[lowcrawler]

How do I check how much capacity my cells have?

I finally drove mine far enough in one burst to get down to the bottom, which using my miniBMS I was able to find out which is my 'weakest' cell.

Now I'd like to test HOW much weaker it is than the rest, and potentially replace if if it is, indeed, considerably weaker.

... but I have no idea how to do that.

[major]

Originally Posted by lowcrawler

How do I check how much capacity my cells have?

I finally drove mine far enough in one burst to get down to the bottom, which using my miniBMS I was able to find out which is my 'weakest' cell.

Now I'd like to test HOW much weaker it is than the rest, and potentially replace if if it is, indeed, considerably weaker.

... but I have no idea how to do that.

I would suggest that you get a CellLog8s and connect it to several cells in series including your cell of interest and "log" a discharge (and charge) cycle. You can then compare cells on graphs like these http://www.diyelectriccar.com/forums...ight=celllog8s You won't know the actual current from a test like that so you cannot calculate the Ah.

To find the actual cell capacity (AH), you need to know the discharge current. For the tests I do, I hold the current constant by using a regulated load in the lab (a big resistor and SS regulator).

Here is another way to do it. http://www.westmountainradio.com/pro...avcode=/UPSCBA

[ricklearned]

If you want graphs that show you volts and Amp hours, there is the West Mountain Radio CBA for about $150 or the Revolectric will do an individual cell up to an 8 cell pack for about $250.

For less money I think you can get some data out of certain cellogs and some other RC Watt meters but using the logview software is not real time but a lot cheaper.

[lowcrawler]

I appreciate the link, but I'm not really looking for a $250 dollar solution to test if a $100 cell is in need of replacement. :-/

Isn't there an easy and cheap way to see how much capacity a cell has compared to another? That's really all I'm looking for... I don't need automated logging, graphs, etc. I'll probably do the test once, replace if needed, and never do it again for the life of the pack.

[Ziggythewiz]

Depending on your setup, if you're currently drained you could remove the weak cell, and keep discharging through an AH counter till the next triggers the cutoff, etc. Or if the charger is adjustable, you can do similar charging to the top.

[dougingraham]

Originally Posted by lowcrawler

I appreciate the link, but I'm not really looking for a $250 dollar solution to test if a $100 cell is in need of replacement. :-/

Isn't there an easy and cheap way to see how much capacity a cell has compared to another? That's really all I'm looking for... I don't need automated logging, graphs, etc. I'll probably do the test once, replace if needed, and never do it again for the life of the pack.

The cheap way requires one or two DVM's so you can measure the voltage and current. And you need some sort of load that will support quite a lot of power. If testing a single 100AH cell you want approximately a 30 amp load. You can use a bank of automotive tail light bulbs as the load. You may need to experiment to see how many bulbs it takes to get 30 amps. And you need a clock.

Start with a fully charged cell. Connect one DVM to the terminals and this will be your volt reading. Connect a calibrated shunt (there might be one of these in your car now) between the battery and the load. It doesn't matter if this is on the positive or negative lead. For convention lets put it in the negative cable between the battery and the load. When you are ready connect the positive cable to the load. If you are using light bulbs they might glow some. Jot down the time and the voltage and the voltage across the shunt. Do this every minute until the voltage reaches 2.5. Disconnect the load and recharge the battery so you can put it back in the car in the same fully charged state as the other cells.

Average the current reading voltages (or just use the current at the halfway point as it will be close enough) and then calculate the current based on this average. Voltage / shunt resistance will give the average current. Compute the time in hours. It will be minutes until the voltage dropped to 2.5V divided by 60. If you were using 30 amps the expected time for a 100AH cell would be around 200 minutes. 200/60=3.33 hours. Multiply the time in hours by the average current and you will have the AH capacity you measured. In this case 3.33 hours times 30 amps = 99.9AH.

You might need a bank of 20-30 bulbs to get 30 amps which would make that the most expensive part of the deal. It doesnt have to be exactly 30 amps, but if too low you could get bored and forget the battery and have it go flat making the whole thing a moot point because then you would have to replace it.

A watch, a volt meter, a shunt, and a load is all you need. You can enter the data into a google docs spreadsheet and make a nice discharge graph if you took down the numbers.

Pick another cell you think is a good one and test that too so you have something to compare against.

An alternative to all of this is just replace the cell you think is weak and keep it as a spare for when you have a real failure. Charge it so the resting voltage is a little over 3.30 volts for long term storage.

[eva-michael]

If it's not more than 60V nominal 140A rated. You can directly use a GT power capacity meter.

[GerhardRP]

Originally Posted by lowcrawler

I appreciate the link, but I'm not really looking for a $250 dollar solution to test if a $100 cell is in need of replacement. :-/

Isn't there an easy and cheap way to see how much capacity a cell has compared to another? That's really all I'm looking for... I don't need automated logging, graphs, etc. I'll probably do the test once, replace if needed, and never do it again for the life of the pack.

First, find a data sheet for your cells and find the resting voltage vs. state of charge. Then, after you have run your weakest cell to its low limit, accurately measure the resting voltages of each cell. Finally look up the SOC of each cell from the data sheet. Cost = good DVM.
Gerhard