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

[wkb]

Given that this threat has become the favourite hangout of UCCM owners, regardless of the UCCM vendor..

I am looking for someone who owns a Trimble UCCM and is willing to measure the voltages on the OCXO pins.

Mine is still refusing to go GPS Locked, a fact I am of course trying to understand and hopefully fix.

Wilko

[wkb]

hi,

Thanks for the suggestions, I did already try what you suggested in an earlier stage. And for good measure I tried again today. One never knows.. Alas.. no luck. Oh, and pin 39 is indeed jumpered to ground.

Wilko

It's set to and running on GPS. Most of the things you point to in the display are just because it's designed to also use the 8khz signal from the frame. You can safely ignore it.

I'm not sure it will change the selected reference, or if it will have any real effect, but you can try:

SYNChronization:REFerence:DISABLE LINK
SYNChronization:REFerence:ENABLE GPS
REFerence:TYPE GPS

It might change the display.

You might want to start from scratch using SYSTem:PRESet - that will reset to defaults. If nothing is borked, it should come up in GPS mode by itself. No need to enter any commands. The display will show some references to 'Link' or 'Ext' but it should still come up.

The other thing you can do is try grounding pin 39 on the FPC - this should force GPS mode.

[ve7xen]

Given that this threat has become the favourite hangout of UCCM owners, regardless of the UCCM vendor..

I am looking for someone who owns a Trimble UCCM and is willing to measure the voltages on the OCXO pins.

Mine is still refusing to go GPS Locked, a fact I am of course trying to understand and hopefully fix.

Wilko

My Trimble unit also does not work properly; it fails to PLL lock, but the GPS does lock properly and I am fairly sure the OCXO is working properly.

DC Voltages on the unit on the bench (no GPS antenna) powered at 5.5V, clockwise (bottom) from 5V power pin:

4.86
2.49
2.81
0 (RF)
0

[wkb]

My Trimble unit also does not work properly; it fails to PLL lock, but the GPS does lock properly and I am fairly sure the OCXO is working properly.

DC Voltages on the unit on the bench (no GPS antenna) powered at 5.5V, clockwise (bottom) from 5V power pin:

4.86
2.49
2.81
0 (RF)
0

The GPS is fine on mine too, no issues there. While powering with 5.5V as measured on the 5A fuse I get:

4.96
2.49
varying (EFC of the OCXO?)
0 (RF)
0 (GND)

Hmm..  <scratches head> 

UCCM >stat

               - UCCM Slot STATE-

1-1. #Now ACTIVE STATUS ---------------- [OCXO Warm]
1-2. #Before ACTIVE STATUS ------------- [OCXO Warm]
2-1. #Reference Clock Operation -------- [Not Used]
2-2. #Current Reference Type ----------- [GPS]
2-3. #Current Select Reference --------- [NOT SELECT]
2-4. #Current Reference Status --------- [Good Accuracy & Stable]
     #GPS STATUS ----------------------- [Available]
     #ALARM STATUS
     #H/W FAIL [ Ext ]
     #OPERATION ALARM ------------------ [Antenna]
3-1. #PLL STATUS ----------------------- [DISABLE]
3-2. #Current PLL MODE ----------------- [OFFSET OBSERVATION MODE]

Command complete
UCCM >syst:stat?

-------------------------------------------------------------------------------
57964-86     serial number  S35DB68332   firmware ver  1.0.0.1-01 Ext     mode
-------------------------------------------------------------------------------
Reference Status __________________________   Reference Outputs _______________
XX Ext Ref   : [LOS]
                                              TFOM     2            FFOM      3
                                              UCCM A Status[OCXO WARMUP]

>> GPS: [phase:+9.5E-10, settling]
ACQUISITION ...................................................[GPS 1PPS Valid]
Tracking: 7 ____   Not Tracking: 3 ________   Time ____________________________
PRN  El  Az  C/N   PRN  El  Az                GPS      19:57:08     09 Feb 2021
 20  42 126   38    10  19 156
 23  37 127   36    26  57 182                ANT DLY  0 ns
  8   8 279   36    29  12  91                Position ________________________
 27  45 282   46                              MODE     Hold
 18  59  67   39
 16  72 256   43                              LAT      N  51:XXXXXX
  7  11 330   39                              LON      E   5:XXXXXX
                                              HGT               +27.54 m (MSL)




ELEV MASK  5 deg
-------------------------------------------------------------------------------

[wkb]

I am wondering about that "phase" thingy. This roof mounted antenna also feeds a Samsung UCCM and some other GPS receivers, so the signal to the Trimble's GPS rx should be OK I'd say..

>> GPS: [phase:-1.7E-09, settling]
ACQUISITION ...................................................[GPS 1PPS Valid]
Tracking: 6 ____   Not Tracking: 4 ________   Time ____________________________
PRN  El  Az  C/N   PRN  El  Az                GPS      20:15:17     09 Feb 2021
 20  47 117   35    10  27 153
 23  43 119   35    26  48 179                ANT DLY  0 ns
  8  14 282   41    29   6  95                Position ________________________
 18  51  63   41    27  53 287                MODE     Hold

[wkb]

 
I wanted to check inside the GPS can... prying the lid loose should not be a problem or ... ?
Well, think again: there is some glued foam inside that kept the lid nicely glued to the sides
of the can    The tracks of the PCB came loose, you can see them still attached to the can
on the pictures I attached.

The good news is that the GPS works just as it did before 

[wkb]

The can with its lid pried off. Not glued after all, just impossibly tight fitting lid. Hm.

[thinkfat]

The "settling" in the phase data means, IMHO, that the OCXO is still doing retrace, accelerated aging, whatever, that prevent a stable phase lock. PLL is constantly correcting. Should be visible in the DAC data if you connect Lady Heather and plot the DAC for a couple days. Give it a couple more weeks.

[wkb]

The "settling" in the phase data means, IMHO, that the OCXO is still doing retrace, accelerated aging, whatever, that prevent a stable phase lock. PLL is constantly correcting. Should be visible in the DAC data if you connect Lady Heather and plot the DAC for a couple days. Give it a couple more weeks.

Hm, you think Offset observation mode equates to what LH calls settling? Could very well be. I have left it on before, for (I guess) a week or 2. Maybe I am too impatient, and/or the UCCM/OCXO got upset from its long and maybe rough trip from China to Europe..

Wilko

[Bad_Driver]

Thanks Wilko for the measurement! Fortunately my Samsung UCCM arrived yesterday after 4 weeks from China.
Due to the impact of the bad real world it will take some days to set it up und check it. But it is in very good shape.
Seems that it was one of the last available from Aliexpress.

How sensitive is the board to the voltage of the supply? My available 3A-PSU has 5.2 Volts.

[thinkfat]

The Samsung UCCMs are pretty sensitive to supply voltage. If you connect them through the FPC connector, make sure to tie all ground and supply lines together, or the voltage drop over the FPC jumper will be to high.

[wkb]

I am pretty sure it will moan at 5.2V supply voltage.

IIRC the Linear Technology low noise LDO requires
> 350mV headroom. And it supplies 5V at its output.

I run mine at 5.5V input voltage
Wilko

[wkb]

My Trimble unit also does not work properly; it fails to PLL lock, but the GPS does lock properly and I am fairly sure the OCXO is working properly.

DC Voltages on the unit on the bench (no GPS antenna) powered at 5.5V, clockwise (bottom) from 5V power pin:

4.86
2.49
2.81
0 (RF)
0

The GPS is fine on mine too, no issues there. While powering with 5.5V as measured on the 5A fuse I get:

4.96
2.49
varying (EFC of the OCXO?)
0 (RF)
0 (GND)

Not being the most patient person on the planet this afternoon I desoldered the OCXO from the Trimble UCCM and reconnected the whole thing using some "flying wires".

Interestingly enough when I now measure the voltages (same order as above)
I get

4.970
2.496
0   !!
0 (RF)
0 (GND)

Does anyone know if the OCXO used on the Trimble UCCM generates its own Vref, or if the UCCM PCB has circuitry to produce its own Vref? 

The OCXO produces a quite reasonable on-freq/stable 10MHz output (as compared to my functional Samsung UCCM, and after pressing the MACT switch obviously).

Wilko

[Bad_Driver]

As mentioned before last week my SAMSUNG UCCM arrived.
Even though my prepared PSU in the dedicated enclosure delivers only 5.2 Volts (on the fuse on the board I measure 5.00 volts)
I gave it a try - and voila! - it works from beginning! The log wasn't erased, first entry is from 2016, last from 2018, seems to be
an newer board. First location is a radio tower on an oil field in Elk Hills Elementary in CA, USA, not far away from Bakersfield. One year later it was
in a small house under 3 high voltage lines in South San Jose, CA, USA at the end of a golf course. 

Long way for this board, travelled several times around the world!

It found the first satellites after some seconds, locked after 40 min and after one hour I got my first TFOM=1, what I never have seen
with the SYMMETRICOM UCCM. Syk view isn't that good since my antenna is nearby the window and the roof is covered with a lot of snow these days.
Much better than the Sym-UCCM!

Attached a photo of my not yet accomplished housing - I'm awaiting some nylon spacers.
On the right side my PSU - from an old SAMSUNG-monitor, I removed the high voltage (700V) part.
In the middle a FT232- USB-interface and a UART-V24 level shifting board. Above the breakout board I found @ Aliexpress as well.
Left the SAMSUNG UCCM. I changed the delivered short coax cables with BNC-cables.

I think the 50€ @ ALiexpress for the board has been a good investment!

[wkb]

Yes... the "Sam" is really superior to the "Sym", at least when it comes to the GPS. There are comparisons posted on EEVblog which I think (IIRC) that the same is also very good in the frequency department. After all that is what this is all about.

For the curious I have attached a signal strength/ C/N comparison of a "Sam" versus "Trim" (Trimble UCCM). I suppose we might need to take this with a grain of salt, as I do not expect the C/N numbers of Sam and Trim to be directly comparable. It does, however, give an indication.  Units obviously are fed from the same roof mounted aerial. I no longer own a Sym so cannot post the 3rd missing picture.

The one with the most yellow and green is the Sam.


Wilko

[Bad_Driver]

Hi Wilko,

interesting! Although the SAMSUNG claims "POWER!" as expected it's doing better than the SYMMETRICOM.
I set up Lady H. for both and will keep it running for some days to get such nice pictures as you for comparison.
The new PSU is ordered but it is on the long way from China.... I don't want to complicate it with additional step-up/down converters.
But it is a pity that I can't use the old Samsung PSU, fits perfect and the same vendor 

[cdev]

My Samsung PCB is 70 x 115 mm, it lacks the metal shielding piece, and in the smallest dimension it also is around 17 mm. It's also quite sensitive. It even works in my basement.
I'm very happy with it. I still haven't figured out what I am going to put it into. It deserves a nice case and 2A 5.5 volt power supply.

A large sized electrolytic capacitor might be enough to allow it to use a smaller PS. The spike in current draw at the beginning is short.

[jrsphoto]

I've read through most of the posts on these  Symmetricom GPSDO systems (what a fantastic resource), which lead me to purchase one on ebay from zyt_gps.  I'm trying to get things working however, I'm not getting a response from the console port on J5.  Here is what I've got:

• I've got things powered up, a Lucent GPS timing antenna connected, and the ACT light is blinking at about a 2hz rate (slower than 1hz for sure). The antenna is not in a position to get a lock, mostly due to the (6' of snow I have),  but as I understand it, I should still have console access.
• I've got a DB9 wired to a DuPont connector, connected to a the pin-header on J5, RX going to TX, TX going to RX, and ground.
• I've got two different USB serial converters, a Belkin, and an Adafruit, which both work to my Ethernet switch serial port. They identify as /dev/ttyUSB0 - Prolific Technology, Inc. PL2303 Serial Ports
• I do get a drop in voltage on TX when I press the reset button

I use:

Code: [Select]

minicom -D /dev/ttyUSB0with: 57600 8N1 port settings. 

I get no response from the serial port.  The belkin device has RX/TX led's and I can see the TX light flash when I type.
I've also tried Heather, and it also gets no response from the GPS on ttyUSB0.

I do get 10MHZ out when I press MACT.

Am I missing something?  Is there a chance the gps is in some other mode and needs to be reset?

John

[cdev]

Have you tried reversing Tx/Rx? I often make that mistake.

[jrsphoto]

Have you tried reversing Tx/Rx? I often make that mistake.

Yep, I also always do this.  So much that I have to check it multiple times ;-)

[Bad_Driver]

It's not clear visible in your picture but I remember that you can't connect all the holes with screws to ground with the SYM UCCM.
For that purpose I use nylon spacers.

And you need a UART-V24 level shifter when you use a simple USB-UART FT232 based controller.

Look on the picture of my temporary SAMSUNG solution above, there you see the level shifter between the UCCM break out board and the UART-USB-interface.

[jrsphoto]

It's not clear visible in your picture but I remember that you can't connect all the holes with screws to ground with the SYM UCCM.
For that purpose I use nylon spacers.

And you need a UART-V24 level shifter when you use a simple USB-UART FT232 based controller.

Look on the picture of my temporary SAMSUNG solution above, there you see the level shifter between the UCCM break out board and the UART-USB-interface.

The Symmerticom GPSDO that I have, has a metal shield on the bottom, and a metal bracket over the top, which attaches to the 4 mounting holes so no issues with shorting VCC to ground on my module. I just left them attached and used metal standoffs that screw into the bottom of the GPSDO module, and mount it to the vector board.




I only have so many USB to serial  or USB to TTL serial devices available, but I figured out my computer had a serial header on the motherboard.  I had the header cable to bring that out to a slot and low & behold, the GPSDO is alive.  Now I just need to figure out how to get an antenna up in the middle of winter!

Thanks for the pointers Bad Driver!

-John

[Bad_Driver]

Good to hear!
Since most of all Computers nowadays have no real V24 connector I had to got from the shelf a
very old docking station with a much older HP laptop. This wasn‘t very handy on my bench.

So I ordered this little board from Aliexpress which works perfect with the Symmetricom and the Samsung UCCM.

My Sym UCCM has incredible low sensitivity, you have to take care of your antenna location.
I‘m still looking for a suitable solution for my lab.

[wkb]

Inline LNA might help, or better coax cable (lower loss), or a better antenna which produces more signal.

But I still like my solution: Samsung UCCM 😋

Wilko
(who still needs to fix the Trimble UCCM that is lying around, but there is also a HP microwave counter on the bench that I am trying to repair.  And a Rigol scope with a blown input channel for which the parts arrived yesterday.  )

[Bad_Driver]

Yes, that is the best way! I'm happy that I got @ Ebay a Symmetricom active antenna splitter (see picture) for less money!
Unfortunately I have no N-type connectors by the hand so I'm awaiting the delivery. Than it's time to mount my new outdoor antenna
on the roof 

btw: Today with blue sky and all the snow gone I saw the first time a TFOM=0 with the Samsung and TFOM=2 wit the Symmetricom (identical antennas, same location).