Will this work as a DIY a 2S lipo charger?

[djQUAN]

Diodes and zeners won't work since they have a very wide accuracy tolerance typically 5%. You'd need better than 1% for lithium ion CV termination voltage. Plus, diode / zener voltage drop varies with temperature, current and phase of the moon.

 

[Sigurthr]

Plus, diode / zener voltage drop varies with temperature, current and phase of the moon.

LOL, QFT! You left out how the angle of your tongue also affects it. :na:

 

[rhd]

Diodes and zeners won't work since they have a very wide accuracy tolerance typically 5%. You'd need better than 1% for lithium ion CV termination voltage. Plus, diode / zener voltage drop varies with temperature, current and phase of the moon.

I wasn't considering that route - it didn't make any sense to me. Somewhere in the equation there has to be an IC with some brains - even if that's just a linear regulator.

 

[djQUAN]

I wasn't considering that route - it didn't make any sense to me. Somewhere in the equation there has to be an IC with some brains - even if that's just a linear regulator.

Yes, sorry, that was for The Lightning Stalker since he mentioned using either a string of diodes or zeners.

 

[rhd]

Any thoughts about the viability of this?

Since this is a little bit outside of my usual wheelhouse, I've held off on laying out a board for the time being (hoping that a second set of eyes can glance at this and give me a yay or nay).

 

[djQUAN]

You'd just need a current limited source for the input, also, you'd need to disconnect the pack after charging.

And one more thing, the op amp may not be able to source/sink enough current to balance mismatched cells in bad packs but in a well matched pack, it should be no problem.

Other than that, it will probably work.

 

[rhd]

You'd just need a current limited source for the input, also, you'd need to disconnect the pack after charging.

And one more thing, the op amp may not be able to source/sink enough current to balance mismatched cells in bad packs but in a well matched pack, it should be no problem.

Other than that, it will probably work.

The OP AMP, is capable of 3A - I chose one that was very high current.

Since I have enough voltage overhead, would it be sufficient to nest everything above inside a current regulating setup of another linear regulator? Ie another 1085?

 

[djQUAN]

Ah I didn't check the op amp datasheet. If so, that'd be plenty.

How big are the cells that you plan to charge with this thing?

Yes, with enough voltage headroom, a constant current regulator before the voltage regulator will work to make a CC-CV supply.

 

[rhd]

Ah I didn't check the op amp datasheet. If so, that'd be plenty.

How big are the cells that you plan to charge with this thing?

Yes, with enough voltage headroom, a constant current regulator before the voltage regulator will work to make a CC-CV supply.

Around 100 mAh

 

[The Lightning Stalker]

Very clever. I would put a low value resistor on the output of the amp just in case, maybe 1.5-2Ω.

 

[Sigurthr]

With djQ's inputs I'd say it looks like it should work fine.

If you add series resistance mind there will be a voltage drop, depending on expected current this will or won't be an issue. I'd just go with a front end CC regulator, personally.

 

[Cyparagon]

Looks good - but it may oscillate if you don't stabilize it somehow. I would put a cap from pin 1 to ground, myself.

 

[rhd]

With djQ's inputs I'd say it looks like it should work fine.

If you add series resistance mind there will be a voltage drop, depending on expected current this will or won't be an issue. I'd just go with a front end CC regulator, personally.

Looks good - but it may oscillate if you don't stabilize it somehow. I would put a cap from pin 1 to ground, myself.

With this advice in hand, I've revised the schematic as follows:

Conceptually it bothers me to have a current regulator that I know will (in all likelihood) never actually be in regulation. I can't articulate a reason that this would be a problem though.

 

[djQUAN]

Below the current limit threshold, the current regulator simply conducts to max (lowest voltage drop). Voltage drop increases to regulate current when the threshold is reached.

 

[Illuminum3415]

um.... if you really must go cascading regulators why not consider using a charge controller for the second regulator as opposed to a linear CV?

http://www.ti.com/lit/ds/symlink/lm3420.pdf