Simple current supply for thermistors?

Jun 22, 2005 25 Replies
R

Here is the OP, courtesy of the OP. For some odd reason, it is showing up as the third message in the thread, at least when read at Google Groups:

I'm looking for a basic, easy-to-build constant current source to drive thermistors, nominally at 0.1 mA. The thermistors would be the common

10 k-ohm @ 25 C variety.

I spent time Googling this topic without success. One site mentioned that an LM317 can be used to generate a current, but the current would have to be a minimum of 5 or 10 mA. I need a 0.1 mA source.

I'd like to build something, preferably for under $10.00, with better than 0.5% accuracy (that should correspond to temperature readings within 0.1 or 0.2 C). My plan is to have several thermistors in series driven with the circuit, and then I can read off the voltage of each thermistor with an A/D hooked up to a PC, and be able to monitor temperature readings over several hours or up to a day.

TIA,

Mark

Vote
Twitter Facebook LinkedIn
R

I'm currently considering 3 options, based on what I've read here:

  1. Resistor divider network, with 1 fixed, low temperature coefficient resistor plus up to three thermistors, all in series. Ratio of each thermistor voltage to fixed-resistor voltage gives accurate thermistor resistance (where fixed resistance is known), so a stable current or voltage source is not required.

  1. The LM134/234, probably using the "zero temp coefficient" configuration.

  2. Off-the-shelf temperature sensor.

Thanks,

Mark

Vote
Twitter Facebook LinkedIn
E

Sheesh! Usually, I hate it when I do that. But in this case, I'm glad. I had dismissed the scheme based on the OP's specs and my error. They are *way* less stringent than I figured.

He can buy a chip that will provide the .1 mA current and have at it: REF200AU-ND from Digikey for $5.20.

No more. Thanks John!

Ed

Vote
Twitter Facebook LinkedIn
C

___

Couldn' t find the original post, but the OP should check out the Dalis DS18B20. It's an addressable 1-wire 12-bit serial device with an accuracy of +/- 0.5 deg. C, and a readability of 0.0625 deg. C. Maxim-Dalis has it for $2.57 one-off, but the lead time is 7 weeks.

Vote
Twitter Facebook LinkedIn
R

Wow. Thanks!

Mark (a.k.a. the OP)

Vote
Twitter Facebook LinkedIn
B

One trivial way to do this is using a TL431.

The TL431 uses almost no current in through the adj terminal, so by putting a 25k resistor from adj to ground, and putting the thermistor in the same branch, you can measure the thermistor resistance by noting the voltage at the cathode, as below:

5V | .-. | | 1k or so | | '-' | .----o------ output V = 2.5 + Rt*100u .---o-. | | | | Thermistor | | | '---o-' | | | | | | - o--- ^ TL431 | | | | .-. | 25k | | | | | | '-' | | | | | | | '----' GND

(created by AACircuit v1.28.5 beta 02/06/05

formatting link

Slightly cheaper than a current source.

Regards, Bob Monsen If a little knowledge is dangerous, where is the man who has so much as to be out of danger? Thomas Henry Huxley, 1877
Vote
Twitter Facebook LinkedIn

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required