Ni-MH variable current supply charger

Yes, get a decent 'intelligent' charger ready made unless you want to build one yourself for the educational side of it. If you want to DIY then get a chip that provides the 'intelligence'.

Maha make quite good chargers (among others), see:-

When you look at the spec for this one from La Crosse, it would be hard to beat as a DIY project.

Rob

LM317 voltage regulator IC set up for constant current.

Googling on "LM317 constant current" produced this:

... but it won't do nearly such a good job (as in treating the cells well so they last well) as a proper NiCd/NiMh battery charging chip.

You can buy the chips or beg a couple free off the manufacturer as I did, problem is I can't rememeber the manufacturer's name now.

No need, the circuit is in the data sheet where you'll also find useful info like package choices and thermal data.

MBQ

But a fast charger only costs a few quid.

You canhave my old one f you like. Ex model car racing charger. Does 4-8 cells at about 3A fast, dropping back to trickle when the delta peak is detected. Runs off 12v or the mains. Can';t sell em for love nor money.

Thanks but the problem with the 317T and also the L200 is R=1.25/I Fine for a fixed current but hard to achieve a variable one.

Thanks but I don't want or need a quick/fast/Gadget lovers DREAM charger. also what I failed to say was I will also be using this for AAA,D,C,and subC cells.

Due to the low value R not available as a variable one? How many different currents do you want? Just switch in different resistors.

Alternatively use a simple transistor as a current source.

MBQ

Yes that's how I do it now with fixed value resistors But I would like to build a completely variable supply as per the original post !

Great can you point me to a circuit that works and I can build. Electronics is just a diy hobby I am not a design engineer.

Google "transistor as current source" gives the first hit as

Basically you control the base voltage and assume the Vbe to be 0.6 V which gives you the voltage across the emitter resistor. That sets the current through the emitter resistor. The same current (to a first approximation) flows through the load.

MBQ

You just know that one is going to be copied and be available from all good local Grocers & DIY stores before long ...

Next addition will be some Flash memory and a USB interface, so you can view the charge rate and the condition of the cells graphically.

And now that I've mentioned it, the odds of anyone placing a restrictive patent on the idea should be zero.

Maxim ?

Yes, that's the one, you can (or could) ask for sample chips on their web site. I got a couple of their deltaV charger chips this way, along with loads of documentation.