| Re: IDL Math Expert, clarification [message #9794] |
Thu, 21 August 1997 00:00 |
Brad Gom
Messages: 49
Registered: August 1997
|
Member |
|
|
Ok, thanks to all of you who have helped so far...
if anyone is still interrested, I'll define the problem a little more
clearly.
The data corresponds to a V-I curve of a cryogenic bolometer
(temperature sensitive resistor)
By measuring the voltage across the detector as a function of current
flowing across it, you can get information about its power loading.
From bolometer theory comes the power dissipated by the element:
P=G*(T*(T/t'-1))
where G is the thermal conductance, T is the bolometer temperature and
t' is the bath temperature. (other terms have been simplified)
The bolometer is a semiconducter, thus its theoretical resistance is:
R=r'*exp((tg/T)^.5)
where r' and tg are material dependant parameters, and T is the
bolometer temperature.
Now Ohms law gives V=sqrt(P*R) and I=sqrt(P/R)
T, the bolometer temperature, is a hidden parameter.
My problem is that I have a set of V vs. I data, but my theoretical
initial estimates of the parameters G,t',r',and tg are not very
trustworthy. Manipulating the 4 parameters by hand is too tedious. What
I need is a procedure that can fit a curve defined by two parametric
equations.
I have received a couple of good suggestions that I am followiong up on,
but if anyone has another solution please feel free...
Thanx,
Brad Gom
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
