[time-nuts] Re: TEMEX LPFRS-01 LPRO repair

[Bob Camp]

Hi

The oscillator that generates the 10 MHz is a VCXO. It needs to have a tuning range that “includes” the correct frequency. The sweeping process is trying to line the VCXO output up with the correct Rb “notch" frequency. The key word there is correct, there are other nearby “notches" that you don’t want to lock up on. Just how the VCXO is set up depends a *lot* on various subtle parts of the Rb’s design. 

The adjustment pins on the unit drive the C field on the physics package. As that magnetic field changes, the notch frequency changes. When it does, the locked VCXO output moves as well. If you can tune the unit with the normal controls … it’s locked. 

These telecom Rb’s don’t come with much documentation. Field repair was not part of the plan. If one broke, you sent it back to the factory and they did whatever to make it work or they replaced it with a new one. Things like schematics pretty much don’t exist for any of the modern parts ( = parts designed in the last three decades …). Even things like test point definitions or alignment procedures … not so much. 

If you get into actual debugging or realignment, your first step typically is to start tracing out circuits and figuring out what’s what. Various folks have tried this on various models. Often they then find out that there are *lots* of board revisions, even on the same model part. 

If your unit is 0.249 Hz low, that’s - 2.49 x 10^-8. Typically these Rb’s have an electrical tune range < 1x10^-8. With fancy digital synthesizers, the digital tune range could be pretty large. Either way, that’s pretty far off for an Rb. 

Fun !!!

Bob


> On Apr 27, 2024, at 2:04 AM, VE7HR via time-nuts <time-nuts at lists.febo.com> wrote:
> 
> Changing the 4 47 uF 63V caps did not change anything.
> I tested the old ones and my Sencore LC 102 said they were good so I was not surprised. 
> I will log data from a cold start tomorrow to see if that tells me something. It does lock for a bit from a cold start and when locked the numbers look believable. But I certainly am not seeing the sweeping on the Control voltage but it does change when it locks. 
> 
> This unit has a sticker on it saying Pin 7 is not connected. That’s the analog adjustment input. 
> 
> When I had it apart I was unable to see any trimmer caps on the board. Where should I be looking? There were a few trimmer pots and a couple of tuning cans. 
> 
> It’s not clear to me but what frequency am I supposed to be adjusting?  The 10 MHz output frequency that is on some units adjusted by the external trimmer on pin 7 or something else. ?
> 
> I will have to check calibration but I think the unit outputs a signal that is 249 mHz low in frequency.  I did not have a chance to see if that changes when it briefly locks. 
> 
> Thanks
> Dave
> 
> Sent from my iPad
> 
>> On Apr 25, 2024, at 10:37 PM, Hans Georg Lehnard via time-nuts <time-nuts at lists.febo.com> wrote:
>> 
>> Hi Dave,
>> 
>> I suggest recording the data from the serial interface and displaying it in a diagram.  I did this with my Temex MCFRS-01 at 1s intervals and it looks like the attached picture.
>> 
>> The manual says:
>> 
>> 1. Rb Signal level.
>> During warm-up time this signal is approximately 0V and after it stabilises to a nominal value of 1 to 3.3V.As
>> long as this signal is too low (<0.6V) the internal LPFRS control unit sweeps the Xtal frequency in order to
>> find the Rb absorption dip.
>> 
>> 2. DC-Voltage of the photocell
>> this signal corresponds to the transmitted Rb light level.
>> This is the light of the Rb lamp which is partly absorbed by the Rb cell.
>> The nominal photocell voltage is in the range 2.0 to 3.5 V but must stay stable after the warm-up time.
>> The photo cell voltage is related to the internal reference 5 V voltage.
>> The full scale corresponds to the coded value ‘00’ and the zero (no light) corresponds to the coded value ‘FF’
>> 
>> hgl
>> 
>> 
>>> Am 24.04.2024 um 16:32 schrieb VE7HR via time-nuts:
>>> I have been having lots of enjoyment out of my TinyPFA. It’s so small, easy to carry and quiet compared to my HP 5370B.
>>> 
>>> I am testing it out on my existing pile of frequency sources. One of them is a LPFRS-01. It no longer locks. It locked for a few minutes when I first powered it up.
>>> 
>>> Looking in the archives there has been a long standing thread on repairing these devices with several people in the same spot I am in on a used to work LPFRS-01.
>>> It no longer locks.
>>> 
>>> Output from Serial port on mine is
>>> 8A 1C C3 8A 7C 86 A7 DC
>>> The 4th column is the VCO control voltage that should be sweeping when unlocked. So that is what the other threads seem to indicate.
>>> 
>>> There are some nice teardown pictures of a unit at https://egidy.de/temex_lpfrs_lpro/
>>> With similar data output.
>>> 
>>> I don’t think I have found any service information yet. If I have missed it please point me in the right direction.
>>> 
>>> I have not pulled my unit apart bit I am not seeing the trimmer cap that has been speculated to be the fix in the pictures.
>>> 
>>> Can anyone shed any light on what to twiddle?
>>> If I get it fixed I promise to post my results. It’s somewhat puzzling that there is a common question but no answer. Does this mean not one has fixed it or just not one has told us they have.
>>> 
>>> Dave
>>> VE7HR
>>> Sent from my iPad
>>> _______________________________________________
>>> time-nuts mailing list -- time-nuts at lists.febo.com
>>> To unsubscribe send an email to time-nuts-leave at lists.febo.com
>> <Rb_1265_OK.JPG>
>> <Temex_LPFRS_Manual.pdf>
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at lists.febo.com
>> To unsubscribe send an email to time-nuts-leave at lists.febo.com
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com