As usual, the AMMRL community responded to my question last week en masse,
offering suggestions and/or commiseration. Everyone who offered suggestions
was helpful; the list is too long to post here, but anyone I haven't thanked
individually please know that I appreciate your responses immensely.
A summary of suggestions follows in the hope this will give others a
starting place for tracking down similar problems in the future. However, in
my case it looks like the problem will turn out to have been a wavegen,
which really bugs me - the outputs looked great on the scope Friday but
swapping them seems to have resolved most of my issues.
The PTS is a logical and even likely source of intermittent problems like
the ones I was seeing, with at least 4 facilities having seen very similar
sounding problems and many more having had their frequency synthesizers
develop problems.
Troubleshooting steps in no particular order:
1. Check the 10 MHz (or 5 MHz) oscillator input. At least two others have
had problems there, and it's easy to diagnose.
2. The PTS output should be a clean sine wave of ~1 V peak to peak at the
set frequency plus 20 MHz (or whatever your IF is).
3. If you have access to a frequency counter, check the exact output
frequency of the PTS. One person had a unit that was 100-200 kHz off, enough
to lose signal but too close to see by measuring frequency on a scope - the
best I could do was something less than 625 MHz.
4. In a similar vein, you can check the tune sweep for a drop out using
mtune/qtune on Varian - presumably wobb would do the same on a Bruker.
5. Several facilities have had the electrolytic power supply capacitors
inside the PTS go bad. One symptom is chopped output at 120 Hz, or others
suggested looking for AC ripple which should come at 60 or 120 Hz at least
in the U.S.
6. Another possibility is that one or more of the frequency "pickets" could
have gone bad inside the PTS - IIRC there are typically 7.
7. Several facilities have had the Varian interface boards that plug into
the parallel ports on the back of the PTS go bad. On the newer console, they
have a green LED that should let you know things are working properly, but
the older cards don't. You can swap the channels by changing the dip
switches (and the output cables) to see if the problem switches channels.
You can also physically swap the cards to then narrow it to either the PTS
or the interface board.
It was suggested that it should be possible and not too difficult to swap
the whole guts or just some pickets between units, but one person suggested
testing the donor first to make sure it's worthwhile. If you have a bad
capacitor, it may be hard to find an really good capacitance match from
anyone other than PTS, but one person recalled finding a similar enough part
that worked.
It was also mentioned that PTS is helpful on the phone and has good service
with relatively speedy turnaround - they'll take your unit, fix any
problems, replace any parts that seem to be dying or are likely to (for
example, they'll replace the power supply caps no matter what), and return
it good as new. One fairly recent repair cost was roughly $850.
--
Dr. Andrew Fowler | University of Iowa
Associate Director | B291 Carver Biomedical Research Building
Medical NMR Facility | Iowa City, IA 52242
319-384-2937 (office) | 319-335-7273 (fax)
andrew-fowler_at_uiowa.edu
Received on Tue Jul 06 2010 - 11:44:20 MST