[time-nuts] Troubleshooting Fluke PM6681

Magnus Danielson magnus at rubidium.dyndns.org
Mon Aug 17 02:46:34 EDT 2015

David and others,

I will have to come back on the details. However, I'll give you the 
overview from the top of my head.

The one calibration value kept in CMOS backup is the length of the 
calibration pulse. This can be altered over GPIB. The program will sweep 
over the value range (don't know if it does anything smart) in order to 
find one that gives the minimum RMS error, which is the value then put 
into the counter. The counters auto-calibration then uses that for 
interpolator calibration.

The original DOS program used a particular Philips synthesizer and old 
GPIB interface, but I heard nothing about this being essential to the 
process in modern day. I would guess that a synthesizer able to sweep 
the interpolator is really the key.

I will see what I can dig up. Would be nice to have one of these again.


On 08/16/2015 10:27 PM, davidh wrote:
> Magnus,
> I have a few of these counters and would love to see the details of the
> calibration process you mention.
> Cheers.
> david
> On 17/08/2015 5:16 AM, Magnus Danielson wrote:
>> Ole,
>> I checked with a former Pendulum employee, and free off memory, he
>> recommend trimming up the 100 MHz until Error 2 does not show. Sensing
>> it directly can be difficult, FET-probe essentially mandatory.
>> Indirectly a 10 MHz is possible. A problem is that trimming with the
>> hood off causes a different thermal setup than when the hood is on. If
>> you dare, make a hole in the hood so that you can trim it with the hood
>> on, that is what they did.
>> CMOS backup battery eventually fails, and then you need to replace it
>> and re-calibrate it. I have the details jotted down, but it seems that
>> this is not your issue.
>> Cheers,
>> Magnus
>> On 08/16/2015 12:20 PM, Ole Petter Ronningen wrote:
>>> Hello.
>>> I have a Fluke PM6681 that has issues. When I got it, it gave results
>>> all
>>> over the place, but after adjusting the 100Mhz multiplier-chain, it
>>> seems
>>> to be much better; at least I get std.deviationwell within spec using a
>>> split pulse to input A and B, 100 samples. I'd like to get it
>>> professionally calibrated, but I don't want to send it in with known
>>> issues
>>> - the cost of calibration will not be refunded if the instrument can
>>> not be
>>> calibrated, and I believe it is no longer repairable.
>>> So, the issue is that it fails the ASIC test (test 6), with err 2.The
>>> service-guide lists a number of signals to be checked in test-mode,
>>> but I
>>> am not able to see any of them. Since the counter seems to function, I
>>> can
>>> only conclude that the signals are not present due to this error.. (Or
>>> that
>>> I am mistreading the guide and looking in the wrong place, probably at
>>> least as likely..)
>>> Anyway, I cant seem to find a description of this particular error, does
>>> anyone know what it means?
>>> Thanks,
>>> Ole
>>> _______________________________________________
>>> time-nuts mailing list -- time-nuts-JSkTLETqlTM at public.gmane.org
>>> 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.

More information about the time-nuts mailing list