AMMRL:Problems with SMS-100 sample changer

From: Frederick Morin, Dr. <frederick.morin_at_mcgill.ca>
Date: Mon, 25 Jan 2010 09:57:53 -0500

Hi everyone-



We have a Mercury 300 console recently upgraded from Sun Ultra / Solaris / Vnmr 6.1C to a Dell PC / RedHat 5.1 / VnmrJ 2.2D after the hard disk gave out. Everything worked fine for a month and a half at which point we also installed an SMS-100 sample changer. This sample changer has been giving me problems ever since. There are numerous ways in which it messes up. I will describe a few.



Say I submit 7 samples to the queue to run proton and C-13 spectra - it will fail to run perhaps 3 of them because it will use a solvent different from the one specified when it was submitted to the queue. The result is that the proton spectrum will run after gradient shimming and auto locking but produce a spectrum of course with the wrong chemical shift scale. It then skips over the C-13 for that sample, ejects it, inserts the next sample and runs it resulting in 2 "Active" entries in the Study Queue.



If I create an acqlog file in /vnmr/tmp it shows the error

----------------------------------------------------------

Seq: Findz0

au

PSG



SVPAR

SAVEDATA

auto: Experiment started

auto: ADC overflow

SVPAR

SAVEDATA

auto: Acquisition aborted

signal too large - reduce gain or pulse width

--------------------------------------------------------------



Sometimes it will run a sample fine, eject it and put it back in the rack and then just sit there and the same error appears in the acqlog (even with no sample in the magnet).



There are other ways in which the sample changer does not work but most of the failures are as described above. If I submit one sample at a time and let it run to completion before I submit another I have not seen it fail, but of course this makes it pointless to have a sample changer! I thought maybe the action of submitting another sample while a previously submitted sample was just starting was causing the problem. There is an entry in the Buglist for 2.2C (not reported for 2.2D) that resembles that but it was supposedly resolved. However, if I submit say 7 samples to start later in the NightQ, check each entry to ensure that it is correct and make sure that no one else touches the instrument before they run, there are still problems.



I have checked the sample ready light on the autospin board and it is always showing the correct status, on when a sample is there and off when it isn't.



I have swapped autospin boards with another console to no effect. I have reinstalled VnmrJ and as expected that did not help.



Has anyone had similar problems with an equivalent setup?



Thanks,



Fred Morin, PhD

NMR Facilities Manager

Department of Chemistry

McGill University

801 Sherbrooke St. West

Montreal, QC

H3A 2K6



Phone: 514-398-6219

Fax: 514-398-3797

frederick.morin_at_mcgill.ca
Received on Mon Jan 25 2010 - 04:58:17 MST

This archive was generated by hypermail 2.4.0 : Fri Jun 16 2023 - 16:16:47 MST