[RASMB] RE: radial calibration [originally from SEDFIT user group]

[John Philo]

RASMBers, I am re-posting this thread from the SEDFIT user's list (which
concerns AUC hardware issues rather than SEDFIT itself) to RASMB at Walt
Stafford's suggestion.
 
John
-----------------------------
Mathew,
 
The issue of whether rotor stretching affects radial calibration is is a
very common source of confusion. Remember, the optical components that
measure the radial positions for both optical systems are permanently
mounted on the base plate of the instrument. Thus the rotor stretching does
not affect the measurement accuracy (yes, the sample actually moves out to
higher radii when the rotor stretches, and hence experiences higher forces,
but the optics correctly measure that fact). 
 
The radial calibration is based on the accurate machining of the rotor and
the precision edges in the counterbalance. Based on that machining the edges
will be exactly 5.85 and 7.15 cm from the center of the rotor, but only when
the rotor has not stretched. If you calibrate on a stretched rotor, and
assign those positions as still being at 5.85 and 7.15 cm, the calibration
will be wrong.
 
John

-----Original Message-----
From: sedfit-bounces at bilbo.bio.purdue.edu
[mailto:sedfit-bounces at bilbo.bio.purdue.edu] On Behalf Of Parker, Matthew
Sent: Wednesday, May 10, 2006 7:51 AM
To: Sedfit User's Group Mailing List
Subject: RE: [Sedfit] radial calibration



John,

 

            Now that you mention it, I have a question about the radial
calibration procedure. I understand the reason for doing the calibration at
3000 rpm, where there is minimal rotor stretching, but my question is:
wouldn't the calibration be more accurate if it were done at the speed at
which the data is collected, where the calibration would presumably correct
for the rotor stretching that occurs? I am sure that the answer is "no," but
I am curious about exactly why that is.

 

            Thanks,

 

            Matt

 


  _____  


From: sedfit-bounces at bilbo.bio.purdue.edu
[mailto:sedfit-bounces at bilbo.bio.purdue.edu] On Behalf Of John Philo
Sent: Wednesday, May 10, 2006 10:38 AM
To: 'Sedfit User's Group Mailing List'
Subject: RE: [Sedfit] radial calibration

 

Marina, it is possible to do runs without the counterbalance, but for
interference scans only. The counterbalance must be present for absorbance
scans in order to calibrate the timing of the lamp pulses (the 'delay'
setting). Sometimes the absorbance scans will actually run okay with a
sample cell in the counterbalance position, but often this fails totally or
gives noisy scans because the delay setting is not quite right.

 

I am surprised though that the error message you get is about radial
calibration---I would expect it to fail on the delay calibration. It sounds
like you may have turned on the 'calibrate radius before first scan' option.
You generally do not want to use this option. That option is dangerous
because you may inadvertently do the calibration at high rotor speed (when
the rotor has stretched) and this will cause the calibration to be wrong.
Basically the radial calibration only needs to be done if the optical
components permanently mounted on the instrument have been serviced (and
normally your service technician would do this).

 

John

-----Original Message-----
From: sedfit-bounces at bilbo.bio.purdue.edu
[mailto:sedfit-bounces at bilbo.bio.purdue.edu] On Behalf Of Fasolini, Marina
[Nervianoms]
Sent: Wednesday, May 10, 2006 3:39 AM
To: sedfit at bilbo.bio.purdue.edu
Subject: [Sedfit] radial calibration

I'm running a velocity experiment without the couterbalance cell, because
I've understood that it is possible, but during the run I have always an
error message for the radial calibration.

Is it correct? 

MARINA FASOLINI 
Structural Chemistry 
Nerviano Medical Sciences 
Viale Pasteur 10 
20014 Nerviano (MI) 
ITALY 
tel. +39 0331 581462 
fax. +39 0331 581360 
marina.fasolini at nervianoms.com 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20060510/bebe15ed/attachment.htm>