[time-nuts] What do people use for measuring temperature?

[Dan Kemppainen]

Hi,

Darn, Bob beat me to it! I was going to suggest the AD590 and a suitable 
ADC. The ceramic part is relatively small.
https://www.digikey.com/product-detail/en/analog-devices-inc/AD590JF/AD590JF-ND/611802
It's not as small as a thermistor, but not overly large. They have other 
packages as well.

The smallest device is probably still a TC (40 gauge TC wire is small!) 
But honestly getting down below .1 Deg C can be hard to do. (If the 
icepoint reference doesn't have the resolution you'll get jumps in the 
curve as ambient changes.)

For thermistors, at least a ways back GE Thermometrics sold glass 
packaged parts with calibration coefficients. They weren't really cheap, 
but were small and accurate enough.

Honestly, if it were an experimental setup and not production I'd just 
grab some glass packaged parts from digikey (looks like a lot of options 
around a few $ ea.) . Bias it with a resistor from a fixed supply, then 
run a three point calibration (Icepoint, boiling point, and somewhere in 
between) and generate calibration coefficients. Pair that with a 
inexpensive-ish 24bit ADC, and you'll have all the resolution you want. 
You should get sub .05 deg accuracy, at least short term. Skip the 
calibration and just use the published B value, and you'll still have 
the resolution.

Let us know what you settle on.

Dan



On 9/26/2020 9:45 PM, time-nuts-request at lists.febo.com wrote:
> Date: Sat, 26 Sep 2020 11:23:45 -0400
> From: Bob kb8tq<kb8tq at n1k.org>
> To: Discussion of precise time and frequency measurement
> 	<time-nuts at lists.febo.com>
> Subject: Re: [time-nuts] What do people use for measuring temperature?
> Message-ID:<1E3256CF-83C4-4932-8621-27597E0D6DB1 at n1k.org>
> Content-Type: text/plain;	charset=utf-8
> 
> Hi
> 
> One interesting ?oops? using RTD?s:
> 
> They are close cousins of strain gauges. Some RTD designs are*very*  close.
> Mount them to this or that and they may tell you more about the stress / strain
> in the mount than about the temperature.
> 
> You do*not*  want to know how many (hundred) temperature test chamber sensors
> that particular goof messed up ?? 4 sensors per chamber, 8 in a pod, three pods
> in the factory here, four pods in the factory down south, Two partials in engineering,
> two partials in QA ?.
> 
> The answer ultimately was to tear out all the RTD?s and replace them with carefully
> tested (= sorted) AD590?s.
> 
> Bob