After ham-fistedly breaking a few things, I've finally got my GPSDO working. (does happy dance)  Now that it's had some time to warm up I see that it's started wandering around +/- a couple of counts from sample to sample (16 seconds).  Is this because the UT+ is not set to timing mode?  Putting in an RS-232 (or USB) to TTL adapter, switchable between the UT+ and the VE2ZAZ board, is pretty much next on my schedule.  I've got an adapter I made many years ago, but it's not suitable for mounting in the box, and there's no simple way to hook it up to the UT+ at the moment. Bob

No, it doesn't depend on the timing mode. What reference are you using? That is, your counter is using its internal reference or else. On Fri, Jul 12, 2013 at 5:59 AM, Bob Stewart <bob@evoria.net> wrote: > After ham-fistedly breaking a few things, I've finally got my GPSDO working. (does happy dance) Now that it's had some time to warm up I see that it's started wandering around +/- a couple of counts from sample to sample (16 seconds). Is this because the UT+ is not set to timing mode? Putting in an RS-232 (or USB) to TTL adapter, switchable between the UT+ and the VE2ZAZ board, is pretty much next on my schedule. I've got an adapter I made many years ago, but it's not suitable for mounting in the box, and there's no simple way to hook it up to the UT+ at the moment. > > 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.

Hi On Jul 11, 2013, at 11:59 PM, Bob Stewart <bob@evoria.net> wrote: > After ham-fistedly breaking a few things, I've finally got my GPSDO working. (does happy dance) Now that it's had some time to warm up Time as in minutes, or as in a couple of days? If you look at the plot in the QEX article, a day or so is the expected "warmup" period. > I see that it's started wandering around +/- a couple of counts from sample to sample (16 seconds). If it's + *and* - then the gizmo is driving down the middle of the road. If it's always + or always - then it's likely not yet locked. > Is this because the UT+ is not set to timing mode? The UT needs to be in "position hold" mode with a proper (24 to 48 hour) survey on it's location for best performance. It should not be +/- 100 ns even without this. Is your VCXO a 10811? Does it have the same sensitivity as a 10811? If not, you may need (per the article) to make some firmware changes. Bob > Putting in an RS-232 (or USB) to TTL adapter, switchable between the UT+ and the VE2ZAZ board, is pretty much next on my schedule. I've got an adapter I made many years ago, but it's not suitable for mounting in the box, and there's no simple way to hook it up to the UT+ at the moment. > > 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.

Hi Bob, OK, you got me: time as in about 120 minutes.  =)  It's mostly "+", with the occasional "-" for about an hour or so.  Then, it wavers back and forth between "+" and "-", but the long term trend is still very slowly "+".    So, I guess the OCXO needs more time to "settle in"?  My OCXO is a Trimble 34310-T. As soon as I can get the serial port wiring done I'll be able to tell if it's in "position hold" mode.  I suspect it's not, as it doesn't have a battery on-board.  I did wind up mounting a Lithium 3V backup battery, using a holder from an old motherboard. I'll have to look around to see if anyone has any software to graph the DAC track for this board. Bob - AE6RV >________________________________ > From: Bob Camp <lists@rtty.us> >To: Discussion of precise time and frequency measurement <time-nuts@febo.com> >Sent: Friday, July 12, 2013 8:47 AM >Subject: Re: [time-nuts] GPDSO is working > > >Hi > >On Jul 11, 2013, at 11:59 PM, Bob Stewart <bob@evoria.net> wrote: > >> After ham-fistedly breaking a few things, I've finally got my GPSDO working. (does happy dance)  Now that it's had some time to warm up > >Time as in minutes, or as in a couple of days? If you look at the plot in the QEX article, a day or so is the expected "warmup" period. > >> I see that it's started wandering around +/- a couple of counts from sample to sample (16 seconds). > >If it's + *and* - then the gizmo is driving down the middle of the road. If it's always + or always - then it's likely not yet locked. > >> Is this because the UT+ is not set to timing mode? > >The UT needs to be in "position hold" mode with a proper (24 to 48 hour) survey on it's location for best performance. It should not be +/- 100 ns even without this. > >Is your VCXO a 10811? Does it have the same sensitivity as a 10811? If not, you may need (per the article) to make some firmware changes. > >Bob > >> Putting in an RS-232 (or USB) to TTL adapter, switchable between the UT+ and the VE2ZAZ board, is pretty much next on my schedule.  I've got an adapter I made many years ago, but it's not suitable for mounting in the box, and there's no simple way to hook it up to the UT+ at the moment. >> >> 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. > > >

Hi Azelio I'm not tracking it with a counter.   The VE2ZAZ board I mentioned in my post has a serial port interface.  Every 32 seconds you get a line of tracking information, which includes the right 4 digits of the hex count of how many oscillator transitions occurred during the previous 16 second sampling period.  With an oscillator of 10 MHz there will be 160,000,000 transitions during the 10 second sampling period for a dead accurate oscillator.  It also gives an error count for each period so you don't have to do the math to subtract it.  This error count value is the one I'm referring to. Bob - AE6RV >________________________________ > From: Azelio Boriani <azelio.boriani@screen.it> >To: Discussion of precise time and frequency measurement <time-nuts@febo.com> >Sent: Friday, July 12, 2013 3:50 AM >Subject: Re: [time-nuts] GPDSO is working > > >No, it doesn't depend on the timing mode. What reference are you >using? That is, your counter is using its internal reference or else. > >On Fri, Jul 12, 2013 at 5:59 AM, Bob Stewart <bob@evoria.net> wrote: >> After ham-fistedly breaking a few things, I've finally got my GPSDO working. (does happy dance)  Now that it's had some time to warm up I see that it's started wandering around +/- a couple of counts from sample to sample (16 seconds).  Is this because the UT+ is not set to timing mode?  Putting in an RS-232 (or USB) to TTL adapter, switchable between the UT+ and the VE2ZAZ board, is pretty much next on my schedule.  I've got an adapter I made many years ago, but it's not suitable for mounting in the box, and there's no simple way to hook it up to the UT+ at the moment. >> >> 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. > > >

Hi You will need to fine tune the firmware settings for that OCXO. Bob On Jul 12, 2013, at 11:27 AM, Bob Stewart <bob@evoria.net> wrote: > Hi Bob, > > OK, you got me: time as in about 120 minutes. =) It's mostly "+", with the occasional "-" for about an hour or so. Then, it wavers back and forth between "+" and "-", but the long term trend is still very slowly "+". So, I guess the OCXO needs more time to "settle in"? My OCXO is a Trimble 34310-T. > > As soon as I can get the serial port wiring done I'll be able to tell if it's in "position hold" mode. I suspect it's not, as it doesn't have a battery on-board. I did wind up mounting a Lithium 3V backup battery, using a holder from an old motherboard. > > I'll have to look around to see if anyone has any software to graph the DAC track for this board. > > Bob - AE6RV > > > > > >> ________________________________ >> From: Bob Camp <lists@rtty.us> >> To: Discussion of precise time and frequency measurement <time-nuts@febo.com> >> Sent: Friday, July 12, 2013 8:47 AM >> Subject: Re: [time-nuts] GPDSO is working >> >> >> Hi >> >> On Jul 11, 2013, at 11:59 PM, Bob Stewart <bob@evoria.net> wrote: >> >>> After ham-fistedly breaking a few things, I've finally got my GPSDO working. (does happy dance) Now that it's had some time to warm up >> >> Time as in minutes, or as in a couple of days? If you look at the plot in the QEX article, a day or so is the expected "warmup" period. >> >>> I see that it's started wandering around +/- a couple of counts from sample to sample (16 seconds). >> >> If it's + *and* - then the gizmo is driving down the middle of the road. If it's always + or always - then it's likely not yet locked. >> >>> Is this because the UT+ is not set to timing mode? >> >> The UT needs to be in "position hold" mode with a proper (24 to 48 hour) survey on it's location for best performance. It should not be +/- 100 ns even without this. >> >> Is your VCXO a 10811? Does it have the same sensitivity as a 10811? If not, you may need (per the article) to make some firmware changes. >> >> Bob >> >>> Putting in an RS-232 (or USB) to TTL adapter, switchable between the UT+ and the VE2ZAZ board, is pretty much next on my schedule. I've got an adapter I made many years ago, but it's not suitable for mounting in the box, and there's no simple way to hook it up to the UT+ at the moment. >>> >>> 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.

The periodic +/- might be caused by periodic error in the PPS. The UT+ has an error "sawtooth" function. That is the error is the PPS is left to accumulate and then it is reset. So if you DAC counts goes up and down on a few seconds or few tens of seconds period it could be tracking the error in the PPS. The UT+ keeps its PPS within spec. Once you gt the serial line connected you will be able to read the sawtooth corrections. The UT+ out puts the estimated error in the PPS and I guess a very sophisticated GPSDO would use that data to correct the phase measurement. From memory, I think with a UT+ the error in the PPS goes up to about 50 ns. (I have a couple UT+) -- Chris Albertson Redondo Beach, California