Cell balancing? Any idea?

[Wicks]

OK lets see what we can put together.

Here is a great demo pic of what's going on:

POSITIVE is your battery voltage.

RL1 is the leakage resistance of the first cap.

RL2 is the leakage resistance of the second cap, and so on...

RB1 is the first balancing resistor that you would add.

RB2 is the second balancing resistor that you would add, and so on...

The Maxwell spec gives a ballpark maximum leakage current of about 5mA.

With an applied voltage to each cap of about 2.4V, that gives 2.4/5mA = ~480 Ohms which is technically internally across each cap.

RB1 is in parallel with RL1 so for the voltage drop across RB1 to take over, the resistance of RB1 needs to be lower than RL1.

In the same manner, the current through RB1 needs to be higher than the leakage current through the cap (~5mA).

I like to think of it this way: If two resistors are in parallel with R1 being 5 Ohms and R2 being 1M Ohm, then R2 will barely affect the circuit and almost look like it wasn't there. Take R2 to 100M Ohms and it is so close to holding your ohmmeter probes in the air that it wont affect the circuit.

In fact the resulting resistance of the two (5||100M) is 4.99999999999 = No affect.

Now if you go the other way and lower R2 to 1 Ohm then it is the dominant resistor.

(5||1) = 0.8333 Ohms.

Back to the topic.

Here's a good article:

http://www.nmr.mgh.harvard.edu/~reese/electrolytics/

Scroll down to the section on series connected capacitors.

They state that the balancing current should be 10x the leakage current.

If so then the current through RB1 must be ~50mA or higher.

That gives a resistance of 2.4/0.05 = ~48 Ohms.

Power in that resistor would be 0.05^2*48 = 120mW which isn't too bad.

Your total parasitic draw would be 50mA + 5mA (cap leakage) = ~55mA.

With a ~100AH battery at a draw of 55mA you'd have a capacity of ~1800hrs (not counting other draws on the battery).

Not sure if you saw it, but here's a paper from Maxwell themselves:

http://www.maxwell.com/products/ultracapacitors/docs/an-002_cell_balancing.pdf

In it they state that balancing with a 10:1 ratio could take hours so you may need a 100:1 ratio.

The higher you go, the better the balancing will be at the expense of higher parasitic draw.

Ultimately this is all theoretical and up to you how you want to implement.

If it were me and my equipment, I'd contact Maxwell directly and give them your scenario and have their tech experts give you a recommendation.

Just tell them you dropped "$XXX" on a bunch of their 3000F caps and would like a recommendation on how to implement them in you car.

Informing companies like this that you bought a bunch of their product always helps get a better response then somebody just asking a question.

Keep me posted on where you go. I'd be interested in what they say or what direction you take.

[SnowDrifter]

Wicks... You are AMAZING. Like.... Seriously.... That was extremely helpful. I appreciate it

I called up Maxwell today and they suggested a 26 ohm resistor as per their 10:1 ratio. Radioshack didn't have any 25 ohm ones and they didn't have enough 50 ohm resistors to parallel them. So for now I'm using 50 ohms for a 5:1 ratio as my starting point. I'll be checking each cell with a multi meter periodically and adjust as needed.

[Raptorman]

My brain is full of fuck just attempting to read this shit.

[SnowDrifter]

My brain is full of fuck just attempting to read this shit.

It's not too hard to understand. You just need to take a different reading style - typical skimming doesn't work. You need to read it line by line and only go on to the next once you understand it. If you skip something or skim, you will get lost. It's just a matter of looking at the whole picture in detail

[SnowDrifter]

Quick update on this: The 50 ohm resistors seem to do their job. The difference between the caps with the highest and lowest reading is .04v.

So... Mission success

I'm going to go outside and measure the current draw of the bank in an hour or so. I put my car on a charger and I'm giving it a couple hours to settle and equalize so I get a decent reading

[Wicks]

Awesome news!

That would be interesting to compare the current from the caps vs the Alt/batteries when the amp(s) hit.

Although I don't think a normal clamp meter is fast enough to give you a real answer.

[SnowDrifter]

Awesome news!

That would be interesting to compare the current from the caps vs the Alt/batteries when the amp(s) hit.

Although I don't think a normal clamp meter is fast enough to give you a real answer.

I seem to recall something about D'Amore coming out with one of their LED style shunt meters. But I can't find it on their website.

And I measured parasitic draw with the car off. Came back with .06a, which is about what I expected. Nothing big. Enough to make my resting voltage go from 15.15 down to about 14.9 after sitting for a day and a half.

Given how much time my car spends on the charger though I'm not concerned. It's just something to keep in mind if the car should sit for a week or more

[Wicks]

Yep, the AM-1. I saw it also but yeah, it's disappeared now.

I was thinking about getting one myself.

Would be handy to have an accurate view of the current draw.

Sounds like a small price to pay (parasitic draw) vs the benefit from the caps, that's for sure.

Especially if you're charging your batteries like you said anyway then it's no trouble.

As long as it doesn't sit too long.

Tried thinking about a way to avoid that but there isn't a clean solution.

You could disconnect the cap bank by a large solenoid, relay or switch but the extra resistance and whatnot would just take away from the caps.

Hmmm, one solution............more batteries and/or caps!

[SnowDrifter]

Haha

It's not even a big deal. Even if my car sits for a couple days it's not that much of a discharge from the battery. It's less than it was when I pulled it out of the box from sitting in storage lol. I'm not concerned about it unless the car sits for 5+ days

It's funny.. The caps will hold their voltage for about 45mins to an hour before they drop to 15v after I'm done driving the car. So even if I make a short trip, that backfeed into the battery helps to make sure it's as topped off as much as it will be without a charger.