[time-nuts] Temperature sensors and bridge amps

[Bruce Griffiths]

Poul-Henning Kamp wrote:
> In message<20101112110627.488cbc8f at vz127.worldserver.net>, "Florian E. Teply"
> writes:
>
>    
>> In a bridge circuit, you don't measure resistance directly, but use the
>> voltage that appears across the bridge. So for a 100 ohms element,
>> you'd usually have ten times the current flowing in that branch
>> compared to a 1kohm element.
>>      
> This was one of the things that I wondered about:  How large currents
> are used ?
>
> Can't be too much because that would lead to self-heating...
>
> Poul-Henning
>
>    
If the 100 ohm and and 1k sensor elements have the same thermal 
resistance then having 10x the current in a 100ohm element will increase 
its self heating by a factor of 10 over the 1k element.

Its usually better to use an AC bridge (to minimise the effect of 
thermoemfs) or equivalent technique with lower sensor dissipation.
There are examples of such bridges using platinum RTDs to achieve 50uK 
or better stability in the literature (about 40-50 years ago).

If the sensor element has a thermal resistance of say 100K/W and a 
stability of 1mK is desired, any self heating should be stable to better 
than 1mK. Thus limiting the sensor self heating to perhaps 10mK or less 
is perhaps advisable.
If the sensor thermal resistance is 100K/W then the the sensor 
dissipation be 100uW or less.
With a 100 ohm sensor element the corresponding sensor current is 1mA or 
less.
The sensor voltage then varies by about 385uV/K for a 100 ohm platinum RTD
With a 1K platinum RTD dissipating the same power the sensor voltage 
tempco is 3.16x larger.

One can also switch the bridge source polarity to achieve a similar 
effect to that of an AC bridge, some high resolution ADC chips have this 
capability (albeit using external excitation switches) built in.

Bruce