I am curious about new subpack mAh values

[jetmon135]

Hi all,

I have tested three of my subpacks so far and all three have had mAh of 7310 across the board on the charge cycle and around the mid to upper 6k up to 7k range on the discharges. My pack overall so far does not seem too out of whack but I still have 17 more subpacks to test. I am very interested in subpack pairing and what discharge/charge values would definitely indicate a failed subpack so I can replace the bad ones. If all my subpacks test out within limits, I guess it is safe to assume that the problem lies with some other component(s).

Thanks,

Alex

[Hybrid-Battery-Repair]

At what discharge rate?

Originally Posted by jetmon135

Hi all,

I have tested three of my subpacks so far and all three have had mAh of 7310 across the board on the charge cycle and around the mid to upper 6k up to 7k range on the discharges. My pack overall so far does not seem too out of whack but I still have 17 more subpacks to test. I am very interested in subpack pairing and what discharge/charge values would definitely indicate a failed subpack so I can replace the bad ones. If all my subpacks test out within limits, I guess it is safe to assume that the problem lies with some other component(s).

Thanks,

Alex

[jetmon135]

8 amps, the Superbrain cannot handle 10 Amp discharge rate without overheating. At least mine cannot.

[Hybrid-Battery-Repair]

Originally Posted by jetmon135

8 amps, the Superbrain cannot handle 10 Amp discharge rate without overheating. At least mine cannot.

It's been a very long time since I ran anything that low, but 6000s are good. Do they match each other?

[eq1]

7310 must be some kind of limit and/or some kind of quirk. A few of my sticks hit that on the first or second charge, but subsequently dropped (though not much). Two of them turned out to be the best sticks, one dropped more but was still among the best. Most of the sticks in Eli's and Steve's packs were hitting 7310 on the first charge, but they also dropped on subsequent charges. I'd still cycle these sticks 3 times, but that's probably it. If the rest of your sticks are similar your pack's probably really good and just needs this sort of cycling - or a grid charge, had you not already taken the pack apart... Check out the first spreadsheet in that book I posted in another one of your posts the other day...

Here's a graph of my cycling. Most of it is 10A discharge, 7A charge. The 5th cycle happened after some time passed, and was a 3.5A discharge, 6.5A charge...


As far as the exact figures that indicate which sticks to replace - it depends. Ideally you'd have the cycling data plus internal resistance and self discharge figures. You'd compare them all and throw out the weakest sticks - you'd try to make all the sticks as similar as possible. Comparing the 'curves' in my chart, for instance, sticks 12 and 13 are not hitting the the charge/discharge values like the others. I would have replaced those, and a few others, if I cared more. I ended up rearranging my sticks in the pack so that the weakest sticks were paired with the strongest, since the voltage taps happen for pairs of sticks rather than singles... To-date my pack's been working great - likely weak compared to a new one, but it's been lasting, assisting, charging, not throwing codes or recal-ing oddly...

[jetmon135]

I am still in the process of learning and I really wanted to dig in and understand what makes the IMA system work. I am currently getting about 64 mpg on a tank of gas with the IMA pack removed. I am interested in getting the back sorted out and then seeing if I can get in to lean burn on my commute. I really do not want to mess with the wires too much on the four subpacks that have the sensors. I am going to try to run the superbrain on them in place and hope that they are decent. How do you pair a stronger subpack with a weaker one?

Thanks,

Alex

[eq1]

I didn't mess with the sticks with the wires on them, either. I basically left all sticks in the case when I charged, but at the end I shuffled them around, figuring out a scheme where I could leave the sticks with the wires attached alone, while pairing the worst with the best...

If you look at the 3rd worksheet of that excel book I posted in your other thread, the top two or three rows show how I re-shuffled the sticks. The lower row is labeled 'original' and has the sticks numbered; the upper row shows where each stick ended up.

As I mentioned before, it's all a little messy, but you could probably figure it out. If you look down each column you'll see the various results, calculations, for each stick and you'll probably get some sense of why some sticks were moved. It's all kind of taken with a grain of salt though as my data collection and calculations are likely only good enough for something just a little better than leaving things alone, or making a guess in-the-dark... Not all the results pointed to clear-cut 'bad' and 'good' sticks. For example, my internal resistance calculations show a couple low values, but other things, like the charge and discharge pattern, show a stick that's fine, or among the better ones... I usually went with the charge/discharge pattern, that is, that was the most important factor deciding which sticks were weakest... Like sticks 12 and 13...

Here's a snip of that spreadsheet:


The row with "wire" or "w" shows which sticks have the wire attached.
The row with "t" shows which sticks have the voltage tap. Pretty sure this is right. But anyway, the colored cells show which sticks form pairs, that is, which sticks form a pair for which the BCM monitors voltage. The re-pairing, shuffling idea - I must have read it somewhere, or at least something similar - follows the logic that if 2 sticks are treated as one when it comes to being monitored for voltage, while voltage has some repercussions on what the BCM decides to do in terms of battery management or output, performance, etc., then pairing the weaker with the stronger will produce pairs that look more alike - in terms of voltage, at least... Seems to make some sense to me, but I have no idea whether it makes much if any difference...

[jetmon135]

Thanks,

It looks like it worth a try.

Alex