References: <3DD2EC20.2FB10158@maine.rr.com> <firstname.lastname@example.org>
Subject: Re: Temperature Measurement Stability
X-Newsreader: Microsoft Outlook Express 6.00.2800.1106
X-Inktomi-Trace: pc1-port1-4-cust150.cos2.cable.ntl.com 1037306129 5556 184.108.40.206 (14 Nov 2002 20:35:29 GMT)
Date: Thu, 14 Nov 2002 20:27:24 -0000
NNTP-Posting-Date: Thu, 14 Nov 2002 20:35:30 GMT
Organization: ntl News Service
One method that I have been involved in where the requirement was for highly
accurate temperature control ( +/-0.1DegC is just about realistic) involved
dual pt100 measurements and some software to compare them and return the
Take the average, but if the two deviate too much raise an alarm.
If one is obviously wrong use the other value and raise an alarm.
"Tony Williams" wrote in message
> In article ,
> Spehro Pefhany wrote:
> > On Wed, 13 Nov 2002 19:19:44 -0500, the renowned "Raymond E. Rogers"
> > wrote:
> > >Platinum RTD
> > Yes. Particularly if they won't be exposed to temperatures outside
> > the normal ambient range, otherwise they may show some small
> > hysteresis effects.
> I believe that Bill Sloman might know something
> about sourcing precision Platinum RTDs, about
> 10x better than DIN41617.
> > What's wrong with an accurate mercury-in-glass thermometer? Not
> > expensive, and it sounds like you have a very controlled situation, so
> > the long response time and immersion depth requirement shouldn't
> > bother you.
> This is the problem.... you need some temperature
> reference that is traceable. We used to buy certified
> mercury-in-glass thermometers, narrow-range, only about
> +/- a few C around the specified temperature.
> Tony Williams.