Mark Willis wrote: > You can make a current limiter out of an LM317 with a resistor across > the "Output" and "Reference" terminals, Limitation here is the combined minimum input-output differential *plus* the 1.25V reference voltage, means minimum drop is about 3.25V or so. I suspect the CRDs may have a lower minimum drop? And as mentioned, the minimum current in this mode is set by the reference current which for an LM317 is 50 to 100 ľA, plus the swamping current which is usually at least 1 mA. You *can* therefore use an LM317 for as little as 1 mA with some limitation on accuracy. I'm sure the LM317 has a far higher power dissipation than the CRDs. > Nice discrete, it makes GOOD sense to use those for LED's, wish I had > some here! Great for a number of applications. Cost has been mentioned. That slows down the enthusiasm. We can't all visit Aki. :-(( > Car tail light LED units, for one. That is *definitely* a high-power application. I'd look to the LM317 if you want simplicity, but the traditional 3-terminal circuits can achieve a 1V or so minimum drop which is difficult to beat. > Related question; Wonder what happens when one parallels these > devices? I'd think the "conductances" would add, 2 paralleled > 20mA units should let about 40mA flow, but I haven't tested it yet. It *must* work fine - by definition. -- Cheers, Paul B.