P3009 battery Frankenstein project

Here's a page on the MRC "Superbrain" 989. Both Bob and I have this model. You can program it to a variety of different kinds of cells (nicd NiMH...) and tell it how many cells it is charging at a time. Also, you can do charge cycling, by telling it the minimum voltage (per cell) it should discharge to, and it automatically determines how much to charge based on a "DV" number which is in millivolts per cell. I generally have it do 3 cycles, in order to get the final charge and discharge numbers that I'm using to determine these modules charge characteristics. You can also tell it how many milliamps to charge before cutting off, and you can use it as a power supply.

MRC - BATTERY CHARGERS Series - SUPER BRAIN 989, Item: RB989, RC PRODUCTS/BATTERY CHARGERS

Oh, and yes, I have used it to charge about 20 modules at the same time by paralleling them. The charger thinks it's charging 6 cells, and doesn't seem to notice the difference.

 

Here's my plan for assembly. Red numbers indicate modules taken from the newer p3009 battery, Black is from the older battery with some bad cells (uneven state of charge)

Module replacements were selected based on the greatest discharge during cycle testing. The selected modules were discharged to ~7.69 volts each, and then assembled into a parallel battery pack.

The pack was then charged and discharged several times, and then allowed to sit for a couple of hours to equalize voltages between the modules. Then the parallel connections were taken off, and the voltages were allowed to settle for several hours, at which point the voltages in the following table were recorded.

Modules were paired into blocks so that the voltage of each module adds together to create blocks with relatively equal voltages.

	Column 1	Column 2	Column 3	Column 4	Column 5	Column 6
0	Block #	Module #	Voltage	Module #	Voltage	Block voltage[/FONT]
1	1	19	8.23	31	8.19	[/COLOR]16.42[/FONT]
2	2	20	8.23	35	8.2	[/COLOR]16.43[/FONT]
3	3	10	8.23	34	8.2	[/COLOR]16.43[/FONT]
4	4	24	8.23	33	8.2	[/COLOR]16.43[/FONT]
5	5	7	8.23	32	8.2	[/COLOR]16.43[/FONT]
6	6	9	8.23	27	8.2	[/COLOR]16.43[/FONT]
7	7	12	8.23	23	8.2	[/COLOR]16.43[/FONT]
8	8	14	8.23	17	8.2	[/COLOR]16.43[/FONT]
9	9	14	8.23	5	8.2	[/COLOR]16.43[/FONT]
10	10	34	8.23	38	8.21	[/COLOR]16.44[/FONT]
11	11	1	8.22	36	8.21	[/COLOR]16.43[/FONT]
12	12	22	8.22	30	8.21	[/COLOR]16.43[/FONT]
13	13	4	8.22	29	8.21	[/COLOR]16.43[/FONT]
14	14	16	8.22	28	8.21	[/COLOR]16.43[/FONT]
15	15	26	8.22	8	8.21	[/COLOR]16.43[/FONT]
16	16	36	8.22	7	8.21	[/COLOR]16.43[/FONT]
17	17	1	8.22	6	8.21	[/COLOR]16.43[/FONT]
18	18	37	8.22	4	8.21	[/COLOR]16.43[/FONT]
19	19	2	8.21	3	8.21	[/COLOR]16.42[/FONT]

The parallel connections were then remade, and the pack will sit overnight to maintain the voltage balance before assembly and installation tomorrow.

 

For some reason some of the red numbers changed to black while being transferred from Excel to the table tag format. The only modules above that are from the old battery are 1,4,7,9,14,34, and 36 in the left column.

 

It's aliiiiive!

 

	Column 1	Column 2	Column 3	Column 4
0		As installed	After test drive
1		SOC 59.5	SOC 73.0
2	Block	Voltage	Voltage
3	1	16.2	16.9
4	2	16.1	16.8
5	3	16.2	16.8
6	4	16.1	16.8
7	5	16.1	16.8
8	6	16.2	16.9
9	7	16.1	16.8
10	8	16.2	16.9
11	9	16.2	16.9
12	10	16.1	16.9
13	11	16.2	16.9
14	12	16.3	16.9
15	13	16.2	16.9
16	14	16.2	16.9
17	15	16.3	16.9
18	16	16.4	17
19	17	16.4	17
20	18	16.4	17
21	19	16.3	17

 

What about it?

 

I drove it on my business trip today, about 250 miles since I installed it. SOC has settled into 65-68 and voltages are within 0.1 except for 1 module.

I have two MRC's, and I left the modules in the assembled battery for compression, but removed the buss bars to access the terminals. Most of the time I charged 2 modules at a time, using setting of 1.5 Amp charge/discharge, 5 minutes rest between charges, and 5 mV peak voltage (although I don't quite understand what that means). I cycled each module 3 times to get the maximum mAhr discharge/charge numbers, and used these to choose which modules to select for the final pack.

I only parallel charged the whole pack in an attempt to balance the charge before assembling the battery. I had tried it earlier with the old battery to see if it was possible, and then again with the 38 modules that I had chosen for the final assembly. I charged and discharged it several times to get the charges even. Seems to have worked.

I've got way too much time into it, but I learned a lot.

Now I've got an extra battery case, ECU, some possibly usable modules and some really dead modules that need recycling. And two chargers that I doubt I'll have much use for until the next time I decide to do something crazy like this. What to do with it all?

 

Well, this weekend it took us to D.C. for the rally, and while I sat in the passenger seat I programmed the scanguage for the block resistances. According to the codes from the scangauge thread, I guess these measurements are in milliohms. But I don't have any basis for comparison. What are good numbers here?

	Column 1	Column 2
0	Block	Resistance
1	1	21
2	2	21
3	3	19
4	4	20
5	5	21
6	6	21
7	7	20
8	8	21
9	9	20
10	10	20
11	11	20
12	12	20
13	13	20
14	14	20
15	15	20
16	16	20
17	17	20
18	18	20
19	19	20

Also I looked back at the reject modules, and found that the heat pattern from Bob's photos is considerably more evident in photographs than it is to the eye. I don't know what the "still good" modules look like, because they're assembled into a parallel block and on the charger, but this is module 8 and 11 which were among the worst of the modules from the old battery.

 

From what I've seen, the resistance numbers don't change much. Are the voltages relevent? Unfortunately with the scangauge I have to reprogram the xgauges to get the voltages back, so I can't get resistance and voltage from the same time period.

Also, looks like my numbers are similar, maybe a little better than yours. Does anyone have data on a younger battery? Or an older one?

 

I'm bumping this thread, partly because I got a message from Joekingiam asking how to balance a battery, and partly just to give an update on my car to anyone who's interested.

The car is still running fine, with the highest voltage module pair generally reading about .2 or .3 volts higher than the lowest voltage module pair. My fuel mileage actually is about 1 or 2 mpg better than it was before the battery rebuild. Other than that I changed the inverter coolant and pump. I need to do the struts and I will need 2 new tires when I take off the snows. Anyone have a preference for summer tires on a 1st gen?

 

Re: P3009 battery Frankenstein project. Coming up: 1 Year!

Bumping the thread again, since I'm coming up on a year since the battery has been rebuilt. It's still running great. ScanGauge still shows about .2 to .3 volts difference between the highest and lowest modules.

I haven't hung around here at priuschat since I haven't had any problems with the car, which is now coming up on 180,000 miles.

I did get a recall notice from Toyota on some bolts in the steering rack, and took it in to get the recall done. They said that since they couldn't get the rack to zero, they couldn't do the recall without replacing the whole rack. So they gave me a brand new rack. Can't argue with that on a car this old.