# Precision constant current source (1 mA) required

Discussion in 'General Electronics Chat' started by Allen_N, Dec 21, 2009.

1. ### Allen_N Thread Starter New Member

Dec 21, 2009
3
0
I'm trying to build a constant current source as part of the signal conditioning circuit for a Pt RTD temperature sensor. (I want to be able to measure temperatures, within a narrow jet of air, up to 600 deg C -- with a precision & accuracy of a few degrees.)

The RTD should get a supply current of 1 mA (typical), which will be held constant over the range of resistance developed by the RTD (as a pretty-much linear function of sensed temperature); this range is about 100 to 175 Ohm. The current must be constant to within a tolerance equal to or better than that required from the sensor itself (i.e. less than 0.5 %).

I tried building an op amp-based current source, but I couldn't get it to work. (I reckon the circuit diagram I used must have contained an error, and I'm too rusty to figure it out). I also looked at a Zener diode-based approach, but the info was too vague (too many unknowns in the equations).

In any case, I plan to power the device from a 9 V battery (possibly via a 7805 regulator -- or something more stable, if need be), so I can't use a solution that sinks much current.

Another issue is that the signal conditioning circuit itself will probably require temperature compensation so that variations in ambient temperature don't affect the gain of the signal conditioner itself (independent of the temperature of the jet in which the sensor is immersed).

Cost is not critical, so I'd be happy to go with an IC-based solution for the constant current source, but I haven't found anything with the precision and stability that I require.

I would be most grateful for any advice or insights from the Forum.

Thanks!

- Al

2. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
783
You could try something like the LM134/234/334 3-terminal adjustable current source. Data sheet is available at various web sites.

3. ### russ_hensel Distinguished Member

Jan 11, 2009
820
47
An op amp can easily do this. Google for circuits.

4. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
t_n_k's suggestion is the way I would go.
National Semiconductor's datasheet for the LM134/234/334 is here:
http://www.national.com/ds/LM/LM334.pdf
See page 7 for a zero temperature coefficient current source.

5. ### Allen_N Thread Starter New Member

Dec 21, 2009
3
0
Thanks everybody,

I already looked at the LM134/234/334 data sheet, but I got the impression that it wasn't accurate enough ('maximum set current error' of 5%). Maybe I'm not interpreting this correctly.

And, I've Googled lots of circuits. For weeks. Most are high-power, low accuracy, or not even constant-current. Google has not been my best friend on this one.

6. ### eblc1388 Senior Member

Nov 28, 2008
1,542
102
Sure you didn't.

The said current error is related to the error of the calculated resistor value for a certain constant current value, owing to manufacturing processes.

This error can be trimmed by user to match your application requirement giving you the exact constant current value required.

It does not mean in operation there will be 5% error(fluctuation) in providing a constant current when your load changes.

7. ### peranders Well-Known Member

May 21, 2007
87
0
Last edited: Dec 22, 2009

Dec 21, 2009
3
0