Hello all,
I hope you are all well. I am having a complicated and compound issue with
our 400 Avance I instrument that is hard for me to understand. The instrument
will sometimes in the middle of run give the error:
"rcuerror: Scan speed overrides accumulation. Increasing DW or D1, decreasing
TD or switching off the FID display may help fifo cycles [none of which actually
help]"
After this error, the lock signal develops aberrant signals that look like
power fluctuations (see image attached
<
https://drive.google.com/file/d/1ShWqCrU3JfF_TBBSvtRbXt96k9X15oFU/view?usp=sharing>).
Sometimes I have seen these signals occurring during a POPT in which case the
phase of my signals changes mid-run and makes it difficult to optimize the
parameter
(second image<
https://drive.google.com/file/d/1Nk7WyDgIZEOo3F2VUBv_wakGIKOxRloq/view?usp=sharing>).
In any case if this error occurs during automation with ICON, acquisition fails
but queued up samples continue to attempt to run leading to successive errors
involving other components. The most common of these is "GCU error: not enough
time for on the fly calculations, no points collected during the last scan"
Clearing the error which lights up on the LCB is usually done by power cycle of
BSMS/2 and/or AQS but often times if we try to run right after power cycle,
error will come up again. If we wait a couple of hours, the instrument will run
usually run just fine for several days to a week. Error seems to occur most
commonly on long overnight 13C or 2D experiments.
A possibly related error that comes up fairly often (but not always) is
"topshim failed: fieldmap signal to noise too low". Even though using topshim
plotall I can see that the field map looks fine and the actual spectra are
perfectly normal-looking. 2H pulse length is normal as checked by running 2H
spectra with the with the lock off, lineshape shimming is within spec, and we
are mostly using just CDCl3 for solvent with the "convcomp" option.
Both the 20 MHz and 10 MHz signals look normal when the error occurs. Running all
of the board tests (rcutest, etc) only one memory error on the GCU shows up. I
suspect this board may be the root cause but after talking to Bruker email support
for months they are fairly certain it is due to the AQS power supply failing.
Replacing both parts is extremely expensive for our fairly small facility so
I'm wondering if anyone has any ideas about what might be causing this issue or
further diagnostic recommendations.
Most of the evidence seems to be pointing to either the GCU, or the GAB and
not the AQS power supply but I don’t have the technical expertise to confirm.
We are running Topspin 2.1 on Windows XP Pro SP3
Thanks for any advice on this,
David O.
Dr. David Onofrei
NMR Facility Director
San Diego State University
chem-nmr_at_sdsu.edu<mailto:chem-nmr_at_sdsu.edu>
Received on Mon Jun 14 2021 - 07:59:56 MST