Workaround for peak picking problem
Frank Holger Foersterling (holger@uwm.edu)
Wed, 17 Mar 1999 16:57:07 -0600
Thank you very much for the many replies to my question
concerning peak picking problems in carbon spectra.
Some answers pointed me into the correct direction that all our
problem spectra had a rising baseline at the edges, even after abs.
This appears to be a problem for peak picking even if the plot
region is restricted to the part of the spectrum with a flat baseline
and all peaks are within that region.
The workaround I came up with for routine use is to just throw away
the edges of the spectrum by setting the parameters STSI and
STSR; i.e. for a spectrum with SI=32768 I set STSI=30720 and
STSR=1024, which causes the first and last 1k points not to be
included in the processed spectrum. That solved the problem in all
three of the troubled specta I tried. One just has to make sure
there are no peaks near the edge of the spectrum, i.e. use a large
enough SW.
Holger
)|(
(o o)
---------------------------------ooO-(_)-Ooo---------------------------
Frank Holger Foersterling, Ph. D.
Director of NMR laboratories
(hf@krypton.org.chemie.uni-frankfurt.de)
--------------------------------------------------------------------
Office: Home:
Department of Chemistry 4420 N. Wilson Dr. #2
University of Wisconsin-Milwaukee Shorewood, WI 53211
3210 N. Cramer Street B-10
Milwaukee, WI 53211
Phone:(414)-229-5896 (414)-962-6334
Fax:(414)-229-5530
-----------------------------------------------------------------------