> -----Original Message-----
> From: piclist-bounces@mit.edu On Behalf Of Forrest W Christian
> Sent: Tuesday, December 23, 2008 11:40 PM
>
> Based on the responses (which were really not what I was expecting), I
> perhaps need to better define what I am looking for, and ask a different
> question.
>
> First of all, I should probably better define inexpensive.   Inexpensive
> in this case would be something in the low-three figure range...  Like
> around $200 or $250.  Maybe a bit more, and preferrably a bit less.
> Think about what you would pay for a good multimeter capable of 0.01%
> (or thereabouts) readings in DCV, that can be calibrated at a
> NIST-traceable lab.   Look at http://www.omega.com/pptst/HH40.html, for
> an example.

One thing to watch out for is that many (if not most) units will specify the
readout and sensor accuracy separately. This Omega unit is like that, the
ON-403-PP sensor probe is only accurate to +/-0.1 degC over 0 to 70 degC.
http://www.omega.com/pptst/ON-403-PP.html

Add that to the readout's accuracy and for $470.00 you get +/-0.115 degC
accuracy over 0 to 70 degC. They don't specify the accuracy below freezing
but, typical of thermistors is to have accuracy drop off sharply below
freezing to +/-0.4 degC at -40 degC.

Paul Hutch

>
> Second, the fact that +-0.05*C is difficult really surprises me.   After
> all from -40*C to +60*C (the range I'm really looking at), this is only
> 0.05% of full scale, or is only 1000 steps from -40*c to +60C, since
> +-0.05 corresponds to nearest 0.1C...  Is there a reason that it's so
> hard?
>
> What I am really looking for is something that is simple to use, and is
> accurate throughout it's range.  The purpose specifically being to
> verify the calibration of less accurate temperature sensors.   Or in
> other words, I really need something which I can trust means "0.5*C"
> when it says "0.5*C".
>
> -forrest

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist