REF200 current source

Sharon;

As long as the voltage does not exceed the REF200 compliance voltage, there will be almost no effect whatsoever.

I built a test circuit for the REF200 (shown below) and when I connected Vs+ and Vs- to +/-2.5v I got 64uA at the output.

Raising the voltage sources to +/-5v the current was reduced to 54uA.

Is it possible to get an accurate 50uA?

Sharon;

I am assuming that you are measuring the output current to your +/-2.5V common and using a low-resistance microammeter. If the current meter has a significant voltage drop, this can cause an error due to the compliance voltage being reduced by that drop. Your getting an output closer to 50uA at +/-5V suggests that compliance voltage is probably the problem.

To verify this, can you use a +/-15V supply and measure the output current to see if it is 50uA?

The circuit of Figure 4a in the REF200 data sheet may be another possibility but the output current is in the opposite direction.

I measured the current from Iout to GND using a Fluke multimeter.

I didn't quite get the comliance voltage thing. Can you elaborate on that?

I can try with +/-15V just to verify but my circuit has only 2.5v and 5v.

Circuit 4a gives 200uA and I need 50uA.

Sorry, Sharon-- that should have been 5a.

"Compliance" is how far the output can swing toward it's supply voltage. Try measuring the voltage across each of the 100uA current sources and the mirror when the output is tied directly to ground. The REF200 spec shows that it can operate down to 2.5V across the current sources; the mirror can operate with a few hundred millivolts.

Check your circuit with +/-15V as a temporary diagnostic to see if the problem is compliance related. Have you tried more than one REF200 in your circuit?

Do you mean to short Iout to GND?

5a circuits are not good for me since I need a current source and not sink.

Sharon;

Yes, short the Iout pin to ground-- this is a current source so this will not damage it.; it simply puts 50uA into ground.

Why the posts on the REF200, a precision 100uA current reference have been moved to this POWER forum is a mystery-- isn't is more properly a precision device?