Voltage comparator for lipo batteries

Hello all. Looking to get some insight on how to compare the differences in voltages for charging some batteries. I had in mind to use a comparator, but most examples use a reference voltage set by some sort of resistor divider network, and I don't think that will work in this case.

So, i thought I would provide some detail on what I know, before getting to my question. I use lipo batteries in my r/c car. The ones I use are a two cell model rated at 7.4v. That is the nominal voltage.. as the individual cells can range in charge from 3.0 volts (empty) and 4.2 volts (full). That puts the pack at 6.0 volts to 8.4 volts. If you let the voltages get beyond those ranges bad things happen.

Here is a basic pack:


When you charge these batteries you apply 8.4 volts across the back, and attach a voltage monitor that watches the individual cells. It keeps them from exceeding 4.2 volts, as otherwise one cell could be 3.4 volts and the other 3.0 (bad).


Now, I would like to be able to charge two batteries in parallel. This is generally a fine practice, you keep the same voltage and just double the current. The downside is, when wired as noted below if the packs (or cells) are different states of charge in the 3.0-4.2 volt range you can get a large surge of current as they balance with their parallel cells in the neighboring pack


So, what I (think) i want to do is insert a wire-would resistor in the .5ohm range between the cells initially to keep the current flow between backs < 5Amps or so. Once the cells get balanced I want to kick in a relay to bypass the resistors for the duration of the charging process. This is what I had in mind for the final product.


So, that brings me to my questions:

• Should I use two resistors as depicted, or just one for the whole pack.
• How to detect when the packs are near 0 volt differential to know when to engage the relay

I had though of using a comparator and some sort of zero crossing detection.. but I couldn't get my head around that. I have never used logic components, but look forward to trying them out. Any thoughts on a simple method to detect when the cells are withing .2 volts or so of each other and how to keep the relay 'latched' until the cycle is over?

Thanks in advance,
Rob

Fixed inline links above. Went from referencing images in my picasa album to using attachments.
---
Seem to be striking out on in-line image links. They work in preview, but not when posted. They are reverting from picassa to a local image cache..

Would a current limiting circuit like this work?

**broken link removed**

I am guessing not as the voltage drop in rsense would affect the end charge of the second pack?

I think the Lipo batteries are the Lithium polymer batteries. I am currently working on lithium batteeries and i think you should not charge them like this. you could have two seperate connections for charging see the simple schmatics below.

**broken link removed**

**broken link removed**
See which link works. donot attempt to do it without a regulator or in parallel or series. this could result in explosion due to abuse of charging conditions. the SC 824 is a IEEE standard. and you can search other ics that charge 2 cells in parallel. but i recommend not to do so..

Roltex,

Thanks for taking the time to respond. Your assumption was correct that lipo was short fo lithium polymer, I should have clarified that. I think we may be working on different scales here, the packs noted are 5000mah and have charge rates of 5amps. The SC824 maxes out at 1A. Regarding the parallel charging, lots of RC packs are made of combination's of series and parallel cells to reach the targeted voltage and mah ratings. Please correct me if i am wrong, but the important aspects of charging are to limit the current to a reasonable rate (1C) and prevent over-voltage. I think my proposal accomplishes both objectives, as long as I prevent the large current surge upon initial connection of two packs at different voltage levels.

So, please correct me if my assumptions are incorrect, I am just a hobbiest in this arena. Still looking for a way to identify when the voltage differences between the packs reaches zero.

Thanks,
Rob