see http://www.instructables.com/id/DIY-Constant-curre...

watch the video, he shows the circuit schematic at about 1:44.

Alternately, a current mirror or any number of constant current sources:
https://en.wikipedia.org/wiki/Current_source

TTFN
I can do absolutely anything. I'm an expert!
FAQ731-376: Eng-Tips.com Forum Policies forum1529: Translation Assistance for Engineers

You should be able to use three resistors. A voltage divider to set the voltage and then the third resistor to a separate ground.

+----res----res-----~
|
Res
|
Ground

Not sure how a voltage divider will keep constant current from the varying battery voltage?

Brad, if you spend a little more, you can use a simpler circuit, see http://www.ti.com/lit/ds/symlink/lm317.pdf figure 14. These have a minimum load current of a few milliamps, but you should be targeting current draws above that level I think?

A simple constant current can be made with a 316 Voltage regulator.
The regulator will hold the Voltage between output and reference to 1.2 Volts so all you need to do is calculate the resistor you need to drop 1.2V at the current you want.

e.g. 50 mA 1.2/0.05 = 24 Ohms

Did you mean LM317?
LM316 is an OpAmp.

A LT3092 is a nice solution I recently used. One IC and two resistors is all you need.

Z

I did a test on many different makes of 9 V batteries once. I was running them simultaneously and the data I was interested in was for rather low discharge currents. So, I used 10 kohm resistors and recorded voltage from start to en of test. I then put the data (.txt file) in an Excel sheet and calculated the total charge from voltage, resistance and time. No big deal and real simple if you want to test many batteries at the same time. Cost for the set up: around 1 USD.

Gunnar Englund
www.gke.org
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.