Quincy Teng
_________________
Question:
We have problems with the sensitivity of our Bruker amx-400 instrument. I
can obtain weak fid only when using the maximum receiver gain (32K). The
fid signals are very weak for both proton and carbon using both BB and
inverse probes. I have swapped preamp (1H and 13C), SE451, PTS interface
board and the receiver cable. The 90 degree pulse width is okay and
ecoupler output power is normal.
Does anyone have suggestions what could cause this problem? If there is
interest in this question I will post a summary of responses to the group.
Thank you very much in advance for all your help.
___________________
Responses:
I had the same or similar problem on my AMX400 and the fault was traced by
Bruker to the FH mixer in the PTS which I subsequently had repaired by
our UK PTS agent, and all seems well now.
The part name used was MSA 0404 integrated circuit (so I was informed)
I hope this helps.
Well, that lets you eliminate from consideration the probe and everything on
the transmit side. It's somewhere in the receive path!
That doesn't leave much left to try. There are some RF switches in the box
which receives the preamps. We have unmounted ours and moved it away from
the magnet - the idea being that the stray field from the magnet was
interfering with the switch working. That may or may not have been the
problem, but the users won't let me move it back. Or you may have bad
switches. Another possiblity is that the input signals to the SE451 aren't
there. We traced another problem to a bad 10 MHz reference in the PTS
which affected the 451 MHz reference input to the SE451.
Since all the RF stuff seem to be in order, why don't you check the AF
(audio freq) amplifier and the digitizer?
I assumed that you have scoped the output of both the decoupler and
broadband channels, even if you get decent 90 deg pws. There may be noise
in the pulse. From the description I think that it may be in the pre-amp.
Somewhere in the back of the console you should be able to 'capture' the
signalfrom the pre-amp before it gets to the reciever. This can be put on
the scope and you should be able to see an FID of the lineshape (proton)
sample. I hope this helps.
I once had a version of uxnmr that shut the digitizer down to 2 bits
(or maybe it was 1) if I changed the spectral width from a default parameter
set that we had been using. (Actually you didn't even have to change SW but
if you simply typed in the value that was already there it caused the
problem.) Using the unaltered default parameters we had normal sensitivity
(due to normal sampling). I believe that you could see that it had done
something to the choice of digitizer in the parameters but I don't recall
what. I have not experienced this sort of thing in xwinnmr. Since I often
see people on this bulletin board saying that they are using old uxnmr
versions, I thought I would point this out in case you might happen to be
using it.
Other than that, I would suggest looking at and/or replacing cables. I
have definitely had the inner pin break off. It's amazing that you can
observe a signal at all that way, but I've definitely seen it happen. After
wasting countless hours on just such problems, I have taught myself to look
for the pin whenever I put a BNC cable on something. Good luck.
_______________________
Quincy Teng, Ph.D. Office (706) 542-1502
Director, NMR Facility NMR Lab -9314
Department of Chemistry Fax -9454
University of Georgia
Athens, GA 30602-2556 teng@sunchem.chem.uga.edu