¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io
  1. QRPLabs
  2. Topics

ProgRock/ProgRock2 vs GPSDO (trimble)

 

Hi Hans,

any plan to compare ProgRock/ProgRock2 used as GPSDO vs Trimble and/or others GPSDO for providing 10MHz reference signal? Still I am not sure it can stay within well less than 1Hz if set for 10MHz. Having just one GPSDO, even assuming is reliable, I could not simply test it.

Could set up this test for us?

I could volunteer lending you mine, you would need at least three...

Giuseppe Marullo
IW2JWW - JN45RQ
 

Hello Giuseppe

The best reference I have is a 10MHz OCXO, I don't have any Rubidium or Caesium GPS-disciplined standards. So I'll do the best with the tools I have and I am sure once the thing is shipped, somebody will have access to a better lab than mine and can make good comparisons.

73 Hans G0UPL



On Fri, Jan 27, 2023 at 12:07 PM Giuseppe Marullo[iw2jww] <giuseppe@...> wrote:
Hi Hans,

any plan to compare ProgRock/ProgRock2 used as GPSDO vs Trimble and/or
others GPSDO for providing 10MHz reference signal? Still I am not sure
it can stay within well less than 1Hz if set for 10MHz. Having just one
GPSDO, even assuming is reliable, I could not simply test it.

Could set up this test for us?

I could volunteer lending you mine, you would need at least three...

Giuseppe Marullo
IW2JWW - JN45RQ
 

Over the last few weeks I have been 'watching' a gps conditioned progrock on
42Mhz with a statistical frequency counter whose timebase is locked to GPS via
a second GPS receiver. The progrock deviation from 42000000.0Hz is averaged at
0.7Hz with occasional far larger deviations caused, I believe, by power supply
and progrock gps's 1pps issues. Not awesome performance, but adequate for most
'sensible' people but not for 'Time-Nuts'.
hth 73 Mike W

On 27 Jan 2023 at 13:07, Giuseppe Marullo[iw2jww] wrote:

Hi Hans,

any plan to compare ProgRock/ProgRock2 used as GPSDO vs Trimble and/or
others GPSDO for providing 10MHz reference signal? Still I am not sure
it can stay within well less than 1Hz if set for 10MHz. Having just one
GPSDO, even assuming is reliable, I could not simply test it.

Could set up this test for us?

I could volunteer lending you mine, you would need at least three...

Giuseppe Marullo
IW2JWW - JN45RQ
 

n 29/01/2023 10:31, Mike wrote:
Not awesome performance, but adequate for most
'sensible' people but not for 'Time-Nuts'.
Mike

Quite a few people who seem sensible are talking of using the QDX with FST4W.
Taken to the extreme "FST4W-1800 has a FSK tone spacing of 0.089 Hz and does not track slow frequency drift, "

I gave up thinking of using a Progrock as a standard long ago, a modern OCXO does better although not cheap nowadays. It will be interesting to see how the new version works.

I guess generally I am a Time Nut but there are some radio uses that need Time Nut accuracy.

73 Alan G4ZFQ
 

A group of us are exploring the use of FST4W-120/300/900/1800 on the 80-1M bands.? Because?the WSJT-x jt9 decoder which searches for FST4W does not include the drift compensation used by the wsprd?WSPR packet decoder, the frequency stability requirements for even FST4W-120 are significantly greater than for WSPR-2 packets.? That means that the FST4W transmitter stability and receiver stability cannot be together more than .5 Hz on any band.? That is fairly easy to do on 2200M and 630M, but progressively harder on the higher bands.? While the TXCO in the QDX is stable enough for most of the longer modes, ?the heat generated by transmitting introduces about 2 Hz of drift 20M which is enough that even FTS4W-120 packets will not be frequently ?on 30M and above.

The KiwiSDR includes a GPS section which uses a GPS "referencing" algorithm (not GPSDO) which achieves better than .05 Hz stability?over the duration of a WSPR-2 packet, which is 10-100 times worse than a GPSDO but good enough for FST4W-1800 spotting on 10M. ? There are more than 30 Kiwi-equipped Wsprdaemon sites which are continuously recording FST4W transmissions from over 30 transmit sites.

However the Kiwi's GPS "referencing" algorithm is not easily implemented on other devices, so for most FST4W transmit and receive sites the best option is to use a true GPSDO oscillator.? The Bodnar mini is an excellent choice, but it is far more accurate and alone triples the cost of a QDX FST4W beacon station.? So it would be great if Hans were to offer such a GPSDO at a QRP labs price point, and of course an easy way to connect it to the QDX ;=)
??


On Sun, Jan 29, 2023 at 3:09 AM Alan G4ZFQ <alan4alan@...> wrote:
n 29/01/2023 10:31, Mike wrote:
> Not awesome performance, but adequate for most
> 'sensible' people but not for? 'Time-Nuts'.

Mike

Quite a few people who seem sensible are talking of using the QDX with
FST4W.
Taken to the extreme "FST4W-1800 has a FSK tone spacing of 0.089 Hz and
does not track slow frequency drift, "

I gave up thinking of using a Progrock as a standard long ago, a modern
OCXO does better although not cheap nowadays. It will be interesting to
see how the new version works.

I guess generally I am a Time Nut but there are some radio uses that
need Time Nut accuracy.

73 Alan G4ZFQ










--
Rob Robinett
AI6VN
mobile: +1 650 218 8896
 

On Fri, Jan 27, 2023 at 12:07 PM, Giuseppe Marullo[iw2jww] wrote:
Hi Hans,

any plan to compare ProgRock/ProgRock2 used as GPSDO vs Trimble and/or
others GPSDO for providing 10MHz reference signal? Still I am not sure
it can stay within well less than 1Hz if set for 10MHz. Having just one
GPSDO, even assuming is reliable, I could not simply test it.

I think the there are limitations of how well the ProgRock1 is disciplined by the GPS.? As far as I can tell, it alters the Reference Frequency value to achieve "lock" with the 1PPS signal.? Since this is limited to changes at a minimum of 1Hz (at the reference frequency of 27 or 25MHz) this means a similar jump at your required frequency, e.g. about a 0.5Hz jump at 10Mhz.? I believe that the software attempts to maintain an average frequency, by appropriate reference frequency changes of +/- 1Hz.
I expect that Progrock2 will implement a similar arrangement.? Obviously, it would be better if small changes could be made to the fractional divider values which would allow smaller steps of frequency adjustment.
For Progrock1, I have achieved better short term stability by fitting a 25MHz (QRP Labs) TCXO, and also thermally insulating the board by wrapping the synth board with plastic foam to reduce the rate of temperature change of the synth chip and TCXO.? The result is a satisfactory performance in my "bottom of the garden" remote receiver setup, where temperatures vary from near freezing to up in the 30 degrees C over the year.? I have ordered a Progrock2 board, as, even if it uses the same control arrangement as Progrock1, the more compact unit will make it much easier to thermally insulate it to slow down the rate of frequency adjustment needed.
--
Peter Lee
G3SPL
 

Hi Peter

ProgRock and Progrock2 don't use the same technique for GPS discipline. Progrock2 should have far higher accuracy than ProgRock. We shall see...

73 Hans G0UPL
http://qrp-labs.com
https://www.buymeacoffee.com/g0upl


-------- Original message --------
From: G3SPL <leepd60@...>
Date: Wed, Feb 1, 2023, 11:58 PM
To: [email protected]
Subject: Re: [QRPLabs] ProgRock/ProgRock2 vs GPSDO (trimble)
On Fri, Jan 27, 2023 at 12:07 PM, Giuseppe Marullo[iw2jww] wrote:
Hi Hans,

any plan to compare ProgRock/ProgRock2 used as GPSDO vs Trimble and/or
others GPSDO for providing 10MHz reference signal? Still I am not sure
it can stay within well less than 1Hz if set for 10MHz. Having just one
GPSDO, even assuming is reliable, I could not simply test it.

I think the there are limitations of how well the ProgRock1 is disciplined by the GPS.? As far as I can tell, it alters the Reference Frequency value to achieve "lock" with the 1PPS signal.? Since this is limited to changes at a minimum of 1Hz (at the reference frequency of 27 or 25MHz) this means a similar jump at your required frequency, e.g. about a 0.5Hz jump at 10Mhz.? I believe that the software attempts to maintain an average frequency, by appropriate reference frequency changes of +/- 1Hz.
I expect that Progrock2 will implement a similar arrangement.? Obviously, it would be better if small changes could be made to the fractional divider values which would allow smaller steps of frequency adjustment.
For Progrock1, I have achieved better short term stability by fitting a 25MHz (QRP Labs) TCXO, and also thermally insulating the board by wrapping the synth board with plastic foam to reduce the rate of temperature change of the synth chip and TCXO.? The result is a satisfactory performance in my "bottom of the garden" remote receiver setup, where temperatures vary from near freezing to up in the 30 degrees C over the year.? I have ordered a Progrock2 board, as, even if it uses the same control arrangement as Progrock1, the more compact unit will make it much easier to thermally insulate it to slow down the rate of frequency adjustment needed.
--
Peter Lee
G3SPL

1 - 7 of 7