Symmetricom TrueTime XL-DC GPSDO

[pacAir]

@ZigmundRat:  I agree with your assessment regarding the high likelihood that the firmware in each unit is custom for the unit configuration at the time of shipment.  Most TrueTime devices seem to be like this (based on the Model Numbers).  I happen to have two XL-DC units that came from the same batch with the same model number configured exactly the same but different than any of my other XL-DC units (I have 5 or 6).  I also have six TCU-800 Time Code Generator/Translator units... two of them came from the same batch with two identical model numbers (identically configured) but different than all the others.

This would seem to dovetail with my observation regarding all of the chassis number labels that change over the shelf-life of each chassis until it acquires its final configuration before shipping.  I do not know if the end digits in the model number are actually indicating specific features or just a sort of incrementing numeric code that must be looked up in a shipping log file.  This might be determined by collecting all of the data as I laid it out and comparing.  I was going to create a spreadsheet of my unit's data so if anyone wants to participate they could compile the data from their unit(s) and send them along for inclusion.

To make this a bit more understandable, here are the specifications from my very first (XL-DC #1) XL-DC receiver along with relevant photos:

XL-DC#1

1.   Model Number:   151-602-817-12
2.   Warranty Expiration:  No Warranty Label affixed to chassis.  (However, chip date codes indicate 1999 timeframe)
3.   TrueTime XL System Version:   TRUETIME XL  SYS VER 029
4.   GPS-XL Version (Shown at boot-up):    GPS-XL v1.046   
5.   EPROM Part Number:    182-6497  v001
6.   Antenna Type & Voltage setting:    Standard, 12Vdc
7.   GPS-XL CPU Board (Model, Firmware & Version marks):    87-611-139 Rev A
8.   GPS Carrier Board (Model, Firmware & Version marks):    87-733
9.   GPS Receiver Daughter Board (Model, Firmware, Chip & Version marks):    87-ACE3-2 Rev B,    GPS RX: 39818 Rev C (USA)
10. OCXO Option Board (Model & Revision) (If installed):    87-399-8
11. OCXO Oscillator (Manufacturer and all data) (If installed):     10.0 Mhz, Wenzel Associates,  500-06449 Rev D, +12Vdc, 13717-0048
12. Type of power supply (AC, DC or both):    AC Switching (standard)
13. List of all installed option boards (Name, Model, Firmware & Version marks):   Network Interface / RS-232 Card, 87-6004-XL Rev A
14. List of all installed options (Oscillator type, Plug-in Card Names, LCD Backlight, Power Supply, etc.):   No large LCD Backlight.  Oscillator, Power Supply & Option Card noted above
15. Revision History  (Didn't note this at the time, will have to open it up and document.)
16. Internal & External Photos:

GPS-XL Assembly:
http://www.pacair.com/Transfer/GPS/XL-DC_1/DSCF3459.JPG

Power Supply:
http://www.pacair.com/Transfer/GPS/XL-DC_1/DSCF3463.JPG

Rear Connectors:
http://www.pacair.com/Transfer/GPS/XL-DC_1/DSCF3201.JPG

OCXO Board Assembly:
http://www.pacair.com/Transfer/GPS/XL-DC_1/DSCF3450.JPG

OCXO:
http://www.pacair.com/Transfer/GPS/XL-DC_1/DSCF3461.JPG

Option Card Internal:
http://www.pacair.com/Transfer/GPS/XL-DC_1/DSCF3457.JPG


Steve

[Colin1256]

@ZigmundRat: Thank you! I believe I misunderstood you. I thought it may be possible to connect an external OCXO to two of the BNC Jacks, provided they would be connected properly within the unit. and use one of them to feed in 10 MHz and use the other to output a trimming signal.

I have exactly the type you are referring to; I just have the VCTCXO on the main PCB, the area behind the large (backlit) Clock LCD is empty on mine.

In an ideal world, I would find an OCXO board somewhere or design and make one myself (provided I can find some pinouts of the connectors I'd need to tie into, as well as determine if my unit's firmware supports it) and upgrade my base-unit to include it, but I could also live with connecting one externally provided the unit supports this somehow.

I will open my unit up and get you the full list of what's inside, but here is the preliminary data:
1.   Model Number:   151-602-247-12
2.   Warranty Expiration:  No Warranty Label affixed to chassis.
3.   TrueTime XL System Version:   TRUETIME XL  SYS VER 030
4.   GPS-XL Version (Shown at boot-up):    GPS-XL v1.047
5.   EPROM Part Number:    182-6225  v002
6.   Antenna Type & Voltage setting:    Standard, 12Vdc (currently reconfigured to 5V as I am testing a different antenna)
7.   GPS-XL CPU Board (Model, Firmware & Version marks):    don't have it recorded right now
8.   GPS Carrier Board (Model, Firmware & Version marks):    87-733 Rev A
9.   GPS Receiver Daughter Board (Model, Firmware, Chip & Version marks):    87-ACE3-2 Rev C,    GPS RX: 39818 Rev C (China)
10. OCXO Option Board (Model & Revision) (If installed):    none
11. OCXO Oscillator (Manufacturer and all data) (If installed):     none
12. Type of power supply (AC, DC or both):    AC Switching (standard)
13. List of all installed option boards (Name, Model, Firmware & Version marks):   none
14. List of all installed options (Oscillator type, Plug-in Card Names, LCD Backlight, Power Supply, etc.):   Large LCD with Backlight.


Cheers,


Colin

[ZigmundRat]

Here is the data for my XL-DC #3:

1.   Model Number:   151-602-922
2.   Warranty Expiration:  No Warranty Label affixed to chassis.  (However, date codes indicate 1997-2000 timeframe)
3.   TrueTime XL System Version:   TRUETIME XL SYS VER 030
4.   GPS-XL Version (Shown at boot-up):    GPS-XL v1.047   
5.   EPROM Part Number:    182-6534  v003
6.   Antenna Type & Voltage setting:    Standard, 5Vdc
7.   GPS-XL CPU Board (Model, Firmware & Version marks):    87-611-40 Rev N/C
8.   GPS Carrier Board (Model, Firmware & Version marks):    87-733 Rev A
9.   GPS Receiver Daughter Board (Model, Firmware, Chip & Version marks):    87-ACE3-2 Rev B,    GPS RX: 39818 Rev C (USA)
10. OCXO Option Board (Model & Revision) (If installed):    87-399-11
11. OCXO Oscillator (Manufacturer and all data) (If installed):     10.0 Mhz, TrueTime (Oscilloquartz) 8666-C2S,  ?45.866.602 42A (label degraded), +12Vdc
12. Type of power supply (AC, DC or both):    AC Switching (Standard Artesyn NFS40-7608)
13. List of all installed option boards (Name, Model, Firmware & Version marks):   Low Phase Noise, 87-357-24 Rev C
14. List of all installed options (Oscillator type, Plug-in Card Names, LCD Backlight, Power Supply, etc.):   Large LCD Backlight.  OCXO, Power Supply & Option Card noted above
15. Revision History  (None in particular noted. Various crossouts/handwritten changes)
16. Internal & External Photos:

GPS-XL Assembly:
https://www.dropbox.com/s/d8h0su5jt3zv37h/2016-07-27%2021.37.39.jpg?dl=0

Power Supply:
https://www.dropbox.com/s/i7j89svxifisffn/2016-07-27%2021.38.37.jpg?dl=0

Rear Connectors:
https://www.dropbox.com/s/ss1sonu01q25h3c/2016-07-27%2021.40.03.jpg?dl=0
https://www.dropbox.com/s/eoytanw0jfnxvjq/2016-07-27%2021.40.14.jpg?dl=0

OCXO Board Assembly:
https://www.dropbox.com/s/9yhhc1yxklr98iw/2016-07-26%2023.06.14.jpg?dl=0

OCXO:
https://www.dropbox.com/s/1g63hy8hbvc6xuz/Photo%20Jul%2027%2C%2021%2039%2018.jpg?dl=0

Option Card Internal:
https://www.dropbox.com/s/yshs2dknc9yc2lx/2016-07-27%2021.38.03.jpg?dl=0

Added Note: The OCXO here is the same as the OCXO on the 86-357-3 Low Phase Noise module (TrueTime/Oscilloquartz 8666-C2S)

[gpsdo_nut]

For my (older) TCXO-only XL-DC:

1.   Model Number:   151-602-579
2.   Warranty Expiration:  7/2000
3.   TrueTime XL System Version:   TRUETIME Mk III SYS VER 020
4.   GPS-XL Version (Shown at boot-up):    GPS-XL v1.036
5.   EPROM Part Number:    182-6103 v003
6.   Antenna Type & Voltage setting:    Standard, 5V
7.   GPS-XL CPU Board (Model, Firmware & Version marks):    87-610 Rev H(?)
8.   GPS Carrier Board (Model, Firmware & Version marks):    87-731 Rev 04
9.   GPS Receiver Daughter Board (Model, Firmware, Chip & Version marks):    87-CM3-3 Rev B,    GPS RX: 37424-61
10. OCXO Option Board (Model & Revision) (If installed):    None
11. OCXO Oscillator (Manufacturer and all data) (If installed):     None
12. Type of power supply (AC, DC or both):    AC Switching (Standard Artesyn NFS40-7608)
13. List of all installed option boards (Name, Model, Firmware & Version marks):   NTS-XL, 87-6003-XL, 187-6009 V0006, TI sync. 87-6007TI-78
14. List of all installed options None
15. Revision History  None


@ZigmundRat - the ? in your 8666-C2S is a "945..."

[gpsdo_nut]

@pacair - Just out of curiosity - what's the maximum time constant configured on your Wenzel OCXO?

[pacAir]

@gpsdo_nut:  By "Time Constant" I am assuming (by the previous messages) you are referring to "Oscillator Response Time".  I cannot give you that info presently because none of my units are currently hooked-up and running (rooftop antenna was taken down for house maintenance and tree-removal recently and has not yet been put back up).  When I try to use an indoor antenna it generally takes forever (if at all) to get all the way to a fully-working state at my home location.  I am in the Pacific Northwest and have a somewhat limited southerly "view" to typical satellite orbits because I live on the northern face of a large chunk of dirt & trees.  I had somewhat attenuated operation on both "C"-band satellite and Dish Network for this same reason.  An outdoor elevated rooftop antenna works well enough, though.

I checked my OCXO units to see if they stored their last known oscillator values (some specs are) but Oscillator Phase & Response Time values must be considered as "real-time" values only and are not stored for off-line reference.

[ZigmundRat]

And for contrast, one of my Rb units:

Here is the data for my XL-DC #2:

1.   Model Number:   151-652-699-1
2.   Warranty Expiration:  Warranty Expires 07/2001
3.   TrueTime XL System Version:   TRUETIME XL SYS VER 027
4.   GPS-XL Version (Shown at boot-up):    GPS-XL v1.043   
5.   EPROM Part Number:    182-6222 v006
6.   Antenna Type & Voltage setting:    Standard, 12Vdc
7.   GPS-XL CPU Board (Model, Firmware & Version marks):    87-611-139 Rev N/C
8.   GPS Carrier Board (Model, Firmware & Version marks):    87-733 Rev N/C
9.   GPS Receiver Daughter Board (Model, Firmware, Chip & Version marks):    87-ACE2 Rev NC,    GPS RX: PII-M1, Trimble 36576-A Rev C (USA)
10. OCXO Option Board (Model & Revision) (If installed):    N/A
11. OCXO/Rb Oscillator (Manufacturer and all data) (If installed):     FEI 5650A Options 25,29,32,42,54
12. Type of power supply (AC, DC or both):    AC Switching (Standard Artesyn NFS40-7608) plus Artesyn NFS40-7624 for Rb (24V)
13. List of all installed option boards (Name, Model, Firmware & Version marks):   T1-8, 87-6000T1 Rev A, NTS-XL 87-6003-XL FW:182-7009 v007
14. List of all installed options (Oscillator type, Plug-in Card Names, LCD Backlight, Power Supply, etc.):   Rb Osc., T1 and alarm card, NTS NTP server card
15. Revision History  (None in particular noted. Various crossouts/handwritten changes)
16. Internal & External Photos:

GPS-XL Assembly:
https://www.dropbox.com/s/fqdkzy4tjl7wzzc/Photo%20Jul%2030%2C%2019%2056%2015.jpg?dl=0

Power Supply:
https://www.dropbox.com/s/cfw6bhm8wc4o8g0/Photo%20Jul%2030%2C%2019%2048%2023.jpg?dl=0

Rear Connectors:
https://www.dropbox.com/s/xvcjyk6ploqfqkq/Photo%20Jul%2030%2C%2019%2049%2025.jpg?dl=0
https://www.dropbox.com/s/cg53a50g63i1lq3/Photo%20Jul%2030%2C%2019%2050%2024.jpg?dl=0

OCXO/Rb:
https://www.dropbox.com/s/ngf053df97icfja/Photo%20Jul%2030%2C%2019%2047%2051.jpg?dl=0

T1 Option Card Internal:
https://www.dropbox.com/s/tq2t0rqti7nfyqc/Photo%20Jul%2030%2C%2019%2051%2019.jpg?dl=0

NTS-XL NTP Card Internal:
https://www.dropbox.com/s/elijthay6xj7i25/Photo%20Jul%2030%2C%2019%2053%2046.jpg?dl=0

Other Photos:
Front Panel:
https://www.dropbox.com/s/muzv5zoplqlxjg7/Photo%20Jul%2030%2C%2019%2044%2054.jpg?dl=0

Overall Top View:
https://www.dropbox.com/s/m573zt8byltnauk/Photo%20Jul%2030%2C%2019%2045%2026.jpg?dl=0

A couple of notes on my Rb units:
+) Rb was adjusted via serial port on the bench to be as close as possible to 10MHz @ 5V EFC, then replaced in the XL-DC. Reported DAC values are currently around -9000 
+) Note the stacking of the base unit and Rb power supplies - not a fan of that but nothing has cooked yet
+) LPN module is not recognized if installed 
+) Temex LPFRS will work as a FE-5850 replacement, but it draws more power, and creates more heat.
+) As compared to the 'slimline' 1 RU XL-DC with Rb option and another 2RU unit I've seen (where the bottom of the chassis is used as the heat sink), these have an external heat sink for the Rb. That heat sink was noticeably warmer when testing with the LPFRS, but the LPFRS was within its temp spec.

[Raoul Duke]

@pacAir : do you have any protocol information for your RD-2?  I just bought one on fleabay and while it is wicked cool, there is a paucity of information available for it online.  And, in true Datum "Hey, let's name a bunch of incompatible things the same and be surprised when we go under" form, it looks like there are a dozen different versions.  I have the RS-422 version with nine 16 segment characters and no colons.  And there are 4 undocumented dip switches on the CPU board - do you know if there is any documentation on these?

I would like to use the RD-2 with my 9390-6000; I have plans to stick a Raspberry Pi 2 into the expansion port to create an NTP server.  (Long term plan is to remover the 19MHz resonator on the RPi and drive the SOC clock gen with the 10 MHz from the 9390-6000 so the NTP should have unmeasurable drift).  The serial port of the RPi can drive the RD-2.

(I also have a Datum 9100 timecode generator which I use as a rack mounted digital clock - it has the cool HP BCD decoding LED displays that look like 5x7 displays but it seems to drift at 2 - 3 seconds / week).

[pacAir]

@Raoul Duke:    It has been awhile since I have played with my RD-2 displays so I don't remember all the info on the dip-switch settings (and mine may even differ from yours!) other than:

Switch # 2:  If turned ON, then the 3 leading digits (which show Days of the Year when displaying IRIG-B and a few other "time" codes... as opposed to "Range Timing" or "Countdown" displays) will be turned off and not displayed.

I seem to recall that one of the switches may determine whether or not the colons are flashing (per-second) or fixed continuously "ON".  Another function that comes to mind is whether the display BLANKS when signal is lost or whether it simply freezes at the last display update value.  That is 3 out of 4 switches possibly. 

Also, I think you will find that your display does have Colon LEDs, you just can't see them until they are ON because they are not mounted against the rear of the bezel panel as the actual digit devices are.  If you never see them (even if time is being displayed) perhaps that is the 4th switch function... Colons ON/OFF?

As with most other gear, I believe that the RD-2 only reads the position of the DIP switches at boot-up time. So, if you want to experiment, you have to cycle power after each configuration change before the unit will "see" the changes and operate accordingly.

The option switches are for configuration of features and not selection of supported input codes.  Selection of codes is automatic and those that are available depend on how the board is stuffed and what firmware is in it.  They may all be RD-2 displays but just like all TrueTime gear there are many custom flavors!

Here is a copy of the documentation I have on the RD-2 (Operation Manual, Drawings, some parts lists, etc.):

TrueTime RD-2 Manual:
http://www.pacair.com/Transfer/GPS/RD-2/TrueTime_820-202.pdf

I got a good laugh regarding your Datum 9000-series comments.  I have had a few of their devices (older, newer, rack-mount, displays, time code generators, aircraft-mount, etc.) and all of them have 9000-series model numbers!


Steve

[Raoul Duke]

Thanks for the information on the dip switches.  Unfortunately, the 820-202 manual doesn't help much with my RD-2 which is an 820-263: it has no IRIG input, and the DB25 connector is labeled RS-422 instead of RS-232.  I see no reason for 16 segment LEDs unless it can display text messages, which would be nice because I hope to alternate between time and date.  I suppose I'll have to brute force it some day when I have the time.

[pacAir]

Doesn't your RD-2 power up with the text messages in the display?  Mine comes on with "Truetime" and a "Version 3.2" (or something like that) on boot up.  These displays are fully alphanumeric capable.

Steve

[Raoul Duke]

Yes: "TRUETIME"  "560-3138" and "VER 1_3" in series after power on and lamp testing all the vertical and horizontal segments (but not the diagonals) before blanking.  The EPROM on the CPU board is likewise labeled "560-3138" and "ver 1.3".  Pins 3, 7 & 8 of the DB25 connector are tied to J3.   These are nonsensical for RS422 as the connector is labeled but they do correspond to RXD, DSR and GND for RS232.  So which is it?  RS232 or RS422?  Who knows?

Quite frankly, Datum & TrueTime  seem like they were confused 95% of the time and it must have been so painful to work with their equipment.  My RD-2 is missing the IRIG code input BNC as well as T1 & R2 internally that would present the correct termination impedance for the IRIG signal, saving like what, $2?  Would it really have been that hard to make a single, universal RD-2 that accepted both IRIG and serial inputs?  One that could be configured by DOCUMENTED dip switches to perform any of the dozen incompatible specialized tasks, making the units interchangeable in the field?  Is it really surprising that their equipment has been orphaned and the companies absorbed?  I wonder if they ever actually printed service manuals for all the different variants of the same item, or did they rely on long term employee Larry who knew all the versions?  Perhaps Larry retired (or went insane) and there was no one left who knew how to produce or repair the equipment.

[Wilton]

I know this topic hasn't been visited in a while, but maybe some of you could give some feedback.
I just purchased a used XL-DC (with NTP)  GPS works fine.  Time works fine. IRIG-B works fine, NTP works fine.
But the primary reason I got it was to serve as a 10 MHz reference for test equipment.
I connected the 10 MHz out to my HP 8594E Spectrum Analyzer.  It was displaying a clean carrier (happened to be another 10 MHz reference).  As soon as I connected the XL-DC as a reference oscillator, the displayed carrier looked really dirty.  It had noise sidebands extending out at least 500 Hz on either side almost as high as the carrier.
I've evaluated a lot of thing:  Tried different signal levels into the Spectrum Analyzer Reference input, using attenuators, AC coupled it.  I thought maybe it didn't like the fact that it was a square wave, but I tried another square wave source I have and it caused no problem.  I finally started analyzing the 10 MHz output of the XL-DC.  It didn't look too bad on the spectrum analyzer, but when I put it on my digital scope and looked at it many cycles after the trigger, I could see a couple of ns of jitter.  I also mixed it to zero IF and looked at the noise on the scope, using FFT.  There were noise sideband spurs about 100 hz away that were only down 20 dBc.  My measurement noise  floor was -60 dBc.  I also did the jitter test on the 1pps output, and it was clean.
So my question is, is this sort of jitter on the 10 MHz output normal, or did I get a defective product?  Is it not appropriate to use the 10 MHz output as a reference for precision test equipment?  I assumed it was.

Thanks,
Wilton Helm
Embedded System Resources

[pacAir]

There are MANY different configurations for the TrueTime XL-DC.  For example, only two of the 5 or 6 I have are equipped with a TCXO or OCXO.  I have not evaluated the 10Mhz output on each of these for comparison but I would be willing to bet the units with a full TCXO or Rubidium oscillator have better noise and phase-jitter specs. 

Also, there is a plug-in card for this series that provides high accuracy, high-stability and low phase-noise 10Mhz output signals.  The card has a 5Mhz TXCO mounted on it.  Here is one currently on eBay:

http://www.ebay.com/itm/TrueTime-Symmetricom-XL-DC-86-357-24-Quad-10MHz-Card-Low-Phase-Noise-BNC-Output-/361986177463?epid=901632620&hash=item54480eb9b7:g:u8oAAOSw42dZIeo8

The "TrueTime Symmetricom XL-DC 86-357-24 Quad 10MHz Card Low Phase Noise BNC Output" provides 4 BNC output ports to drive multiple receiving devices such as test gear.

Steve

[Wilton]

I think it is going to have to go back.  I still don't know if it was defective or just never intended to meet my specs.  I really don't want to spend the money for the extra plug-in card.  The seller sent me a second main card to try, but it ended up not having 10, 5 and 1 MHz outs at all (and stupid me, was jumpered for 12 V so I lost an antenna).

But I have learned some things, so am posting here in case it helps someone else.  The 10 MHz signal comes from a different clock oscillator (I don't think it is even a TCXO).  The replacement board didn't even have it populated, so it obviously isn't necessary for the functioning of the rest of the circuit.  It is apparently disciplined to 1 PPS or some other signal from the GPS board.  There are a lot of FPGAs in this device, and I'm sure some are frequency comparators for such things.  It sort of looks like 1 PPS (or possibly some other signal) is propagated with high precision and then everything else is locked to that.

[ZigmundRat]

FWIW I just put my unit on the spectrum analyzer, with a 1MHz span centered on 10MHz. Taking the 10MHz off the back of the main card shows nothing ugly. The worst spur is at ~9.81 MHz and is ~66dB down. There are some closer in spurs at about +/- 50kHz, I am guessing from the switching power supply. This is on an OCXO unit with a low phase noise option card.  Looking at the output of the LPN module, all artifacts are down in the grass and not visible. Never looked at it before but nice to see it *is* low(er) noise.

So just for giggles, I fired up my TCXO unit. Same spur on the 10MHz output, slightly different level but still about -66dB.

Switching references for the analyzer (now the XL-DC square wave) and looking at the Thunderbolt OCXO, the TBolt looks reasonable. in fact, switching back to the TBolt as ref, and looking at the TBolt signal I don't see any real difference regardless of the reference used. But we could be comparing different fruits here. What were your spectrum analyzer RBW/VBW and span? I'm using an Advantest R3261 here and went down to +/- 10 kHz and 30 Hz RBW. I also have the ref oscillator going through a variable attenuator and a distribution amplifier. Input to the spectrum analyzer was around 1.9v PtP. But as you say, looking at it on the spectrum analyzer didn't reveal a lot.

I don't think I'll be able to duplicate your other tests (i've got other things going at the moment), but it might be educational to do so at some point. I was wondering: did you look at your other square wave source the same way?

[texaspyro]

I am in the process of adding Truetime XL (DC and AK) support to Lady Heather.  It looks like I can also support the XLi (does NTP), but the XLi manual has some vague/conflicting info for the output of these commands:

F18 - version
F72 - faults
F73 - alarms

Does anybody have an XLi and can capture the serial port output response to these command?

[pacAir]

I don't have an XLi but I wanted to jump in and thank you for the "Lady Heather" application!  I have used it on a number of GPS devices and would LOVE to see TrueTime support added!  I have a number of TrueTime XL-DC, XL-AK and several older TrueTime models in my collection.

Steve

[texaspyro]

I don't have an XLi but I wanted to jump in and thank you for the "Lady Heather" application!  I have used it on a number of GPS devices and would LOVE to see TrueTime support added!  I have a number of TrueTime XL-DC, XL-AK and several older TrueTime models in my collection.

A guy is loaning me an XLi,  but it does not have the GPS board.   I think it will answer my questions, though.

My XL-DC has the FTM-III power line frequency analyzer card in it.   Attached is a screen shot of it monitoring the power line freq and calculating various statistics and a histogram and ADEVs of the frequency.

I have some code in there to also support the oscillator frequency measurement option, but am unable to test it.

[Kean]

I know this is an old thread, but it is one of the few here mentioning TrueTime GPSDO's.

Just in case anyone is following this thread and knows anything about the TrueTime GPS-DC, I make a post recently looking for more info.

https://www.eevblog.com/forum/testgear/looking-for-info-on-truetime-gps-dc-416-rubidium/

[Ulaker]

Hello,

anybody of the people familiar with TrueTime xl-dc gpsdo around? Mine does not behave as expected, it never gets lock stays in lookung for satellites status for ever even with the time and  lon/lat/alt correct after a few minutes.

Any help welcome,

Ulaker 

[ZigmundRat]

There could be several causes for this. Most probably:

+) Do you have the right antenna? It could be 5V, 12V, or require a down-converter. This is all usually marked on the back by the antenna connector. You might include a picture of the back, and antenna details.

+) GPS receiver could simply be old/defective. It’s probably a Trimble Ace, or maybe a SVee 6,  and those are somewhat hard to find these days.

+) You didn’t wait long enough for a fix. These receivers are not particularly sensitive, and require a good antenna placement. Also, it may take 15 minutes or more for the receiver to find the strongest satellites and then download an almanac.

+) Does FN 60 ever show satellite signals?

Hopefully this is some help. Let’s see where it goes.
 

[Ulaker]

Ok fourth attempt to post a follow up Tried few time to craft a followup with quotes and photos but failed to comply to the rules and regulations...

I did change the battery after initial attempts to get this old horse saddled, but the fresh battery did not help, the only thing that changed is that the current year stays in the NVRAM when power cycled.


Would like to believe that this antenna is the right one since date time and location do get on the displays whithin a few minutes after power on.

The unit is acting quite fast within a few minutes to show on both displays the current time and on the small display the correct longitude/latitude/elevation, The status is enumerating a few integer numbers that I take to be the staelite currently in field of view. Than transits to looking for satelites and eventually not enough satellites. So far it was understandably but even after two weeks with several restart (power cycling) The FN60 does not contain tracked or bad satellites.

OCXO is unfortunately missing, does that matter for satellite fix?



Is the 3/4" PVC pipe stub compulsory under the antenna? Is proximity to metallic structure detrimental to signal integrity?
Is the Firmware possibly obsolete? do the currently active staellites of the constellation use a newer protocoll for almanach distribution? Why is the serial port spitting out only day of year and current hour minute second? Not reacting to any input?
Is there a unlock procedure for the serial port?

Questions over questions to which I am more than happy to have somebody to talk about.

Thanks for helping solve this puzzle,

Greetings from Germany,
Ulaker

[ZigmundRat]

Yeah, that looks like the down converter antenna, lucky you were able to get it. And I see the receiver is marked as 'Downconverter required'. So you're off to a good start 
It's possible to convert the board to standard GPS (no downconverter) but it's not easy, and you need another receiver (practically unobtanium).

You say that it does track for a while and then loses the satellites. (And yes, those integer numbers on the bottom line are the tracked satellite numbers.) Apparently something is warming up and drifting, causing it to lose lock. I don't think the OCXO is required for GPS to get satellite fixes, but it would of course never go to locked state. FYI: All of my units have OCXO or Rb options and require them to be present or the system does not initialize. So I think you must have the TCXO version since it starts up OK.

Question: How do you know you are missing an OCXO?

Answers:
+) No, the PVC stub is not an absolute requirement.
+) Yes, proximity to metallic objects is detrimental to signal integrity. Best to place above and away from them, best as you can.
+) Yes, the firmware is obsolete We're talking mid-90s here. That said, the unit should work (I have 4 of them, different build options) with the only real effect of the date/time being wrong, which you can fix with the FN68.
+) There are no changes to the almanac distribution
+) The default is to send the time every second
+) You probably aren't sending it the right commands. Have you read the fine manual? 
+) AFAIK, there is no serial port lockout, but you can lock the keypad.

I doubt its the cause, but check FN65 and make sure all satellites are enabled.

If you're feeling adventurous, you could check the oscillators on the CPU board. These are under the GPS board. Actually there should be two. One a 16.368 MHz for the GPS clock, and a 10MHz TCXO. It's possible if the 16.368 oscillator is off frequency enough, it could cause poor sensitivity and other symptoms. I don't exactly recall 100% but I *think* the MCX connector closest to the board edge has this signal on it, the other one goes to the GPS antenna connector.

I also notice you have the NTP card. Are there other modules installed?

[Ulaker]

Question: How do you know you are missing an OCXO?

The dangling connector stub between the receiver & NTP card and the display would have been my guess that it is for the OCXO it has supply
and coax as well as wires that I deemed to be the control/adjustment to trim the 10MHz to track the GPS time. I do not really know, it is a guess.

I locatet the 16.368MHz signal path and  it is 16.3860517  is it close enough? Should it be tweaked closer to 16.368000?
There is provision to adjust the frequency in the can of the oscillator of  but with the stacked board in place not accessible.
Not really clever design. 

The serial communication is still a bit of a mystery to me at this time since my naive expectations to interact with the module did not produce
any gain of insight. Maybe need to really closer verify that the Rx/Tx is really what I think it should be. The PC terminal emulation is receiving
the datagram for DOY.HH.MM.SS followed by a "?" nothing more than that and any keys I attempt to  send to the device are not producing any
change in behavior. The stream of DOY.Time is coming steady. Only at the beginning it is counting from zero then when it gets the decoded
signal from the first Satellite it changes to the correct current DOY and time.

I will try to reposition the antenna to be a bit further away from metallic structure, so maybe the signal/noise improves and the normal sturt up
procedure completes.

No other modules and I have not yet checked if the NTP card is responsive, I still needs a converter to talk to it over the LAN port. If the sturtup ever completes
the next step might be to illuminate the large display and if I get to attach the NTP card to the LAN will try to set up a local NTP service?

I am doing this just as a challenge to see if I still can get old stuff to work again.

Thanks once more for assistance,
Ulaker