Author Topic: ($40 GPSDO working well now, my mistake)  (Read 60504 times)

0 Members and 1 Guest are viewing this topic.

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
($40 GPSDO working well now, my mistake)
« on: August 12, 2017, 02:15:11 pm »
Edit: Old title was: GPSDO loss of satellites+fix troubleshooting
New title: True Position GPSDO working well now

Original post: What would you look for as broken/fixable in a GPSDO that:

1.) fails to lock on more than 1-3 satellites at a time, and that with extreme difficulty.
2.) wanders in terms of temperature and seems very hot (upper 40s C.)

The antenna and signals are known to be good, so its not that. It currently has around a 50% sky view.


Note: I should have known better.
The GPSDO was never broken, it was my dumb mistake, its a semi old GPSDO, and it needed to have a big and expansive sky view to be happy, which it wasnt getting in my windowsill. Moved to the middle of the back yard, it got seven sats.

its happy and working well now (and these GPSDOs, which were used in cellphone E911 radiolocation, are a great value.)
« Last Edit: August 13, 2017, 02:57:18 am by cdev »
"What the large print giveth, the small print taketh away."
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1380
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #1 on: August 12, 2017, 02:33:27 pm »
What the c/n numbers (signal strength or SNR potentially) for the 3 satellites you do lock? A good antenna/lna/rx should pull C/N numbers in the high 30s to low 40s for satellites in good positions, and in the 30s for the rest of a reasonable constellation, with a few satellites (maybe on the horizon) in the high 20s.

If you're only getting lock on 3 or so and they are in the 20s, something is potentially wrong with your antenna/lna/cable/rx front end.
« Last Edit: August 12, 2017, 02:40:26 pm by ConKbot »
 

Online edpalmer42

  • Super Contributor
  • ***
  • Posts: 2260
  • Country: ca
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #2 on: August 12, 2017, 04:43:52 pm »
Cdev, you know better than this!  We can only give vague general suggestions when you give us vague general info.  What type of GPSDO is it?  Can you talk to it?  Any error messages?

Ed
 
 
The following users thanked this post: cdev

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #3 on: August 12, 2017, 05:35:28 pm »
Its a True Position cellphone radiolocation board. Telecom surplus, apparently.

It looked like too good of a deal to pass up.

And now its working better, much better.

So, (crossing fingers) maybe that was it!

Wow, now seven satellites.. things are looking good.. 


Thank you!
« Last Edit: August 13, 2017, 03:12:24 am by cdev »
"What the large print giveth, the small print taketh away."
 

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #4 on: August 12, 2017, 06:09:45 pm »
Its a telecom surplus unit that was made in 2008. It contains a Bliley OCXO and Furuno GT-8031 timing GPS.

It cost me around $40! 

What do you get?

 The bare board. It requires 12 volts of power and you need to talk to it via 3.3 volt UART or RS-232. When it boots, you have to enter in $PROCEED and then it starts receiving.

There are a number of commands that are possible which are detailed in a PDF at packratvhf.com  There was a thread in May mostly about them on the Time-Nuts mailing list.

Now the fun part starts, building a device around it.

There is an "$EXTSTATUS" message which is not documented in the PDF which contains the number of satellites tracked.




Quote from: edpalmer42 on Today at 10:43:52
Cdev, you know better than this!  We can only give vague general suggestions when you give us vague general info.  What type of GPSDO is it?  Can you talk to it?  Any error messages?

Ed
"What the large print giveth, the small print taketh away."
 

Offline Nuno_pt

  • Frequent Contributor
  • **
  • Posts: 435
  • Country: pt
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #5 on: August 12, 2017, 06:16:19 pm »
I'm considering to buy one of this also, to check my 2 Trimble units  :-//  |O, and to round all also a NEC one just to check the others  :palm:
Nuno
CT2IRY
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1380
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #6 on: August 12, 2017, 07:21:48 pm »
I've seen receivers that are off for a while take 15-20 minutes to lock up while almanac data is updated, and some time receivers take well past 20 minutes to be happy and enable output. Not too unusual for a unit powered off for an unknown but probably long period of time. I wouldn't be too worried. :-+
 

Online edpalmer42

  • Super Contributor
  • ***
  • Posts: 2260
  • Country: ca
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #7 on: August 12, 2017, 08:39:34 pm »
Many (most?) GPS receivers remember where they last were.  If you move the receiver, it starts looking for where the satellites are supposed to be and they're not there.  They also don't know what time of day it is.  This can result in a *very* long time to find satellites and download the current almanac and ephemeris.  If the receiver allows it, you can drastically speed up this process by manually entering the approximate time/date/location.  It doesn't have to be perfect, just close.

Ed
 
The following users thanked this post: cdev

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1380
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #8 on: August 13, 2017, 02:43:16 am »
Many (most?) GPS receivers remember where they last were.  If you move the receiver, it starts looking for where the satellites are supposed to be and they're not there.  They also don't know what time of day it is.  This can result in a *very* long time to find satellites and download the current almanac and ephemeris.  If the receiver allows it, you can drastically speed up this process by manually entering the approximate time/date/location.  It doesn't have to be perfect, just close.

Ed
Tangentially related, but still related. Some time receivers have a fixed position mode. If you have a surveyed in accurate antenna location, or let the receiver self survey (averages location for 24+ hours) and record the location. With location fixed, you no longer need 4 satellites for a fix (1 for time, 3 for triangulation) so you have an "extra" 3 satellites adding to precision of the time fix.
 

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #9 on: August 13, 2017, 03:18:45 am »
After letting it have its antenna outdoors to do its survey now its chugging away indoors - I don't know how accurate it is in this mode yet. I think my GPS timing antenna is getting mounted outdoors fairly soon.

Quote from: ConKbot on Today at 20:43:16>Quote from: edpalmer42 on Today at 14:39:34
Many (most?) GPS receivers remember where they last were.  If you move the receiver, it starts looking for where the satellites are supposed to be and they're not there.  They also don't know what time of day it is.  This can result in a *very* long time to find satellites and download the current almanac and ephemeris.  If the receiver allows it, you can drastically speed up this process by manually entering the approximate time/date/location.  It doesn't have to be perfect, just close.

Ed
Tangentially related, but still related. Some time receivers have a fixed position mode. If you have a surveyed in accurate antenna location, or let the receiver self survey (averages location for 24+ hours) and record the location. With location fixed, you no longer need 4 satellites for a fix (1 for time, 3 for triangulation) so you have an "extra" 3 satellites adding to precision of the time fix.
"What the large print giveth, the small print taketh away."
 

Offline metrologist

  • Super Contributor
  • ***
  • Posts: 2199
  • Country: 00
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #10 on: August 14, 2017, 07:30:36 pm »
Many (most?) GPS receivers remember where they last were.  If you move the receiver, it starts looking for where the satellites are supposed to be and they're not there.  They also don't know what time of day it is.  This can result in a *very* long time to find satellites and download the current almanac and ephemeris.  If the receiver allows it, you can drastically speed up this process by manually entering the approximate time/date/location.  It doesn't have to be perfect, just close.

Ed

It always intrigued me on how these receivers can look for SV's where it thinks they should be. It's easier to understand not know the time and then not knowing which SV's to expect to hear.
 

Online edpalmer42

  • Super Contributor
  • ***
  • Posts: 2260
  • Country: ca
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #11 on: August 14, 2017, 08:37:32 pm »
Many (most?) GPS receivers remember where they last were.  If you move the receiver, it starts looking for where the satellites are supposed to be and they're not there.  They also don't know what time of day it is.  This can result in a *very* long time to find satellites and download the current almanac and ephemeris.  If the receiver allows it, you can drastically speed up this process by manually entering the approximate time/date/location.  It doesn't have to be perfect, just close.

Ed

It always intrigued me on how these receivers can look for SV's where it thinks they should be. It's easier to understand not know the time and then not knowing which SV's to expect to hear.

It always amazes me that GPS works at all!  A receiver starts off knowing nothing.  Somehow, it listens for multiple signals on the same frequency, finds one, gets the date and time, then downloads the almanac and ephemeris that tells it what else to look for and then calculates its location based on triangulating those signals.  It makes my head hurt to think about it!

Arthur C. Clarke said that "Any sufficiently advanced technology is indistinguishable from magic."  For me, GPS is proof of that statement!

Ed
 

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: ($40 GPSDO working well now, my mistake)
« Reply #12 on: August 14, 2017, 09:07:35 pm »
GPS is an incredible example of a great many different scientific disciplines being used together to solve an age old problem.

Anybody interested in how GPS works should check out RTKLib.

Which you can learn about at http://rtklib.com it's author's home page, and his github.

Here is a tutorial..
"What the large print giveth, the small print taketh away."
 

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #13 on: August 14, 2017, 09:15:26 pm »
A little update: You totally nailed it in this post, ConKbot.

Now that it was able to download its survey info, and has been running for awhile, its now running okay with even an indoor antenna, and tracking between 2 or 3 and 7 sats. In other words, its managing using an indoor patch antenna. Freeing up my other antenna for other stuff.

It seems to me that at the beginning, having its location reset, thinking it was back in Japan, really impacted its ability to function.


Quote from: ConKbot on 2017-08-12, 20:43:16>Quote from: edpalmer42 on 2017-08-12, 14:39:34
Many (most?) GPS receivers remember where they last were.  If you move the receiver, it starts looking for where the satellites are supposed to be and they're not there.  They also don't know what time of day it is.  This can result in a *very* long time to find satellites and download the current almanac and ephemeris.  If the receiver allows it, you can drastically speed up this process by manually entering the approximate time/date/location.  It doesn't have to be perfect, just close.

Ed
Tangentially related, but still related. Some time receivers have a fixed position mode. If you have a surveyed in accurate antenna location, or let the receiver self survey (averages location for 24+ hours) and record the location. With location fixed, you no longer need 4 satellites for a fix (1 for time, 3 for triangulation) so you have an "extra" 3 satellites adding to precision of the time fix.
"What the large print giveth, the small print taketh away."
 

Offline CJay

  • Super Contributor
  • ***
  • Posts: 4136
  • Country: gb
Re: ($40 GPSDO working well now, my mistake)
« Reply #14 on: August 15, 2017, 05:12:54 am »
I picked one of these receivers up too and have just received a cheap patch antenna with the correct connector:

http://www.ebay.co.uk/itm/3M-28dB-LNA1575-42MHz-RP-SMB-GPS-Signal-Active-Antenna-Aerial-Connector-Cable-FE/252329705481?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

Is this similar to the patch you're using cdev?

 

Offline CopperCone

  • Super Contributor
  • ***
  • Posts: 1415
  • Country: us
  • *knock knock*
Re: ($40 GPSDO working well now, my mistake)
« Reply #15 on: August 15, 2017, 06:17:55 am »
don't forget the RF forum. I get paranoid thinking about how many responses I don't get posting my threads there!!!
 

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: ($40 GPSDO working well now, my mistake)
« Reply #16 on: August 15, 2017, 02:03:36 pm »
Cjay,

That's good to know about- thank you!

I am using a home-made adapter from SMB to SMA. Both of the powered antennas Ive used with the True Position (one is a Lucent/Max-Rad/PCtel, that runs off almost everything you throw at it, the other is a Motorola SMA patch antenna that runs off 3.3 or 5 volts) have 'worked' i.e. registered as being there.

Isn't the True Position a great deal?

Have you put it into a case yet? I'm looking for an inexpensive case.

I think this project deserves something a bit nicer than my usual cases. (I often use recycled cases from other devices or even cardboard boxes..  that's what it's in now)
« Last Edit: August 15, 2017, 02:05:37 pm by cdev »
"What the large print giveth, the small print taketh away."
 

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: GPSDO loss of satellites+fix troubleshooting
« Reply #17 on: August 15, 2017, 02:10:22 pm »
That would depend on downloading an ephemeris and almanac, which they can do usually within 15 minutes if they have a decent antenna and even a marginal sky view. However, it still had to deal with the configuration which was telling it it was in Kyoto, Japan. Not being in California, or Hawaii or Alaska, I suspect that NONE of the SVs visible from there were visible at my QTH at the times they were supposed to be above.


Quote from: metrologist on Yesterday at 13:30:36

It always intrigued me on how these receivers can look for SV's where it thinks they should be. It's easier to understand not know the time and then not knowing which SV's to expect to hear.
"What the large print giveth, the small print taketh away."
 

Offline CJay

  • Super Contributor
  • ***
  • Posts: 4136
  • Country: gb
Re: ($40 GPSDO working well now, my mistake)
« Reply #18 on: August 15, 2017, 02:17:45 pm »
Cjay,

That's good to know about- thank you!

I am using a home-made adapter from SMB to SMA. Both of the powered antennas Ive used with the True Position (one is a Lucent/Max-Rad/PCtel, that runs off almost everything you throw at it, the other is a Motorola SMA patch antenna that runs off 3.3 or 5 volts) have 'worked' i.e. registered as being there.

Isn't the True Position a great deal?

Have you put it into a case yet? I'm looking for an inexpensive case.

I think this project deserves something a bit nicer than my usual cases. (I often use recycled cases from other devices or even cardboard boxes..  that's what it's in now)

So far, I've not let it run a full survey with that antenna so hold off buying one of that particular type, especially as the PackRat document seems to say it won't work. The True Position does seem to recognise it has an antenna and the serial output does seem to indicate it can see sats but I've no idea if it's working properly as yet, for the price it's worth a try but I can lay hands on a timing 'bullet' if need be.

I've got a lovely anodised aluminium rack case complete with LCD cutout and bezel that it will go into I think, there's already cutouts for BNCs on the rear so adding a distribution amp will be simple, as always it's a 'when I get to it' job.

I reckon the True Position is a lovely little board, a bargain if it all works (so far so good).

 

 

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: ($40 GPSDO working well now, my mistake)
« Reply #19 on: August 15, 2017, 04:43:11 pm »

Does it ever show a 1 in the third field (I think) of the $STATUS sentence? If not you are good to go as far as it being recognized.

$EXSTATUS shows number of sats being tracked.

Why havent you done a survey?

If you can- just for a few moments- get that antenna into an optimal position with a decent ground plane under it.. and type in $SURVEY

a CD size, even, would likely do wonders over no ground plane, but bigger is better.. and do the survey which once it starts running, may even complete faster

Ideally, either find a piece of flat steel and place it outside just to get the first survey done, or find some large appliance - say a refrigerator, thats near a window or south facing wall, and try placing the antenna on top of that. 

A magnetic antennas optimal placement is on top of a car.

Without the ground plane, indoors or out, it won't work well.




Quote from: CJay on Today at 08:17:45

So far, I've not let it run a full survey with that antenna so hold off buying one of that particular type, especially as the PackRat document seems to say it won't work. The True Position does seem to recognise it has an antenna and the serial output does seem to indicate it can see sats but I've no idea if it's working properly as yet, for the price it's worth a try but I can lay hands on a timing 'bullet' if need be.



I've got a lovely anodised aluminium rack case complete with LCD cutout and bezel that it will go into I think, there's already cutouts for BNCs on the rear so adding a distribution amp will be simple, as always it's a 'when I get to it' job.

I reckon the True Position is a lovely little board, a bargain if it all works (so far so good).
« Last Edit: August 15, 2017, 04:45:09 pm by cdev »
"What the large print giveth, the small print taketh away."
 

Offline pigrew

  • Frequent Contributor
  • **
  • Posts: 680
  • Country: us
TruePosition GPSDO Commandset
« Reply #20 on: August 16, 2017, 06:20:51 am »
I just bought a three of these boards from Taobao. I'm running them at 12VDC input (which had been suggested on the time-nuts list), though the board seemed OK at 9V, too. The oven uses 5V, and I couldn't find any regulated voltages above 5V. Someone had written 9V on the power input connector.

Example GETVER strings:
$GETVEROCXOGPSNote
$GETVER 12.0.1 0.4850266.0.1 19 fbde 7437 06162200B0000A2004183ACCBliley NV47A1282GT-8031AMy board #1
$GETVER 12.0.1 0.4850266.0.2 19 fbde 7437 06162200C1000A200442A35FBliley NV47A1282GT-8031BMy board #2
$GETVER 12.0.1 0.4850266.0.5 19 fbde 7437 06162200C2000A2009350093CTS 1960017 GT-8031FMy board #3
$GETVER 10.1.2 BOOT          10 9e05 7437 06162200C1000A2005411C55Bliley NV47A1282GT-8031FPackrat Board

I put together a quick control program for Windows, for ease of debugging. I thought that I should post the commands that I've discovered, as I've not seen any list other than the PackRat PPT which was for older firmware.


        // Based on FW 12.0.1
        // Commands:
        //
        // $PROCEED [Send at startup to get past the bootloader]
        //
        // $FACT [Factory preset]
        // $GETBDELAY [returns board delay, nanoseconds]
        // $GETDELAY [returns cable delay, nanoseconds]
        // $GETPOS [return position ]
        // $GETSCALEFACTOR [Returns a float, such as 3.742106e-3]
        // $GETVER [returns version info]
        // $KALDBG <0|1> [Enable reporting of Kalman filter parameters]
        // $PPSDBG <0|1> [Enable or disable timing information every second]
        // $RESET [Unit software reset]
        // $SETBDELAY <n> [-32 <= n <= 32, Set board delay, PPS4 units (roughly 5 or 6 ns). PPS4 is controlled to equal this value]
        // $SETDELAY <n> [-32768 <= n <= 32767Set cable delay, nanoseconds]
        // $SETPOS <n> <n> <n> [set position to Lat/Long/Elevation_MSL, send value returned by survey]
        // $SURVEY <n> [survey for n hours, default is 8]
        // $TRAINOXCO [Start OXCO Training. This restarts the board ($PROCEED needed), and measures freq
        //             change with 500 ADC count]
        // $UPDATE FLASH [update flash memory settings]
        //
        // Other unknown commands:
        // $GETA [returns -1; Attenuator?]
        // $GETP [returns -1 255; Potentiometer?]
        // $SET1PPS  ["$SET1PPS 0"/"$SET1PPS 1"] seems to go to a manual holdover mode, and status changes to 3]
        //           [Seems to return to normal a few minutes after "$SET1PPS 1 1"???  (Status goes 8,16,17,18,0]
        //

        // Messages:
        // $STATUS
        // 1: (Maybe 10 MHz bad, based on packrat docs)
        // 2: (Maybe PPS bad, based on packrat docs)
        // 3: Antenna is bad? 0=good
        // 4: Holdover duration (secs)
        // 5: Number of sats tracked (different than, but within 2 of $EXTSTATUS, perhaps only counts channels 0-7???, range is 0-8)
        // Status [Locked = 0, Recovery = 1, (Forced holdover?)=3, Train OXCO=7, Holdover = 8,
      //        [Startup A/B/C/D = 10/11/2/19 ]
        //        [ (transition from 1 to 0) = (14,15,16,17,18) ] Wait states when transitioning
        //        [ (transition from 0 to 1) = (20,21,22) ]  Wait states when transitioning
        //  (6 = locked, but unknown location????)
        //
        // $PPSDBG 1187153266 3 25.28081e3 -253 -6 2 2 0.0
        // 1: same as clock (GPS Time)
        // 2: Same as $STATUS status, but updates much more often (and seems to skip states less often)
        // 3: Floating point number. Output voltage. Tends towards 29e3 on my board. Proportional to the DAC voltage
        //         On my RevC CTS board, Vbias ~= 6.25e-5*PPS3. This may make sense for a 4.096 V reference: 4.096/2^16=6.25e-5
        //         During startup, it is not put in the result string (this field is blank, so two sequential space characters are in the string)
        // 4: Measured phase offset? Units seem something like 6.5*ns
        // 5: Looks like a saw-tooth between -15 and 15 (or so). Perhaps the quantization error reported by the GPS module?
        // 6: Normally 0, but sometimes 2 (related to holdover/startup?)
        // 7: Normally 0, but sometimes 1 or 2 (related to holdover/startup?)
        // 8: Always 0.0?
        //
        // $EXTSTATUS
        // 1: SurveyStatus [0=normal, 1=surveying]
        // 2: Number of sats (different than, but within 2 of $STATUS, perhaps only counts channels 0-9, range is 0-10)
        // 3: DOP (maybe TDOP?)
        // 4: Temperature (close to FPGA? close to oven?) (my board reads about 45C)
        //
        // $GETPOS (sent after setting position, or requesting position
        // Latitude
        // Longitude
        // Elevation_MSL
        // Correction to MSL to get WGS elevation (add this value to MSL to get WGS ellipsoid)
        // Status flag [(Normal?)=0 on 196 board or =2 on Bliley board, Surveying=3], or maybe a FOM?
        //
        // $SURVEY 40448488 -86915296 225 -34 7129
        // [sent during a survey]
        // 1: Latitude
        // 2: Longitude
        // 3: Elevation_MSL
        // 4: Correction to MSL to get WGS elevation (add this value to MSL to get WGS ellipsoid)
        // 5: Number of seconds remaining
        //
        // $SAT
        // Channel # (0-based, seems to only return channel 0 through 7)
        // GPS Sattelite number
        // Elevation (degree)
        // Azimuth (degree)
        // SNR (dB*Hz)
        //
        // $CLOCK 1187156731 18 3
        // GPS UNIX-timestamp (secs since 1970), but my board is off by 10 years (reporting 2007 while it is 2017)
        // Count of leap-seconds
        // Time figure-of-merit (1=good, 7=bad)
        //
        // $GETVER 12.0.1 BOOT 10 fbde 7437 06162200B0000A2004183ACC
        // $GETVER 12.0.1 0.4850266.0.1 19 fbde 7437 06162200B0000A2004183ACC
        // [version information, I used this to know if I need to send the PROCEED command, String contains BOOT when in bootloader mode]
        // [During boot, only terminated with LF but not CR(or maybe other-way around?)]
        // 1: Bootloader version?
        // 2: ("BOOT" during boot) or (software or perhaps GPS version info????)
        // 3: Status code (same as $STATUS)
        // 4: CRC of something?
        // 5: Always 7437?
        // 6: (Part number: 06162200)(Board Rev: B)(Assembly version: 0000)(Delimeter: A)(Year: 2004)(Week?: 18)(SN?: 3ACC)
        //
        // $WSAT 4 138 209 38 0
        // [WAAS Satellite info, same format as $SAT]
        //
        // $SET1PPS
        // [Sent at boot, but also in response to a $SET1PPS command. Sent every 20 seconds.]
        //
        // $KALDBG 1187203779 0.08 29.59241e3 0.120e-3 0.568 0 0
        // [Only be sent when reference is locked (state=0)]
        // 1: GPS UNIX-timestamp (secs since 1970), but my board is off by 10 years (reporting 2007 while it is 2017)
        // 2: Floating point number.
        //     Resets to 0 at time of lock (and at end of holdover).
        // 3: Floating point number. Magnitude is similar to PPS3, but does not track it so well.
        // 4: Floating point.
        //     Resets to 0 at time of lock (and at end of holdover).
        // 5: Floating point. Smoothed version of PPS3, seems like ~6.5*(PPS phase in ns)?
        //     Resets to 0 at time of lock (and at end of holdover).
        // 6: Flag, always zero?
        // 7: Flag, always zero?

« Last Edit: September 06, 2017, 03:43:11 am by pigrew »
 
The following users thanked this post: BravoV, cdev

Offline CJay

  • Super Contributor
  • ***
  • Posts: 4136
  • Country: gb
Re: ($40 GPSDO working well now, my mistake)
« Reply #21 on: August 17, 2017, 08:43:09 am »
Mine seems to be working just fine, an absolute bargain.

Despite not having a timing antenna (I may or may not order one now) it's tracking 7-8 satellites and the OCXO is stable, one counter tells me it's ~0.2Hz low, the other tells me it's 0.7Hz low, which pleases me an awful lot, neither counter has been calibrated since 2006 so those option 04b Racal High Stability OCXOs really live up to their billing.

Next task is to build it into a case and try to make sense of the data that comes out of the board, I've gotten really rather confused by some of the messages from it and at one point thought I was in the middle of the Atlantic because I'd misplaced a decimal point...

Pictures to follow as long as CDEV doesn't mind a minor thread hijack?
 

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: ($40 GPSDO working well now, my mistake)
« Reply #22 on: August 17, 2017, 02:57:24 pm »
I have had good luck with a Maxrad "GPS-TMG-26N", which I paid around $20 for. (its often sold as Lucent or PCTel too)

It was a very good investment because it "always works", for GPS, it will also accept and run on any voltage delivered by a GPS, anywhere from 3 volts to >12v.

It has a built in SAW filter so I don't know how well it works with Glonass, (1.6 GHz) it may not work at all there. But for use with a timing GPS its perfect.


---

Would love to see photos of your GPSDO, and the case you're planning to put it into.


Pigrew, thank you for posting your command list! 

There are a lot of intermediate status settings that look useful for understanding how it works.

I definitely want to incorporate some way of tracking that process into my own device, however it turns out.



Quote from: CJay on Today at 02:43:09
Mine seems to be working just fine, an absolute bargain.

Despite not having a timing antenna (I may or may not order one now) it's tracking 7-8 satellites and the OCXO is stable, one counter tells me it's ~0.2Hz low, the other tells me it's 0.7Hz low, which pleases me an awful lot, neither counter has been calibrated since 2006 so those option 04b Racal High Stability OCXOs really live up to their billing.

Next task is to build it into a case and try to make sense of the data that comes out of the board, I've gotten really rather confused by some of the messages from it and at one point thought I was in the middle of the Atlantic because I'd misplaced a decimal point...

Pictures to follow as long as CDEV doesn't mind a minor thread hijack?
« Last Edit: August 17, 2017, 03:09:34 pm by cdev »
"What the large print giveth, the small print taketh away."
 

Offline pigrew

  • Frequent Contributor
  • **
  • Posts: 680
  • Country: us
Re: ($40 GPSDO working well now, my mistake)
« Reply #23 on: August 17, 2017, 03:15:16 pm »
Pigrew, thank you for posting your command list!  There are a lot of intermediate status settings that might be useful for evaluating the intermediate wait states as it works.

How did you set yourself up to derive that information?

It was a combination of trying random commands, looking at the packrat PPT file, and trying to write a controller program. I have my laptop outputting a log message when it receives an unknown command or status.

One guy (Mark Sims) posted on time-nuts that he had extracted the firmware image from the flash, but I don't think that it has been posted, yet.

I'll try to continue editing my post with what I learn. (I just noticed a "$GETA" command, but I don't know what it does).
 

Offline cdevTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: ($40 GPSDO working well now, my mistake)
« Reply #24 on: August 17, 2017, 03:26:52 pm »
So now its possible the optimum voltage may be lower than 12 volts. :)

If it does well with a well-regulated 9 volt supply that would mean I can use a decent transformer containing well-regulated linear wall wart that I have.
« Last Edit: August 17, 2017, 03:30:39 pm by cdev »
"What the large print giveth, the small print taketh away."
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf