[RASMB] Temperature calibration

Kirk C Aune kaune at sbcglobal.net
Thu Dec 23 00:43:01 PST 2004


David Hayes wrote:

> At 12:48 PM 12/22/2004, Kirk C Aune wrote:
>
>> I have read this thread with interest and can no longer contain 
>> myself and am willing to show my vintage and detachment from the field.
>> Surely, the precision of the measurements in the XL machines has been 
>> established by experimentation with the same samples.   In that case, 
>> are there not any Model E's available yet whereby a run can be made 
>> (where the temperature accuracy is known very well by precalibration) 
>> and compare that result to what is obtained in the XL machines to 
>> establish accuracy?   It sure sounds like what is otherwise being 
>> proposed is overkill.
>> Secondly, I find the discussion interesting because in my days of 
>> work with the Model E, I was always more concerned with temperature 
>> gradients in the spinning metal whereby convection could set up not 
>> only erroneous temperatures, but also erroneous transport.  We messed 
>> with aluminum flashing liners in the rotor chambers when we tried to 
>> do higher temperature work simply out of concern for the range of 
>> heating and cooling available for temperature control.   It looks 
>> like I am going to have to familiarize myself more about the 
>> temperature control system of these Non model E systems that are 
>> yielding modern day analytical ultracentrifugation results so that I 
>> can better understand your current concerns.
>>
>> I enjoy following all of the conversations coming through this list 
>> group.
>>
>> Kirk
>>
>>
>> _______________________________________________
>> RASMB mailing list
>> RASMB at rasmb-email.bbri.org
>> http://rasmb-email.bbri.org/mailman/listinfo/rasmb
>
>
> Hi Kirk,
>
> I am a bit surprised myself at how prominent people in the field who 
> obviously have very busy schedules are willing to undertake such a 
> project, but I certainly want to follow it and contribute if I can.
>
> I also wanted to pass on some experiences that make me interested in 
> this large scale trial, perhaps because the XLA is a bit more of a 
> "black box" than the old model E's.
>
> Just as I was checking an equilibrium run at lunch today and looking 
> at our data backup program when we had a power surge and the XLA lost 
> power momentarily.  It beeped, and by the time I  looked at it, it had 
> already regained power, rebooted and was spinning away with the error 
> messages waiting to be cleared.  The software had frozen up though, so 
> I had to reboot the computer, and start the XLI software (twice:  it 
> never works the first time after a reboot) and restarted the 
> absorbance scans.  Meanwhile, the temperature for some unexplained 
> reason went from 4.0, where it had been for 3 days, to 3.5 and the 
> controller started heating.  Of course, it overshot, reaching a 
> recorded temperature of 4.8 and then slowly started to fall again.  As 
> I write this, it is back to 4.2.  It makes me wonder what is really 
> happening in the cells, I expect the fluctuating temperature is really 
> just the surface of the rotor:  but it does point out how limited the 
> temperature sensing equipment is.
>
> At the centrifuge workshop this summer, some students brought data 
> with them to analyze.  In some cases, the instructors told them:  this 
> data just isn't good enough to get more information.  I have seen some 
> differences in using data from our two machines with very similar 
> samples (comparing absorbance scans from each), and it makes me 
> wonder, was the sample prep different, or (more likely) is one of the 
> machines just not collecting data as well.
>
> Writing that last sentence reminds me to change my question to a 
> request.  Please Arthur, send us enough sample to run on both machines.
>
> I am glad that you ended with how you enjoyed reading the 
> discussions.  I will not worry as much about filling people's 
> mailboxes with email "noise", though maybe I have written too much on 
> this topic.
>
> David Hayes
>
>
>
>
 From the temperature fluctuations you appear to have experienced in the 
XLA from the outlined occurences, my concerns for thermal gradients 
radially in the metal imposing thermal gradients in the contained 
samples with concommitant convection is greatly increased for 
experiments conducted on these machines as compared with the Model E.   
I think those concerns of convection could have far more serious impact 
on the results than an accuracy of the temperature.





More information about the RASMB mailing list