I am curious as to whether anyone has observed cases where long term
signal averaging makes the signal disappear. It is a software problem with
xwinnmr 2.0 (including all the patches ever provided) on an ARX500 and an
ARX400. Bruker seems not to have heard about this from anyone else so
perhaps it occurs only with ARX's which is a "select" group. We definitely
did not have this occur on any previous version of xwinnmr or uxnmr on
these spectrometers. After about a year and a half I have finally figured
out the cause. The problem never occurs if you do not display the
acquisition window while the acquisition is going on. Probably depending
on where you are in multiples of 8 when you display the acquisition window
and how many scans you display, the routing of the data needed for phase
cycling appears to get messed up. After we realized that the problem was
occurring it was mostly solved by block averaging of no more than 256 scans
at a time. The curious thing that I observed was that usually my first
file had only noise (not even solvent), but all the rest in an overnight
series of about 50 files were virtually identical. So I figured that it
either it was my aura or else I was doing something to cause it. I would
really appreciate hearing from anyone who has observed anything similar.
Below is a description of one of the many times I did this to make sure
that it is reproducibly caused by displaying the acquisition window. (It
is.) Surely I am not the only person in the world for whom this occurred.
I plan to put on xwinnmr 2.5 sometime soon and what I really want to know
is whether this happens in that version.
Jane
Select parameters such that the signal to noise ratio increases as the
square root of the
number of scans. (For the data shown below, this is definitely the case
when the
acquisition window is not displayed at all during the acqusition.) Type
ii, then zg. Type tr, phase the peaks to be positive absorption. Set the
regions to be used for the signal region and the noise region for the
signal to noise calculation, sino. Occasionally type tr, note the number
of scans stored, type efp, then type sino. All tr commands were typed from
the processing window. Only where noted, the display was switched to the
acquisition window, but no commands were typed. The return button was then
selected after observing ~3 scans in the acquisition window.
ns sino
11 27.42
19 36.59
28 45.50
34 50.04
44 57.13
53 62.98
acquisition window for ~3 scans
63 69.33
acquisition window for ~3 scans
76 73.70
acquisition window for ~3 scans
86 66.20
98 59.48
105 56.62
119 50.70
129 46.39
140 42.17
149 39.39
163 35.85
176 32.30
202 26.28
218 22.90
234 19.55
745 29.16 - the peak is negative
766 30.70 - the peak is negative
776 31.06 - the peak is negative
Dr. Jane Strouse
Dept. of Chemistry and Biochemistry
UCLA
Los Angeles, CA 90095-1569
(310)-825-9841 - voice
(310)-825-0393 - FAX
strousej@chem.ucla.edu