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

  
___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Re: [time-nuts] TICC reference source

[Anders Wallin]

page 48 here has some notes on time-base error:
http://leapsecond.com/hpan/an200-3.pdf

for time-interval measurement the number of digits matter. if you keep
time-intervals 'small', say 123.45 ns (probably can't resolve much below
10ps anyway with a TICC) then a time-base with only about 5-6 digits of
stability is sufficient - even if you are measuring two good
H-masers against each other.
The OCXO may have better short-term stability compared to the Rb.

Anders

On Fri, Jun 14, 2019 at 5:01 AM Taka Kamiya via time-nuts <
time-nuts@lists.febo.com> wrote:

> I am building a TICC-IN-A-BOX.  Basically TICC + TADD-2 mini + reference
> oscillator.
> Initially, I was going to use HP11811-6011 but I am also considering
> PRS10.  (I already own both)  While I believe 11811 is sufficient, PRS10
> can be kept off and reaches usable state (physics lock) faster, while 11811
> has to be kept on or otherwise usable stability will be few hours.
>
> I do have a house clock but since it can be a subject of measurement, I
> want my TICC-IN-A-BOX to have an independent clock.
> Am I thinking correctly?
> ---
> (Mr.) Taka Kamiya
> I'm stuck in a wormhole  Hello, worms!
> ___
> time-nuts mailing list -- time-nuts@lists.febo.com
> To unsubscribe, go to
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to 
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.