A persistent problem with DISNMR over many years has been the way an
integral plot request will hang the computer indefinitely if one of the
'regions of interest' generated by 'AZF' extends up to the edge of the
SW. It took a very long time to pin this problem down, not least because
its occurrence is random, being dependent on an unusual spectrum
content.
To cure this, Bruker can only suggest widening the SW so that none of
the integral regions extend too far, but this is not terribly practical
if you want to give your customers good digital resolution and also wish
not to be profligate with disk space. We have a sufficiently mixed bag
of samples going through our automatic AC200 that we get peaks occurring
anywhere between +20 to -40ppm. While we don't expect to get great
spectra from some of the more extreme things, at least it would be nice
if the oddities did not hang the whole system. (Especially as it often
seems to hang at about 8pm, leaving a whole night's sample-changer load
waiting for you in the morning!)
To save me re-inventing the wheel, has anyone devised a method for
overcoming this problem? It could be done by preprocessing the region
file, to eliminate regions that went off the end, or by generating the
region file in such a way that it would never have a rogue region, or
perhaps you have another way. I'd rather not use a fixed, safe integral
region, because that would make the integrals more difficult to use.
Thanks for any suggestions,
Alan
----------------------------------------------------
Dr A S F Boyd Chemistry Department
NMR Spectroscopist Heriot-Watt University
phone: +44-131-451 3214 Riccarton
fax: +44-131-451 3180 Edinburgh
cheasfb@bonaly.hw.ac.uk EH14 4AS
http://dava.che.hw.ac.uk/ Scotland
----------------------------------------------------