Re: [APML] Thermal Time Constants

[Aplanatic@aol.com]

Hi Chuck:

>When you're talking time constant do you mean the >temperature reaching 63% of it final value, like an RC >time constant where it takes about 5 timeconstants to >reach the final voltage?

Yes, the thermal time constant is defined as the time it takes for the temperature delta to reduce by 1/e (about 37%, or as you say, 63% toward its final value).  As an example, suppose the the starting delta T is 10 degrees.  After one time constant the delta T will be 3.7 degrees, after two time constants it will be 1.4 degrees, after three it will be 0.5 degrees, etc.

The Foucault test shows that thermal currents (in this particular test environment) become almost unnoticable when the delta T between the mirror and ambient is less than about 0.3 degrees C.  Where one defines the thermal effects to be small is a matter of interpretation, of course.

In practice, if the ambient temperature is changing then the mirror will lag behind ambient by the rate of change of ambient temperature times the thermal time constant.  This can have a large effect when ambient is changing fast (like several degrees C per hour) and the time constant is long.  I see this effect when trying to focus my Schmdit Cass early in the evening.  I see lots of tube currents and bad "local" seeing until ambient stops changing rapidly (around midnight, typically).  This is true even though I run a fan after sunset for an hour to try to bring the OTA into equilibrium.  I don't run a fan while exposing, but maybe I should.  I think Tom Krajci has implemented a method of coupling a fan to the OTA without causing major vibration problems.  This also explains something that puzzled me for a while.  Shots taken in the early evening are almost always a bit "soft", and my best shots (in terms of resolution) are often taken early in the morning.

BTW, in my scope, the local seeing (tube currents) can be really huge.  I've seen as much as 5 arcseconds (maybe more) local seeing effects until the ambient temperature stabilizes.  In part, this has been the motivation for performing the bench experiments.  I really didn't believe that the time constants were so long (for the no-fan case) until I did it myself.

Dave Rowe


--  APML Archives at <http://www.system.missouri.edu/apml/>  ---
                Unsubscribe at <majordomo@seds.org>