V G wrote:
> Thanks for the tips. I made the circuit:
> http://postimage.org/image/2emnxdfus/full/

That looks pretty good, except for R3.  Just have the opamp output drive th=
e
base of Q2 directly as I said originally.

You can tie the collector of Q2 to 5V instead of the collector of Q1.  That
will make this current sink work down to lower voltages.

I don't know where the 2N2222 came from, but that's a bad idea.  I'm pretty
sure I said to use a transistor in something like a TO-220 case, like a
TIP-41 or many others.  The 2N2222 could get fried at high currents.  At
V2=3D3.7V and 2A, it could be dissipating up to 600mA.

> What is the purpose of the 10pF capacitor?

To keep the control system stable.  I didn't look up LTC1050, so I don't
know at what gain it's supposed to be stable.  In any case, there is going
to be some lag from the output of U1 until that shows up on its negative
input.  Such a lag can cause instability.  C1 makes the controller more
stable, but also slower responding.  You're not going to get any fast
transients, so slowing down the response in favor of stability makes sense.

You can try it without C1 and see what happens.  If you're just looking at
the result with a meter, you might not notice the instability.  However,
look at the output of U1 with a scope to make sure itsn't not oscillating.
I was actually expecting C1 to be a bit higher.  I originally said a few 10=
s
of pF, not just 10 pF.  That's only a little above stray capacitance.  I
probably would have started at 22pF and then gone up from there to guarante=
e
stability.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.