Gen II Prius Individual Battery Module Replacement

Something a lot of people don't seem to realize is that when you are discharging to below 0.9V/cell, reversals occur very quickly. In many cases, for very weak cells, reversals have been seen at > 1V/cell at least in the IMA world.

Steve

 

Thanks kiwi, i appreciate your honesty and respect your point of views but not your attitude.

Anyways since our last correspondence i have also gotten Jeffs (HA) grid charger and it worked well for me. Jeff is a great guy and i'd love to work with him again. Jeffs charger and earlier acquired 6 hitec X4 chargers have made a good set of tools for my 2 Priuses (Wife & I) and friends too.
My appreciation to all the folks on this thread who have made it one of the most educational experience on PC for me...ryousideways, MTL_hihy, strawbrad, jbmbenjamin, ericbecky, Texas Hybrid Battery, S keith and others as well as PC senior members like Professor Jeff and Britprius. Can't thank you guys enough for all the knowledge share. I also know that many out there have made notes from this thread just like me.

I visit Chico California quite often and was wondering if there are any hybrid battery recon shops within the area, i wouldn't mind dropping by at some point to add more color to my time out there ?

 

I echo everything ikefor said in his second paragraph. Great info here!!!!!! Its not often when you have this many standouts in single thread but damn, you guys are good.

 

I have measured the internal resistance of the modules using 2 55W halogen bulbs (first one, and then two in parallel), and calculated the internal resistance of the modules as follows:
# V1 I1 V2 I2 dV dA Int. Res.
1 7.72 3.17 7.43 6.05 0.29 2.88 0.101
2 7.52 3.29 7.22 6.42 0.30 3.13 0.096
3 7.65 3.27 7.24 6.46 0.41 3.19 0.129
4 7.51 3.28 7.10 6.40 0.41 3.12 0.131
5 7.54 3.24 7.00 6.35 0.54 3.11 0.174
6 7.46 3.26 7.03 6.32 0.43 3.06 0.141
7 7.52 3.24 7.20 6.42 0.32 3.18 0.101
8 7.58 3.30 7.13 6.40 0.45 3.10 0.145
9 6.95 3.07 7.03 6.35 -0.08 3.28 -0.024
10 7.58 3.30 7.13 6.40 0.45 3.10 0.145
11 7.57 3.24 7.21 6.43 0.36 3.19 0.113
12 7.44 3.25 7.06 6.35 0.38 3.10 0.123
13 7.55 3.20 7.06 6.35 0.49 3.15 0.156
14 7.66 3.32 7.25 6.41 0.41 3.09 0.133
15 7.55 3.23 7.23 6.42 0.32 3.19 0.100
16 7.45 3.27 7.03 6.33 0.42 3.06 0.137
17 6.11 2.85 5.20 5.35 0.91 2.50 0.364
18 7.39 3.25 7.16 6.38 0.23 3.13 0.073
19 7.50 3.23 7.17 6.38 0.33 3.15 0.105
20 7.54 3.29 7.14 6.38 0.40 3.09 0.129
21 7.65 3.22 7.31 6.47 0.34 3.25 0.105
22 7.57 3.29 7.11 6.31 0.46 3.02 0.152
23 7.59 3.20 7.18 6.40 0.41 3.20 0.128
24 7.41 3.23 7.15 6.38 0.26 3.15 0.083
25 7.57 3.21 6.82 6.25 0.75 3.04 0.247
26 7.38 3.24 7.08 6.36 0.30 3.12 0.096
27 7.51 3.21 7.03 6.34 0.48 3.13 0.153
28 7.32 3.21 7.09 6.35 0.23 3.14 0.073

Modules 9, 12, 17, 23, and 25 were already identified as weak based on charging capacity. How many additional modules should be replaced based on internal resistance? 5, 13, 27, & 22? What about 6, 8, and 10? What is a reasonable cut-off?

Dave

 

I'm trying to figure out what you did. It's clear you're using DeltaV/DeltaA to compute a number relating to resistance.

Were these done for a time period, or what? Please describe the process.

The first concern I have is that almost all of your voltages are at or below 7.2V nominal at 6A. If these sticks were depleted or you had applied the loads for an extended period of time, perhaps it's not an issue.

Whatever the case, I find it helpful to look at it visually (note I made up numbers for #9 as you clearly had a typo):



I see 3 groups at the bottom and the two outliers. I would also factor capacity into the mix as well.

 

Hi all,
This is great, I'm on my way to HD to get the items above. One ?, why can't I just recharge the bad cell only.

 

You can't access cells. Each module is six cells in series, and you can charge/discharge a single module.

 

Thanks Keith,
Finished 120watt, 60 sec voltage drop test. Here are the results. Not sure if I should be doing more testing or start setting up for charge/discharge. And should I be replacing any of the modules?

 

View attachment 103032

I would say you have 3 zones of modules. The two highest and next six may be of concern. Are the voltages you reported those at the 60 second mark immediately before you terminated the load or after the load was removed?

Please describe your testing process and equipment used, e.g., charger and charge termination criteria, what you're using to accomplish a 120W load, etc.

Steve

 

This would imply you are using a 0.44 ohm resistor as a load and are drawing about 17.1 amps for your tests.
If you are using light bulbs with marked wattage to do the test they will not be using the wattage marked unless the bulbs are rated at 7.6 volts. The voltage of the modules you are testing.
A bulb only uses it's rated wattage at the voltage marked on the bulb. A lower voltage in will lower the wattage used.

John

 

2.25 Ohm/17.1A?

Nominally, a 2.25ohm resistor would pull 7.2/2.25 = 3.25A
Nominally, a 120W load would pull 120/7.2 = 16.7A

 

Yes my mistake, and I have only been doing it for 60 years. I inverted the equation resistance would be 0.44 ohms. However I was calculating at actual module voltage in this case circa 7.6 volts.

 

Thanks...

Sometimes my grasp of things drifts into the "tenuous" zone, so I have to ask for clarification...

I'm hoping he used a 120W 12V bulb and at least got about 7A out of it.

 

Thanks for keeping us on track John.

 

This code does not suggest a problem with the HV battery. I know you are waiting for a Mini VCI. Until then you are working (guessing) in the dark.

Did you use a 120 watt and 120 volt bulb from Home Depot? The actual load of your test is unknown. We need to know what you used.
Recharging does not fix bad cells or modules.

Once in awhile I disagree with Keith. Or look at a problem from a different perspective. Aman has taken a leap of faith to start looking for problems with his battery. My favorite college textbook was a little pamphlet titled How To LIe With Statistics. Keith's bar graph starts at .42 volts. By not showing the graph down to zero the spread from .45 to .63 volts appears larger than it really is. Given that Aman does not have battery code, we do not know the true load applied, and we do not have a good way to interpret the results I would call the test inconclusive.

Brad

 

P0A94 is a dc to dc converter performance problem.

John

 

Thanks all,
Using setup by MTL_hihy page 13. 12.8v, 35/65w bulb and v2 measurements are at 60 sec mark. I currently have p0a94, but toyota had told me a couple weeks ago that bank 10 was going into negative. Got veepeak mini obd ii, waiting on mini vci to arrive.

 

When you get the Mini VCI cable run Steve's discharge and charge test with graphing. Then you will have a good idea of the state of your battery. Techstream starts the module count opposite the battery ECU. Your test results showed you started counting at the battery ECU. So bank 10 is your module numbers 9 and 10. Their voltage drops were .51 and .48 which is right in the middle of the test results. I am not sure what "bank 10 was going into negative" means. Modules and banks of modules do not go negative in the car.

Brad

 

Brad, I looked at it both ways and chose the more illustrative view. Here it is to 0V:
View attachment 103074
The zones disappear (or take a lot more looking), but the 2 and 6 still stick out.

In any case, that's why I asked for more information because the load wasn't clear. I tried to resist, but I was overcome by the urge to make a chart and comment...

So, we have a 120W 12.8V bulb. I would estimate the load at around 7A. Not great, but better than an 800mA discharger.

Again, how did you charge the modules, and what was the charge termination criteria? I'm trying to establish that you fully charged all your modules, or the discharges may be meaningless.

Steve

 

Steve,
I didn't charge. All I did was measured initial votage (v1), than applied light bulb load and took voltage (v2) at 60 second mark prior to taking off load. Should I have done something else?
I'm thinking of just driving it until I get my mini vci. Any suggestions for looking into p0a94 code in the meantime?