www.greenhybrid.com/forums/f50/what-avoid-when-buying-pack-parts-31278/



the link above us for some background on this but IN A NUT SHELL: I bought a battery pack from a 2003 Civic Hybrid which is been sitting for 4 months at the junkyard.
The starting voltage of the entire pack was 70 and each voltage tap was plus or minus a quarter of a volt at 7 volts.

it went from 70 to 100 extremely quickly like in a matter of 15 seconds.

half an hour in the total voltage is a 165 and each voltage tap is within .1v of the rest except for one which is .12 off

Ill be reporting back. Original honda sticks from Panasonic.

before the week is out I'll have the pack completely returned and then it will be time to disassemble and make one good pack from the two!

I don't know if this will work but I am very optimistic and as always your mileage will also vary

 

What you are describing is pretty normal. Ensure you give it a full 30 hour charge. If that pack was truly operational back in April, it has fairly high SD.


You really should do three full reconditioning cycles. You don't need to go any deeper than 96V, and it would be useful to time each to 120V.


Furthermore, I recommend you let the pack sit a week after the last cycle and consult the voltages to see if there is anything that stands out. 7 days is enough to give you an idea if they will hold charge sufficiently well for use in the car.


Lastly, if you REALLY want to do this right and build the best possible pack, you'll need another $20 in hardware, 2 hours more work and about 3 more weeks calendar time.


Get yourself a 100A battery tester from Amazon.


Pack A = pack in the car now
Pack B = salvage pack.


Fully recondition pack B with

1. 30 hour 350mA charge (10,400mAh input)
2. Discharge to 106V
3. Repeat 1 & 2 a total of 3 cycles
4. 30 hour charge
5. Disassemble pack

Allow the pack to sit for 7 days

1. Record resting voltage with separately attached meter.
2. Affix aluminum spacers and bolts to stick ends.
3. Clamp 100A tester leads on spacers.
4. Apply load for 15 seconds
5. Record lowest voltage observed immediately before terminating load on separately attached voltmeter.
6. Place tester in forced air flow from box fan
7. Repeat 1-6 for all 20 sticks allowing at least 3 minutes between tests for the tester to cool.

Once you have the above data, reassemble the pack, grid charge for at least 4 hours, and install in the car.


Repeat all of the above for Pack A.


Once you have all the load data collected, you will be able to reliably select the best sticks from your pack.


The load test is affected by the 3 primary performance parameters of sticks: 1) capacity, 2) SD, 3) IR.


Sticks with lower capacity (bad) will have lower loaded voltages
Sticks with higher SD (bad) will have lower loaded voltages
Sticks with higher IR (bad) will have lower loaded voltages


Thus, by fully charging sticks, letting them sit for 7 days and then load testing them with 60A, the end voltage of each stick will give you a good indication of the general state of health of the stick.

 

see the only problem with three weeks is that I only have the next one during which the car can be out of service. But I will be quite displeased if after sacrificing two weeks of my vacation this year that I still end up with something that isn't functioning properly.

hell, maybe I can do full 3 reconditioning cycles and just swap the battpack out every couple weeks hahaha. Sadly if the back seat removal wasn't such a pita that might be an option for me

letting it sit for a week makes sense to me. I bet if I had done that with pack A it would have been pretty apparent that it wasn't properly reconditioned.

how long does disassembly and reassembly of one of these packs take? I can't find a single thing about it on YouTube so I guess I'll be the first guy recording it or trying to.

mostly because I'm chickenshit I'm going to shut down when I leave for work in an hour and then resume when I get home nine and a half hours later.
I figure it can't hurt to give it a little bit of time to get used to different status quo than slow self-discharge

 

My proposed method only takes the car out of service for the length of time it takes to pull, build, install packs plus the grid charge of the final build. While you're reconditioning B, you're driving with A. While you're reconditioning A, you're driving with B. Your actual touch time won't be that high except when you're building or tearing down.


Your actual total down time will be about 5 hours (3 pack swaps and 1 build) + the amount of time it takes to do the final grid charge (16-24 hours).

 

Detailed as always, COPY I get what you mean now. So basically a long weekend (I'll be very slow and careful.)
I can make that work!

 

Yah, and I forgot to mention that while there may not be a video of the break down, there is a lot of information online.


You will do best by taking pictures as you disassemble it. 3 or 4 of the sticks have thermistors on them... you need to pull these sticks all out simultaneously and be very cautious with them as they can break easily. They are affixed with an extra layer of shrink.


Take lots of pictures as you disassemble as they will be far more valuable than anything you find online.


Critical during assembly is ensuring sticks are installed in the correct direction and clocked properly to allow the hex and square terminal ends to align with the end plates. If you have to deal with the PTC strips, it will be a pain - just be careful. Do not cut the ends off out of frustration. Lastly, make sure you don't under or over-tighten the stick bolts.

 

up to 177.7v now

Yesterday charged for about 4 hours, discharged for an hour (i really wanted to see if one would seriously sag like pack A). started discharge at 165v, at 150 i stopped having extracted a meager .3Ah then at 530am I began charging again and have left it all day, 18.5 hours. It will be on charging another 8 hours before I need to shut down. (Out of house ALL day) so in 38 hours it will have been charging for almost 28. Not ideal but on cycle 2 and 3 it will be.

headed off to bed now with it up at 177.7 volts Taps: highest at 17.82 the lowest at 17.72.

question: if these things stay at 80% or less, and full charge is 168V, doesn't that mean in theory a 100% charged pack would be closer to 210V ?
I see alot online about not breaking the 178V barrier though.

 

I'm not sure where you're getting "full charge is 168V". NiMH voltage is not particularly linear, and a NiMH cell is DEAD at 1.0V under load - not 0V.

All packs are a little different. I have no idea what the 178V barrier is. I've been doing this awhile, and it's the first I've heard of it. Based on what I know, I don't see how it has any meaning whatsoever.

Let me put it to you this way - NiMH voltage exists purely to confuse you. You will not successfully equate voltage to anything meaningful without including at least two other parameters. Resting voltage is worthless except to confirm a problem. LOADED voltage at a given TEMPERATURE have meaning.

You are interested in how the voltage responds to current at the existing temperature. If it's cooler, the pack will go to higher voltages than if the pack is warmer.

The only things you can mostly count on:

1) When the pack voltage stops rising for 5 hours, you can assume the pack is at or very nearly at true 100% SoC.
1A) Sometimes the pack voltage will actually peak and drop. If this occurs, the pack is fully charged 5 hours AFTER the peak is hit. If the voltage drop is accompanied by heating, the pack is at true 100% SoC.
2) By definition and performance, a NiMH cell is DEAD when it hits 1.0V as the chemical reaction can no longer provide any meaningful energy after that point.
3) A significant temperature rise, say 20-30°F is another indication that it's full.

28 hours is fine. If the voltage does not raise by 0.1V for 5 hours, you may terminate the charge early. If the temperature of the pack raises by 20-30°F, you may terminate charge early.

 

https://hybridautomotive.com/pages/pc
the table says the top end of the operating range for this batt is 168V and I've read may other places that the typical usage ranges is between 20 - 80% because not letting it go too high or low increases longevity of the cells (many more cycles before kaput)

as for 178v. I read a 2 lone-wolf type grid charger builders who both who said once it gets to 178.8 discontinue immediately. Can't find links now though....
that said, what you wrote about heating while charging and voltage drop accompanied by heat fits all the white papers I've read about NiMH.

3ish hours left and it's at 180v

 

I don't agree with those as absolutes, and I don't know what he based those on, or the context.

You can actually see voltages ranging from 120-190 in operation.

Not sure where the lone wolves came up with that. It's probably arbitrary based on nothing meaningful.

A pack's peak voltage is based on the sum of the internal resistance of the 120 individual cells, the charge current and the temperature of each of the 120 cells. Many of the LED PSU have a 20mA tolerance, so you can get variation on that alone.

The generalities I mentioned are what holds true from pack to pack.

Voltage drop at grid charger currents isn't particularly common, but it can happen. Heat will almost always happen unless you're moving a lot of air through it. In any case, there will be a measurable temperature increase from the inlet to the outlet.

It is also common for subsequent charges to achieve higher peak voltages when cycling a pack. I see this on the subpack level as well as individual subpack capacity increases as voltage depression is eliminated. Accurate or not, I attribute this to the influence of the voltage depressed chemistry on the overall voltage of the subpack.