From: Karen G. Fleming <kgf@csb.YALE.EDU>
  To  : rasmb@bbri.harvard.edu
  Date: Thu, 24 Jul 1997 17:53:52 -0400

interference noise level

Hi All,

Background:

I have been having problems with small amounts of oil in my chamber. The last
EQ experiment I set up, I collected interference data in addition to the
absorbance data that I normally collect. Both data were quite noisy, and the
absorbance data were not even worth analyzing. (I already know how to clean the
lamp and the slit, so I can address that problem).

The SRV values from NONLIN on the interference data never dipped below 1E-2.
Although I have very limited experience with the interference optics, I was
under the general impression that 1E-2 is not a good number. Therefore, I am
currently trying to determine what I should expect for the noise level using
the interferenc optics to visualize the sample on my Beckman XL-I. Since my
absorbance data looks quite messy, I am wondering if my interference lenses
might also be dirty. I would therefore like to hear any or all opinions on the
following questions.

1. I scanned a blank hole (from 5.9 to 7.1 cm). I subsequently did a linear
regression on the linear portion of the scan. The standard deviation of the
slope of such line was 4.57E-05. The service technician suggests that this
number should be below 0.002, and it is.

2. I cleaned the lense in the chamber with ethanol and lense paper. I scanned
as above, and the standard deviation was found to be 4.89E-5. It would appear
that oil is not the cause of the noise in my data.

3. I removed the sample from my cell, rinsed with water several times, filled
with water, and then scanned my cell. The standard deviation was 1.07E-3. Is
this what I should expect??

4. I took my cell apart, cleaned the windows, put the cell back together and
scanned again. The standard deviation of the line was 1.03E-3. Essentially the
same as before.

In all cases, a blank scan from the scallop portion was subtracted from the
scan. The settings for both the "sample" scan and the "blank" scan are the same
(i.e. same pixels/fringe).

Is this a good procedure for checking my system, and do I have a problem with
my windows perhaps?

Sincerely,

Karen






-- 
******************************************************************
Karen Gibson Fleming, PhD
kgf@csb.yale.edu
Yale University, Department of Molecular Biophysics & Biochemistry
266 Whitney Avenue, PO Box 208114, New Haven, CT  06520-8114    
Voice (203) 432-5174     Fax (203) 432-5175
http://www.csb.yale.edu/people/engelman/kgf
******************************************************************