[time-nuts] [Bulk] Re: Ultra High Stability Time Base Options for 53132A

[Bob Camp]

Hi

> On Apr 12, 2015, at 8:03 AM, J. L. Trantham <jltran at att.net> wrote:
> 
> Bob,
> 
> Thanks for your thoughts.
> 
> It is hard to tell from the pictures that accompany the items but it appears some use the 'Vref' pin to obtain the control voltage and some do not.  If that indeed is the case, I wonder if using the Vref pin would offer a greater likelihood of better performance.

A lot depends on if the example you get *has* voltage on the Vref output or not. You can come up with a stable reference source. without using it. 

> 
> Also, how would you go about testing these Time Bases or 53132A?

Simple approach:

1) Apply power, see if the controller  cuts back, toss out about 1 in 5 because they don’t. Simple check with a DVM.
2) Check output power with a power meter. X mark the ones that have < 2 dbm out into a pile. That’s about 3 out of 5. 
3) Check the output for spurs (not sub harmonics, but true spurs) on a spectrum analyzer. X mark the ones that have issues. That’s maybe 1 in 12.
4) Toss them on a TimePod and compare them to something good. X mark anything with ADEV > 5x10^-12 at 1 to 10 seconds. That’s about 3 out of 5.
5) Check the same data for phase noise. X mark anything with major humps in it. That’s maybe 1 in 5
5) Toss them in a Delta chamber and do a temp run. Look at the data. X mark anything that’s over 1x10^-9 p-p. That’s about 3 out of 5
6) Apply voltage to the EFC pin and X mark any that don’t at least trim +/- 5x10^-8. That’s about 2 out of 5. 

Now try and sort through the pile for any that either were not tossed out at step one or made it through with no X marks. Multiple X marks are
not terribly surprising because there may be a common failure that impacts multiple specs. So far I have not seen any without X marks on them.
It is interesting to note that the “low power out” X marks do not correlate to a very great extent with the other failures. There’s a fairly well known
capacitor failure at the output stage that impacts power without messing up things like ADEV or temperature stability. 

Again, when new or properly installed there would be no toss outs or X marks on any oscillators in a 10 to 20 piece sample. You would have to 
get a much larger test group to have any chance of finding a problem unit (if you could at all). The fact that 100% of the parts had some sort of
issue says to me that the handling of the parts was not at all good. 
 
>  Installing them in the 53132A, leave them on for 24 hours, do the 'CALTIMBAS' routine, then use the 53132A to measure your 'best' source?  If so, which source would you use?  I have GPSDO (TBolt, Z3816A, Z3805A, Lucent Z3810A, Lucent RFTGm-II-XO and Rb), Rb (5065A, LPRO, and others) or CS (5061A).  Also, how would you set up the 53132A and collect the data?

The gotcha is that the 53132 is not accurate enough to take some of the key data above all by it’s self. 
> 
> My question is how can I get the most stable and accurate 'stand-alone' counter, not having to connect an external reference?

If you are going to run a counter stand alone, *and* turn it on and off, old cell site OCXO’s may not be your best bet. Their warmup characteristics will probably get in the way. Those
OCXO’s were designed to be turned on once and run for years and years. A run in of three days was not at all uncommon. 

Bob

> 
> Thanks for your help.
> 
> Joe
> 
> -----Original Message-----
> From: time-nuts [mailto:time-nuts-bounces at febo.com] On Behalf Of Bob Camp
> Sent: Saturday, April 11, 2015 5:04 PM
> To: Discussion of precise time and frequency measurement
> Subject: [Bulk] Re: [time-nuts] Ultra High Stability Time Base Options for 53132A
> 
> Hi
> 
> All of the MV-89’s that I have seen are very much a crap shoot in terms of what you get. Out of a dozen or so I’ve bought from various sources, none has met the original specs on the unit. There have been a wide range of issues. Doing a full test on one involves a lot of work. The problems with some of them are major the problems with others are fairly hard to spot. 
> 
> I do not believe that’s a reflection on the original design of the part. I’d bet that nearly 100% of them meet spec when new. I think it’s purely due to the amount of violence done to them when they are pulled from boards during the savage process. 
> 
> Bob
> 
>> On Apr 11, 2015, at 8:01 AM, J. L. Trantham <jltran at att.net> wrote:
>> 
>> Has anyone done any testing/comparison of the three aftermarket UHS 
>> Time Base Options for the 53132A that are on theBay?
>> 
>> 
>> 
>> One is from Poland:  111643536543
>> 
>> 
>> 
>> Two are from China:  181698758773 and 331420737911
>> 
>> 
>> 
>> All use the Morion MV89A and all appear to be 'plug and play' with the 
>> automated time base calibration process of the 53132A.
>> 
>> 
>> 
>> I recently added the 3 GHz Prescaler from the seller in Poland
>> (111631156891) and was very satisfied.  No connection other than a 
>> satisfied customer.
>> 
>> 
>> 
>> I also recently added the HP Opt 010, one of the 10811 based time 
>> bases, and was very satisfied as well.  Just wondering if there is something better.
>> 
>> 
>> 
>> Thanks in advance for any thoughts.
>> 
>> 
>> 
>> Joe
>> 
>> 
>> 
>> 
>> 
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to 
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.