From: eanders@uconnvm.uconn.edu
  To  : rasmb@bbri.harvard.edu
  Date: Fri, 6 Apr 2001 15:37:13 -0400

reliability at high speeds

Holger,

If you are having problems with noisy data with the Absorbance system, 
there are several things you should check/try.

Insureyou are working at a wavelength where the OD is insensitive 
    to small changes in wavelength.  If you chose a wavelength on the 
    side of an absorption band where there is a sharp rise in OD with 
    wavelength, small changes in monocrometer position cause large 
    changes in OD in the scan.   
Do  not work at multiple wavelengths at the same time.  The 
    wavelength accuracy is higher if it stays at one wavelength then if it 
    switches between 2 or more wavelengths
Use quartz windows below 250 nm.  Sapphire absorbs strongly at low 
    wavelengths and leads to noisier Abs. data.
Adjustthe number of replicates. For Abs. velocity we use 3 
    replicates, for Abs. Equilibrium final scan 10-15 replicates.
Insurethe lamp is cleaned regularly
Periodicallydisassemble the slit assembly and clean the small 
    lenses inside the assembly as well as the slit. Insure you reassemble 
    lenses in the proper orientation.


Additional problems with the Abs system may occur.  The time interval 
between scans is somewhat variable.  Occasionally scans are lost, or 
there is a larger then normal time interval between successive scans.  I 
would think we get 1-2 odd scans per 100 in an absorbance velocity run.  
It does not seem to be speed dependent.


As far as maximum speeds, we usually do not run Interference Velocity 
above 50 KRPM due to the Wagging problem John Philo alluded to.  The 
effects are most noticeable at low sample concentrations.  It can be seen 
in the subtracted pairs DC/DT output from DC/DT+ or similar programs.  
When you overlay a series of subtracted pairs, wagging show up as 
DC/DT curves that appear to have a very different slopes from each 
other.  This is not a vertical displacement which may be due to poor 
jitter/ integral fringe jump, rather a slope to several of the sets.  This is 
noticeable and common in our XLI at speeds over 50 K.

Windows and Double sector cells seem to handle 60 K speeds without 
problems.  We use Carbon and aluminum filled epon cells at 60 K 
although they are rated for 45K.  If one sector of the cell leaks 
significantly you need to stop the run.  The center rib will not withstand 
one empty side and the other full at 60 K.  Both sapphire and quartz 
windows handle 60 K RPM without problems. 

All the Best 
Eric
Eric Anderson
Research Technician II
National Analytical Ultracentrifugation Facility