How to Load Test a lifepo4 pack

[dumbass]

OK, I stopped at Habor Frieght today and picked up a battery tester. It's a 500a 12v tester. I run twin 24v 20ah lifepo4 packs. I am currently not running any kink of on bike meter. I only have the throttle lights but I can tell a major difference betweek the 2 packs. On pack #1 the red throttle light comes on almost every time under full throttle and the acceleratrion feels wimpy. Whereas. on pack #2 I never see the red throttle light and acceleration is great.

So I bought the battery tester to check for a weak cell under load. Please note I am not running a BMS or LVC. When bulk charging with a 12v SLA smart charger all cells on pack #1 come up to 3.45 to 3.39. To me this is a good thing and an indication of a fair balance. So now I want to test the pack 12v at a time. So what should I be looking for and what load amperage should I use? The tester is adjustable. Please remember they are 20ah cells and rated at 2c. I was thinking of testing at 40a per 12v pack. Is this correct and what should I expect to see?

http://www.harborfreight.com/500-amp-carbon-pile-load-tester-91129.html

Bob

 

[amberwolf]

You could test even at lower currents, and probably still see voltage drop on the problem cell groups.

Most likely you have one or more cell groups in the problem pack that either have dead cells or low capacity cells, that aren't charging properly, *or* if they are cylindrical cells some of the tab welds maybe broke, which will take whole cells out of the equation, greatly lessening the capacity of the group they were in.

I'd measure the voltage on each cell group while under a nominal load of like 1-2A first, just measuring with a voltmeter across the + and - of each group. IF you have balance connectors (even if not using htem) on the packs, then you can measure at those connectors. Just meausre from ground to the first cell-group connect point, then from the first cell-group connect point to the second one. Then from the seconed to the third, etc.

Note them all down and the ones that vary significantly downward are the problem groups, which you can investigate further.

If it'sa cylindrical-cell pack, you can look at my Vpower/CammyCC rebuild thread for a bit of pack construction detail, before opening up your own.
http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=22750&start=0

If they're pouch cells, then probably they're made like Pings, and there are a few threads on those.

 

[docnjoj]

I get about 4 amp load @ 36 volts using an old bun warmer tray although many suggest a hair dryer for a load. Works well enough using a VOM for each cell (pouch group) in my Pings.
otherDoc

 

[dumbass]

I tried a 115v 1200w iron last week I guess it would be about 260w at 25v. I checked each cell while under this load but I didn't know what I was looking for. I should have recorded each cell but I didn't. I will redue this test and then a 40a test and record the results. Hopefully someone can tell me what it meand.

BTW, These are Thunder Sky prismatic 20ah cells. They do not have any balancers, BMS or LVC on them of any kind.

Bob

 

[texaspyro]

I get about 4 amp load @ 36 volts using an old bun warmer tray although many suggest a hair dryer for a load.

Hair dryers are usually bad loads... the fan won't run properly at lower voltages (maybe not at all on DC) and the thermal fuse will fuse.

 

[amberwolf]

BTW, These are Thunder Sky prismatic 20ah cells.

Then just measure each cell's voltage (preferably at the terminal ring itself on the cell, rather than on the bolt or the wiring lugs, but checking both places might be helpful) and note that down; any cells significantly lower than the others under load have some sort of problem. Either a bolt is not tight, causing high resistance and voltage drop, or the cell itself has an internal prolbem, or the cell is not being fully charged for some reason (usually one of the first two causes the last).

 

[dumbass]

BTW, These are Thunder Sky prismatic 20ah cells.

Then just measure each cell's voltage (preferably at the terminal ring itself on the cell, rather than on the bolt or the wiring lugs, but checking both places might be helpful) and note that down; any cells significantly lower than the others under load have some sort of problem. Either a bolt is not tight, causing high resistance and voltage drop, or the cell itself has an internal prolbem, or the cell is not being fully charged for some reason (usually one of the first two causes the last).

OK, by the end of this weekend I will recharge and test both my packs and compare the good pack with the problem pack. I'll do it with a small load and again with a higher load. Thanks, Bob

 

[dumbass]

OK, here's my test info. I first charged and balanced all 8 cells to then did 2 tests as follows; test #1 was at a 260w load I tested and recorded each cell. 32 I applied a 30a load and recorded each cell, #3 I applied a 60a load and recorded each cell, #4 I recorded each cell after removal of all load and #5 I bulk charges all cell at one time wit han SLA 24v 2a charger and recorded the results. Right now I would say the cells are all fine and the only problem was they were way out of balance but didn't show it. All thoughts welcome as always.

Bob

CELL # START AT AT AT RECHARGE
CHARGE 260W 30A 60A AT REST

1 3.62 3.50 3.00 2.73 3.32 3.70
2 3.62 3.48 3.00 2.73 3.32 3.70
3 3.64 3.45 3.00 2.73 3.32 3.70
4 3.62 3.46 3.00 2.73 3.32 3.70
5 3.63 3.48 3.00 2.73 3.31 3.35
6 3.60 3.44 3.00 2.73 3.31 3.35
7 3.67 3.46 3.00 2.73 3.31 3.35
8 3.61 3.41 3.00 2.73 3.30 3.35

 

[amberwolf]

It does look like a balancing problem, where the last four cells aren't fully charging, based on the last column.

If you have a single-cell charger, you could recharge them individually with that, because without a shunting BMS or similar, you can't fully charge them along with the other much-more-full cells as they are (or else you'll overcharge the other cells).

 

[dnmun]

or put a load across the first four and drain them down to match the low cells, either before or during charging. but they should not show so much dispersion, those numbers look significant.

 

[dumbass]

It does look like a balancing problem, where the last four cells aren't fully charging, based on the last column.

If you have a single-cell charger, you could recharge them individually with that, because without a shunting BMS or similar, you can't fully charge them along with the other much-more-full cells as they are (or else you'll overcharge the other cells).

Yeah, I did have a single cell charger and did rebalance the cells after I bulk charged them. I just wanted to see the result of how I was charging them in the past. I actually built (in process of) an 8 cell charger from 8 single cell chargers. So I will start using the new charger for now on and see how it works out.

Thanks, Bob