Gen II Prius Individual Battery Module Replacement

After a couple more cycles (depending on the module) all of my modules fell between 6000 and 6300 mah. Pulling them all out and staggering them to average about 6150 per "block" seems to have done the trick. The battery is back in my car and has 100 miles on it with no lights or problems. It's actually getting a few miles per gallon better than before... though it was winter and lower mpg is expected in cold temps.

 

Congrats on a successful rebuild. Did you use my first or second scenario for balancing the blocks?
Gen II Prius Individual Battery Module Replacement | Page 11 | PriusChat


I think averaging capacity per block (not per module) is the most important aspect of rebuilding since that is how the hybrid battery computer checks the health of the overall battery pack (thus no more CELs for weak blocks). I think as long as people are careful about the cycling and do a good job on the balancing capacity, there will be alot more successful DIY hybrid battery rebuilds in the future. I think it also has alot to do with the climate your are in too (ie if you live in a hot climate the battery capacity will go down much faster than if you live in a place that has longer winters).

 

I'm the second owner of my prius. It started life in Florida, then the owner moved to South Carolina 20,000 miles in. I purchased the car with close to 200,000 miles and took a gamble that it would last close to another 100,000 (though it was such a good deal it was hardly a gamble). I live in michigan so needless to say our winters are horrible. According to the original owner the engine, transmission, and battery are all original. I'm surprised that my battery lasted so long living in a mostly hot climate.

 

Hi very smart way to go. Can you please tell me what kind of 55watts halogen bolb you used for battery load, I am trying to used a load tester bolb as you did to know resistance in each module.

 

I believe that module 28 is the one closed to hv computer in gen2. Contrary to gen1, where module #1 is closed to hv computer. If I am wrong please let me know.

 

I think this is the case. Counting from the "non computer" end of my battery it was module "16" that was bad and my code was "Battery block 8 becomes week".

 

I used a headlight from a 1997 silverado... Worked for me.

 

What did you notice about the bad module vs good ones when loaded? Voltage drop? How much?

 

The voltage on good modules dropped about .2v each while the "bad" module dropped over a full volt.

 

Nice to know because after cycling my modules they all seemed to be able to hold capacity well (none less than 6,100 mAh) and they were well balanced for 24 hours (see balancing pic) but when I installed the packs again I got the same "weak bank 1" error I did before so I must have missed something. I wish I knew how the banks were counted on mine so I could narrow down which 2 modules were in question, but no one seems to have figured this out yet (please post if you do). Anyway I figured there must be an issue that only shows up when loading the modules since the capacities all seemed to test fine. I'm going to pull out the packs again and load test the modules individually to see if I find any with possible issues. I'll start by creating a 65W (high beam) headlight harness with a wall switch and multimeter to read voltage drops as you have to hopefully ferret out the bad one.

On the bright side at least I had the smarts not to put the interior back in before I verified everything was working.

Attached is a pic of balancing the packs........be careful of those leads or you'll be welding.

 

MTL, you have 6100mah on all modules, they were balanced for 24hrs and you still got a bank 1 error??? That sounds like it might be the voltage sensing line to me. Do you have miniVCI or something else that you can read the bank 1 voltage in real time? because if everything is as you say it is, there should be no reason to get a bank 1 error.

 

Yes I do have mini VCI and I'll take a few screenshots tonight so people can see what I'm talking about as I drive around. I kept good records on all the charging cycles and I will post those too just so people will have something to reference later.

I wasn't sure if you could have good capacity on the module, but still have voltage drop too low when a load was applied.......sounds like no. I will also try loading the modules to see if that shows me anything. Another option may be to break the pack apart and shift the modules around to see if the issue follows a "bad module" or if it keeps saying there's an error on bank 1 despite the change (hence an issue with the computer or voltage sensing line). It would be nice if the battery turned out to be good after all since they aren't cheap as you all know.

 

Sounds like a good plan.

 

I only did 2 charge cycles on each in most cases since the packs typically went above 6100 mAh on the 2nd discharge cycle when charged to 7000 mAh at 2A current rate. Once they reached above 6100 mAh capacity, I considered them to be "good" so no further testing was done beyond that point besides a final (discharge>charge) cycle of 4500 mAh before balancing all modules together.

Here's the test results on the packs as promised (start voltage, discharge 1st mAh, discharge 2nd mAh, discharge 3rd mAh, etc):

Pack 1 (passenger side, 12 modules)
10.88v, 5621, 6189
10.85v, 5656, 6227
10.83v, 5571, 6133
10.81v, 5385, 6032, 6113
10.79v, 5352, 6005, 6099
10.81v, 5232, 5975, 6140
10.79v, 5424, 6107
10.79v, 5325, 6131
10.80v, 5521, 6124
10.81v, 5451, 6146
10.85v, 5684, 6176
10.86v, 5797, 6308

Pack 2 (middle, 6 modules)
10.87v, 5643, 6284
10.87v, 5529, 6192
10.85v, 5520, 6217
10.85v, 5491, 6176
10.85v, 5562, 6247
10.88v, 5870, 6356

Pack 3 (driver's side, 12 modules)
10.88v, 5799, 6272
10.86v, 5692, 6158
10.85v, 5698, 6271
10.83v, 5508, 6103
10.82v, 5371, 6143
10.82v, 5367, 6083, 6123
10.81v, 5267, 6019, 6119
10.83v, 5374, 5973, 6133
10.84v, 5547, 6217
10.84v, 5634, 6204
10.86v, 5764, 6291
10.88v, 5774, 6307

Note: For people finding this later, these are 9.6V (8 cell, 6,500 mAh) Highlander modules so capacities are similar to the Prius, but voltages will be different

 

Good to note this!
It'd be bad to see Prius modules up in this voltage range!!

 

Yeah, with those capacities, there's no way you should have any faults. I would take a good hard look at that sensor.

 

You mean the sensor wires coming off each block?

I still haven't had a chance to load test them yet, will need to wait until this weekend to try that.

 

The discharge capacity test is essentially a load test. You're showing good capacity. Yes, if those are the wires that measure/sense the voltage that's giving you an error.

 

For some reason I thought the capacity and load tests were different, but hopefully when I test them I will be able to see what you are saying is true.

Do you guys think that 65W would be enough of a load to see any issues with load testing on these batteries? Should I be using larger loads to check voltage drop?

 

Your modules (8 cells in series) should have a series resistance of about 0.02 ohms. so with a 6 amp load as you specified, you will see about 0.1 volt drop in a good module. So a bad module should be detectable with your 65 watt load.

JeffD