A look at my Symmetricom GPSDO / 10MHz reference (OCXO + Furuno receiver)

[Briain]

Hi

It's interesting that what you described so closely echoes what my own (err; I mean my loan) one also displays via LH (see attached screen-shot) so that's why I was wondering if the 'fault' might be based on a vaguely related issue (and also perhaps be largely irrelevant)? Before I received this unit, GM8BJF had it running - alongside a Simmetricom unit - into a dual channel 'scope, and it all looked pretty darned good, so after sparking it all up here, that is exactly what I did, and after a few hours of settling time, the two traces remained 'visually locked' (over a several day period) so despite all of the mildly concerning 'alarms' displayed from both the serial terminal outputs and the LH screen, it does appear to all be working just as it should.

Bri

PS I've just recovered the R Pi from its other nefarious duties, but after just consuming a bottle of Finca Carelio (which is 14.5% ABV, by the way) I think that it might be wise to not scrabble about in my cellar of particularly sharp and dangerously pointy things, right at the moment, but despite these dangers (and my great age) I will attempt to do so in the morning, and then - assuming that I survive the ordeal - I'll report back about the F/W, just as promised (and if not, you'll likely read of my untimely demise in the local newspapers; please do feel free to LOL). 

[metrologist]

That unit firmware is much newer. I was hoping it would be something similar as my Jupiter GPS module, where the earlier firmware just did not implement the 10kHz PLL...

When you say locked for days, I would not have considered my two units to be locked, but I guess as long as it does not drift more than a full cycle, then that is still locked. Mine have been off for a while, but I think it's time to look at them again. Maybe by the weekend I'll achieve good stability and have better confidence in what I can report.

I would like to get my RPi3 running LH - no clue where to begin with that. Can the RPi run two instances and two separate modules?

[texaspyro]

I would like to get my RPi3 running LH - no clue where to begin with that. Can the RPi run two instances and two separate modules?

Yes, the PI can run multiple instances of Heather.

You can download the X11 (linux) version from KE5FX.COM, but it is the v5.0.   I have newer code that adds a lot of new features and devices that I can send you.   Building on Linux is very easy.  Unzip the archive into a new directory,  switch the that directory, check out the readme.txt,  type "make".  The new version supports touchscreens and has some tweaks for optimizing the display on smaller screens (like the 800x480 PI touchscreen).

[edpalmer42]

Below shows a screen-shot from LH with the GPSDO's elevation mask temporarily set at 0 deg (just to obtain a picture of what it could see). I planned it all to give me a 20 deg clearance (and the elevation mask now set to 20 deg) and other than some pesky trees to my east (fetch me a chainsaw, immediately) I pretty much achieved my goal.

Bri (GM8PKL)

Brian, the best way to set your elevation mask is to look at the outputs from LH's SAA and SAE commands and see what setting is best for your situation.  Step carefully because I noticed that for some reason, the SAA command claims that my Tbolt is getting good signal strength from straight north +- ~35 degrees!  Pyro, any thoughts on that?  Is it LH or the Tbolt that's confused?  SAS and SAD correctly show no such nonsense.

Also, Pyro, any idea when we can expect a new official version of LH?  It sounds like there has been a lot of improvements since the last release.

Ed

[texaspyro]

Step carefully because I noticed that for some reason, the SAA command claims that my Tbolt is getting good signal strength from straight north +- ~35 degrees!  Pyro, any thoughts on that?  Is it LH or the Tbolt that's confused?  SAS and SAD correctly show no such nonsense.

Also, Pyro, any idea when we can expect a new official version of LH?  It sounds like there has been a lot of improvements since the last release.

You are probably actually getting the highest signal levels from the north.  SAA shows signal level vs azimuth, regardless of the satellite elevation.  Just because you don't have good satellite coverage to the north does not mean that those that are to the north don't produce good signal levels.  Also, you can have obstructions, etc to the south.

SAW weights the azimuth values by the inverse of the elevation angle. The outline of the SAW plot is very similar to the SAS signal level color map and better shows where the best signal levels are coming from and local obstructions.

The SAE command will show the satellite signal level vs elevation.  It will put a marker on the plot where the signal level falls off rapidly (and a marker at the current elevation mask).   Set your elevation mask to 0 (or a low value),  clear the satellite signals (CS command), let it run for a few (eight?) hours, and check the SAE command.   Set the elevation mask angle to the indicated value.  The &a command will also set the signal level mask for most receivers (on the Tbolt it also sets the OCXO gain, TC, and damping values).

I hope to get the next release out in a couple of weeks.  I need to finish documenting the support for some of the new devices (like rubidiums, HP5071, freq counters, etc) and the new commands.


BTW, try the ZU command.  It will show all 4 signal level plots at the same time.

[edpalmer42]

Step carefully because I noticed that for some reason, the SAA command claims that my Tbolt is getting good signal strength from straight north +- ~35 degrees!  Pyro, any thoughts on that?  Is it LH or the Tbolt that's confused?  SAS and SAD correctly show no such nonsense.

You are probably actually getting the highest signal levels from the north.  SAA shows signal level vs azimuth, regardless of the satellite elevation.  Just because you don't have good satellite coverage to the north does not mean that those that are to the north don't produce good signal levels.  Also, you can have obstructions, etc to the south.

SAW weights the azimuth values by the inverse of the elevation angle. The outline of the SAW plot is very similar to the SAS signal level color map and better shows where the best signal levels are coming from and local obstructions.

The SAE command will show the satellite signal level vs elevation.  It will put a marker on the plot where the signal level falls off rapidly (and a marker at the current elevation mask).   Set your elevation mask to 0 (or a low value),  clear the satellite signals (CS command), let it run for a few (eight?) hours, and check the SAE command.   Set the elevation mask angle to the indicated value.  The &a command will also set the signal level mask for most receivers (on the Tbolt it also sets the OCXO gain, TC, and damping values).

Okay, I think I see where I went wrong.  You're right that there are satellites to the north, just not very far to the north.  See attached picture.    Anything on the 'wings' to the east and west is suspect because the antenna is mounted on my south-facing balcony.  Everything to the north is going through the building.

By the way, the nibbles out of the bottom of the SAS command aren't due to obstructions or multipath.  There just aren't any satellites that orbit in those areas.  As the orbits precess over a period of months, the nibbles fill in.

Brian, this is still the best way to set the elevation mask.

I hope to get the next release out in a couple of weeks.  I need to finish documenting the support for some of the new devices (like rubidiums, HP5071, freq counters, etc) and the new commands.

Great!  Looking forward to it.

Ed

[Briain]

...When you say locked for days, I would not have considered my two units to be locked, but I guess as long as it does not drift more than a full cycle, then that is still locked...

I forgot the version is actually shown on the LH screen (that screen-shot was taken a while ago).

When I say locked for days, I was using an old CRT 'scope, so I didn't leave it running 24/7, but I simply fired the 'scope up several times (basically, whenever I walked past it) over a multi-day period and the two outputs hadn't changed (WTR each other). They were actually almost exactly 180 deg apart and remained that way, but of course, I would most likely have missed any periodic 'glitches'. Once I get the other Trimble PSU built up, I plan to use a LCD 'scope (I'm pondering one of these super-cheap Chinese ones) and leave it in view of my comfy chair for a few days (far more interesting viewing that than UK TV) and thus hopefully, I'll notice any such events.

Bri

[Briain]

...Brian, the best way to set your elevation mask is to look at the outputs from LH's SAA and SAE commands and see what setting is best for your situation...

and

...The SAE command will show the satellite signal level vs elevation.  It will put a marker on the plot where the signal level falls off rapidly (and a marker at the current elevation mask).   Set your elevation mask to 0 (or a low value),  clear the satellite signals (CS command), let it run for a few (eight?) hours, and check the SAE command.   Set the elevation mask angle to the indicated value.  The &a command will also set the signal level mask for most receivers (on the Tbolt it also sets the OCXO gain, TC, and damping values)...

Thank you both for the above information (I was unaware of the above process, but I have just set the mask to zero and tomorrow, I will see what LH proposes) and the very interesting note about the &a option and additional T'bolt configuration.

My initial assumption was that even if you did have a good line of sight, you still wanted to mask the lower angle ones for propagation reasons, so that was in my mind when selecting the modest antenna height (and a 20 deg mask). It will be really interesting to see what Lady Heather commands me to do, but looking at the old survey picture in my initial post, I expect it will suggest something more like 25 deg (and I might raise the antenna height).

I'll report back in 24 hours.

Bri

[cdev]

If you use a timing antenna, it does something similar, typically they are designed to reject multipath and favor satellites which are higher in the sky. 

[texaspyro]

The yellow tick mark on the SAE command is where the signal falls off to around 85% of the peak value.  At that point, the signal generally falls off rather quickly.   The "&a" command sets the signal elevation mask to midway between the 85% level and the current elevation mask...  I felt setting it to the 85% level might have been a little aggressive.

[cdev]

Texaspyro,

I am really looking forward to trying the new Lady Heather program out.

Thank you for doing all this work  to make such a useful program available!

[K1FPV]

I wonder if anyone on here has a copy of the HP/Symetricom/Microsemi 58531A software used on the 58540 receiver? I've got a working receiver that is apparently working fine putting out a solid 10 MHz. signal as well as a 1PPS output.
Thanks,
Bill
K1FPV

[texaspyro]

I wonder if anyone on here has a copy of the HP/Symetricom/Microsemi 58531A software used on the 58540 receiver?

Does Lady Heather do anything with it?  Try it with the /rx5 command line option.  I know a lot of the commands are a bit different than the standard HP GPSDO's.

If you can part with it for a couple of weeks, I can probably get Lady Heather to work with it.

[K1FPV]

I've tried Lady Heather with it. Unfortunately, Lady Heather doesn't see it on the serial port. I'm going to try a couple other things with it and if I have no luck, perhaps I'll go that route and see if you can do it. I know Heather sees regular HP/Symetricom receivers as I've use Heather on my 58503A and 58503B and it works fine.

Bill
K1FPV

[K1FPV]

Too bad I'm having the trouble getting some control software for the HP58540A because I've got the receiver running and the 10 MHz. output has been rock solid on 10 MHz. My counter which is already GPS referenced, measures 10 MHz down to .1 Hz, and the output is rock solid and steady on 10.000,000,0 MHz. (BK Precision 1823A)

[cdev]

Is there any chance that your serial port is one thats incompatible with LH running on the computer you are using, for example, LH's computer may want to see a 12volt rs-232c and your device may have a 3.3 or 5 volt UART or vice versa. Also, its common for Rx and Tx to get mixed up.

If that is the case, poke around on the headers, especially around any chips on the board that might be a level shifter.. because I would be surprised if a GPSDO board of that caliber did not have both TTL and RS-232c level serial IO available somewhere. However the one that is the default might well vary depending on what model the customer ordered.

This is just me speculating based on what Ive seen come through here and my own (also telecom surplus) GPSDO

I've tried Lady Heather with it. Unfortunately, Lady Heather doesn't see it on the serial port. I'm going to try a couple other things with it and if I have no luck, perhaps I'll go that route and see if you can do it. I know Heather sees regular HP/Symetricom receivers as I've use Heather on my 58503A and 58503B and it works fine.

Bill
K1FPV

[texaspyro]

Most HP equipment needs a null modem adapter/cable.  I permanently install a low profile null modem to DE9 adapter on all my equipment that needs one.

[cdev]

Try reversing Tx and Rx. does it work then?

[K1FPV]

Hi Guys,
Yes, I'm using a null cable. I am able to communicate with the receiver using Termite 3.2 but not with PUTTY. the big problem is most of the commands are quite long and I need to have the command sheets in front of me.....and inputting the commands is a slow process. Fortunately, I do get replies though. It obviously would be much easier if I had the 58531A software package.

Too bad Lady Heather isn't communicating with the receiver as is the case with SatStat and GPSCon.

[NivagSwerdna]

Oh dear... after a period of lying idle I decided to fire up my Symmetricom GPSDO but I've just fried it and there is magic smoke.

Please can someone remind me how many volts/amps I was meant to apply to the board and also what the voltage of the onboard boost regular is meant to be.

Thanks

[texaspyro]

Please can someone remind me how many volts/amps I was meant to apply to the board and also what the voltage of the onboard boost regular is meant to be.

6V 2A

[NivagSwerdna]

OK. I can now testify that supplying the board with 12V is not a good idea. Small tantalum fire at C11 and I think the PTN04050CAS is fried... hopefully not too much beyond... we shall see.

[texaspyro]

I fried a Trueposition GPSDO when the 12V wall wart I was running it on went wonky and dumped 20V into it.  I pulled the flash chip and dumped the firmware and figured out some undocumented commands.

[cdev]

You guys have convinced me, even though its arguably redundant, I'm now using LM340T12  for my GPSDO and also an LM340T5 running off of that for the controller's supply.

[NivagSwerdna]

I took a closer look today.. after desoldering the input cap C11 and the PTN04050CAS I re-applied 6V and get a surprisingly encouraging result with various flashing LEDs so I think probably more has survived than I expected.  Testing the PTN04050CAS in isolation shows it is knackered and pretty much pass through now.  I will replace C11 and the boost regulator with equivalents PTN04050CAZ? when I have time over the forthcoming holidays.

I think the PTN04050CAS only supplies 12V to the OCXO and given I only ever supplied 12V I feel that the later stages of that should be OK.  At least one of the other regulators must be working otherwise I would not get flashing sequences.

Fingers crossed.

(There is a 1k35 ohm resistor across pins 1 and 3, this resistor is actually on the other side of the board... 1k35 sets a 12v output).