When it isn't a different state of charge, it's a different level of internal resistance.
It gets a little confusing because when a cell reaches it's full state of charge it gets hot, but a cell with high internal resistance also gets hotter than the average cell(with both charge and discharge) along with the voltage sag during discharge or higher voltage on charge, but of course the end of charge heating occurs with all nickel-based cells.
From everything I've ever read, there is really nothing that can be done once internal resistance develops in a cell for most chemistries, with the exception of lead-acid which can be desulphated, which has its own compromises.
I added a computation of internal resistance to the program based on the voltage rise when switching from 6A to 30A. I am getting numbers like 0.0047 ohms, but it is highly dependant on picking the right points to sample for voltage.
I agree that if high Internal resistance is found, there may not be much that can be done. After logging the first charge so I know the initial subpack conditions, I will try various things to "fix" the issues. I dug out my giant SCR to make a high current switch for the cell zapping and spot welding, and hope to get that operational soon.
A single cell charger is also on the list. Maybe if I can cycle only the cells that are behind the others,I can achieve a better balance than trying to do the whole subpack.
I als need some cooling if I will be testing at 100A. Dont want to cook any more cells. The first 100A test got the subpack to 150F.
A close look at subpack 2 trace, shows that the white #1 cell reaches a plateau and starts to get hot while the rest continue to rise and are running cooler, I am assuming that this means that cell 1 is fully charged and the rest are still charging? http://99mpg.com/ProjectCars/mimapackwhack/subpacks18/
If anyone wants a copy of the actual excel spreadsheet datalog file to look at, just send me an e-mail and I will send them to you.
A single cell charger is also on the list. Maybe if I can cycle only the cells that are behind the others,I can achieve a better balance than trying to do the whole subpack.
I als need some cooling if I will be testing at 100A. Dont want to cook any more cells. The first 100A test got the subpack to 150F.
A close look at subpack 2 trace, shows that the white #1 cell reaches a plateau and starts to get hot while the rest continue to rise and are running cooler, I am assuming that this means that cell 1 is fully charged and the rest are still charging? MIMA Honda Insight Modified Integrated Motor Assist - Subpacks 1-8
If anyone wants a copy of the actual excel spreadsheet datalog file to look at, just send me an e-mail and I will send them to you.
Normally I see a dip that goes along with the temperature rise, I've never seen the voltage hang that steady when I've played with them, so I couldn't tell you if they were done, I'd probably keep charging until that cell reached 120 degrees or I see voltage drop to be sure, unless the drop was missed already. It's hard to tell when you can't see the full picture from discharged to charged at the same amperage throughout.
I played with the idea in my head about single cell charging and then realized that it might not help much. The reason being is that usually we charge these when cycling with slight overcharge, which I've verified by measuring temperature. I played with charging them and measuring temperature with an infrared thermometer and watching the lowest cells to make sure that by the time they terminated the charge(which the lowest temperature cells always did later) that they had the voltage drop to indicate that their charge was complete. Since the cells that have the least capacity are the ones to drop to the .9 volt level first, they will already be getting the full cycle, so charging separately might not help much. Although taking the lowest capacity cell and discharging at a very low amperage(say .6 amps or something) to .9 volts to ensure they are really getting a full cycle might help, but I'm not sure to what extent, I have a feeling that it won't progress things too far, but it's worth a shot. I've read in some places that the actual cell reversal is at 0.6 volts if done at a very low amperage, I can't find this information anymore, so I wouldn't do it unless it was for experimentation on a stick that is already deemed bad in case it makes things worse.
As far as cooling goes, I tried putting a few sticks in the refrigerator for an hour and then charging them at 7 amps(max of my charger) and the voltage was much higher as they can't absorb it as well but at the end of charge I noticed the same stick had a much wider final temperature range when before they ended at a difference of less than 5 degrees C and I'm not sure if what I was doing was a bad idea or not. I've been avoiding getting them hotter than 45 degrees C (113 degrees F), but have noticed that at 7 amps when stopping the charge at 45 degrees C that they have climbed to 50 degrees C in about 5 minutes, even with that relatively low amperage. Most commercially sold chargers that I've come across terminate between 45 degrees C if based solely on temperature and ones based on -dV use 50 degrees C as the 'failsafe cutoff' in case standard end of charge detection fails.
I am running a log term 350ma topping charge on stick #9.
The stick showed moderate cell to cell voltage difference at the start.
I quick charged it to the first cell temp rise, then stuck one of the LPC-20-350 on the whole stick.
After 2 hours the three lower voltage cells are just showing a gradual voltage rise the two highest voltage cells seem to be pretty much stable, and the intermediate cell has risen to just about where the two highest voltage cells are sitting. I will be out all afternoon and will see where things are when we return in 4-5 hours.
When it isn't a different state of charge, it's a different level of internal resistance.
It gets a little confusing because when a cell reaches it's full state of charge it gets hot, but a cell with high internal resistance also gets hotter than the average cell(with both charge and discharge) along with the voltage sag during discharge or higher voltage on charge, but of course the end of charge heating occurs with all nickel-based cells.
The increase in temperature at the end of Charge for a NiMH battery is because you changed what chemical reaction is the dominate chemical reaction you are doing inside the cell.... but it is not actually part of the charging process itself.
While the charging chemical reaction is the dominate chemical reaction in the cell there is usually very little heat given off.
Abnormal internal cell impedance can be measured with varying changes in voltage... at 1 amp what is the dV ... as 2 amp what is the dV ... etc...
Keep in mind even in a healthy cell the internal impedance changes as the SoC changes ... it is not static over the entire SoC, and the cell impedance also changes as other things like temperature change.
Back from the 150 mile trip, and after over 9 hours, the weaker cells seem to have gradually risen in voltage to nearly the same as the stronger cells. The subpack sitting in my tester, where it could cool easily was about 8 degrees above ambient.
I suspect that a whole pack full of subpacks would get much hotter even at 350ma, especially in the center, if the fan was not running.
Note the rather big cell voltage difference when the quick charge started, and also when it ended. The slow constant charge may be required to fully rebaalance the cells?
I will let it soak till I go to bed, and see where things are tomorrow AM. Then I will do a discharge to see if the unbalanced condition returns. We may finally get to see the rebalancing process at work?
Should be very informative. MIMA Honda Insight Modified Integrated Motor Assist - Rebalancing?
The slow constant charge may be required to fully rebaalance the cells?
As appealing to so many as the fast is... I suspect you are correct... The only work around I can think of is to have the charge and discharge circuits doing smaller amounts of cells at a time... individual cells would be ideal ... but we could probably get away with individual 6 cell subpacks.
Or we just have to make do with the slower rate... which will be vastly easier and cheaper.
I was running around trying to get the accident taken care of with not too much luck. The lady's car was a company lese, and the insurance is through the leaser.
After posting the photos last night, I let the charge run for about 3 hours more. I discharged them this morning. Photos here: http://99mpg.com/Projectcars/mimapackwhack/
All the data is there, but what does it mean. I hope to trace a bunch of subpacks over the weekend, and will give #9 a few more heavy cycles to see where the trends go. All ideas as to the best way to use the subpack analyzer will be considered, and if you want to help with and get first hand experience I will be working on this tomorrow.
May dig out the Zapper and see what that does to things. Where is Sherlock Holmes when you need him.
It is still early / preliminary ... we will have to see how it pans out after a few cycles ... and as you posted the slower rate may be needed for parts of that process ... charging and discharging.
But ... from this early data it suggested you might be seeing the difference between different phenomenon ... A lowered voltage due to lowered Capacity ... and the separate issue of voltage depression ... which can sometimes be confused as the same thing , but are not... and a difference in SoC ... but hopefully over time some of these things will begin to filter out.
Baring yet unforeseen events ... it looks like I might get a chance to swing over finally this coming weekend 3-13-10... let me know if you could use a hand then.
Sounds good Ian, I ran stick # 9 through another cycle with all of the cells tracking much better than the first discharge, so the long topping seems to have drastically changed things. http://99mpg.com/Projectcars/mimapackwhack
The AutoGuide.com network consists of the largest network of enthusiast-owned enthusiast-operated automotive communities.
AutoGuide.com provides the latest car reviews, auto show coverage, new car prices, and automotive news. The AutoGuide network operates more than 100 automotive forums where our users consult peers for shopping information and advice, and share opinions as a community.
