[time-nuts] Tuning a Symmetricom 1111C

[Ed Palmer]

On 2017-07-10 10:00 AM, Tom Knox <actast at hotmail.com> wrote:
> Hi Ed;
>
> Thanks for the response. Great logic/detail on potential issues and solutions. The power supply issue seems completely intermittent and when running may be completely fine. That said I am leaning toward the power supply being the issue, it would explain all the issues. I first saw the power supply issue with the batteries installed, I removed the fuse and the missing voltages came up and the unit locked. before with those voltages absent it was displaying the lock light every few seconds after warmup. I had assumed a bad battery had created too great a load on the supply. But it failed to come up again when I rebooted and then came up again the next time I rebooted. It is either zero or correct. I have several problematic 4065's and thought they had 1000B's, I will check if any have a 1111C.

Several problematic 4065s??  I can think of a few people that would like 
to have that kind of problem!

> It would be easy to swap them and would resolve the question of if these problems are related. The unit I have may actually be locking but outside it's acceptable EFC range. Phase Noise is exceptional about -120dB @ 1Hz offset 5MHz.

No, I don't think so.  If you look at the manual you'll see that the 
number of tests and measurements the processor makes are such that the 
lock is either good or it's not there.  What you're probably seeing is 
just the performance of the uncontrolled OCXO which is pretty good.  The 
phase noise spec for the 4065C is -106 dBc @ 1 Hz offset.  Your bad 
power supply might actually be improving the phase noise by preventing 
the processor from affecting the OCXO.  It's interesting to note that 
the Allan Deviation at one second for the high performance tube is less 
than half the value of the standard tube.  That means that even at one 
second, the system is affecting and perhaps degrading the raw 
performance of the OCXO.

> I will look at solder on the bricks, for the symptoms that could be the issue.

Not just the bricks.  The entire board.  I can't remember where on that 
board I found bad connections.  I think it affected the front panel 
processor.  I think the 5V was low so the front panel randomly 
rebooted.  I didn't make any notes about that.  I know, I'm bad!

I don't know why I didn't think of it before, but you should read 
through this message thread:

https://www.eevblog.com/forum/metrology/no-luck-on-my-first-attempt-at-acquiring-a-cesium-frequency-standard/

I helped a guy repair a Datum 4040A Cesium Standard.  The 4040A and the 
4065A are both built around the Datum 5045A Cesium module.  They then 
have different power supply and user interface 'wrappers' around the 
5045A.  His 4040A had many bad capacitors in the power supply section of 
the 5045A module.  The power supply in question is hidden under the 
metal cover on top of the module where the big circuit board is.  
Unfortunately, he deleted his pictures, but there's still lots of useful 
troubleshooting info on that power supply section.  I made a bunch of 
measurements and posted them so he could compare with his.

If you want to see some pictures of my unit, they're here:

http://s701.photobucket.com/user/edpalmer42/library/Datum%20FTS-4065A?sort=4&page=1

In the second picture, the power supply is under the sort-of brass 
colored panel with the black and white label.

  Ed