How I Recondition Cells

I literally couldn't do anything for the time being as as nothing sticks out on this side so it is wedged in the remaining portion and cant come out.

 

Why not drill a hole and thread where the temperature sensor enters the module? I think theres more strength there. And i dont think every module has a temperature sensor in it. Only 3 modules out of 28 have a sensor going in.

 

Pretty sure there are no temperature sensors that go in any module. The three sensors sit on the outside if I'm not mistaken.

 

This is all really cool stuff but my original batteries are making like 13 15 years so for 15 $1,700 cash and carry I can't see where all this re-engineering is going to pay off for me or you or whoever this is taking lots of time lots of experimentation in the fooling around because you like that sort of thing or not etc etc versus just getting a new battery like you do for your flashlight installing said battery now you can spend your time re-engineering the radio and the nav system I guess if that's your thing just seems to me this takes a lot of doing and fooling around so if you have the time and that sort of thing I guess have at it thank God I don't have to do this to power my flashlight or my electric bike or anything when those batteries go bad similar to the flashlight you tend to just buy a new one I noticed that with devices like my tools and things like that nobody really rebuils those batteries unless they really have lots of time they order the sales and re-rack the little case for the DeWalt 20 volt why I buy six amp DeWalt 20 volt batteries from other manufacturers for about $24 a piece I have like seven of them sitting here I know the DeWalt battery cost $130.95 whatever it is now. I'll lock for the $25 and the funny thing is is the $25 battery 6 amp hour I'm running for my DeWalt tools have lasted as long as the DeWalt batteries that came with the tools and some of the DeWalt batteries that I bought early on until the fakes were available The fakesin this genre have done a bang up job I still use them almost daily in every DeWalt tool I own which is about 50.

 

Impressive work, Tracy. I've been trying to get to the why of some of the approaches and often no one can explain why, even with very strong opinions and some trial and error results. You seem to be deep in the why.

I had some questions.

If I get your approach in summary (I may miss something here), you start with the usual discharge/charge cycle but with 8a charge rate and 8.4 volts (100% charge) until you get indication of gas or bloat (sounds funny when as I type this), then continue the charge cycle using 2A, followed by discharge to 6v at 6.5A.

When improvement stops, you then do a series of discharges (no charging in between or charge to 6.5 at 2a? Confused here.)
6v at 5a
6v at 2.5a
6v at 1.6a
6v at .8a
6v at .4a
6v at .2a
5.9v at .1a

(I wasn't sure, because I thought you mentioned charging to 6.5v to discharge to 6v, no need to go higher on the charge)

Charge at .1A until between 7.1 and 7.2v
Then charge at 3a to finish 8.4v
sit 24 hours, balance down to 7.9v

My questions (from curiosity to the why, any why details would be great)

1 - why do you use such high amp until you see gas or bloat - is there more than saving time - is there any benefit to charging at such high amp at the beginning cycles? For instance are you trying to get an accurate DeltaV - a problem with some chargers at low amperage charging?

2 - since you get such high capacity value from discharging at the bottom, do you think you get benefit at the top regardless of amps? i.e. charging to 8.4v, does it really provide benefit or is it all from the bottom level work? Have you tried/compared doing the 6.5/6.0 cycle to compare the benefits (or is there some data or standard that tells why)?

I'm obsessed in the why often, but my exposure was a low-mileage 2008 prius battery fail "coincidentally" with a replaced cat that got stolen, resulting in a long period of sitting idle leading to 2 weak modules failing.

I've been using ev-peak quad chargers and DeltaV is unreliable even at 5A charge, so I've had to max it out at 20 to bypass the early cycle stop. Funny thing is a bad cell will trip a deltaV on first charge, makes for early identification. I can theoretically discharge at 5a but in operation, it rarely hits 2a on the display. Been playing with some marginal/bad modules for about a month as I have a spare car, but that's not normal.

All that to say, for the person trying to save/defer buying a new battery pack (hobby charging and multimeter) and needing to get back on the road in a couple of weeks, with DeltaV not reliable, would the summary above make sense, using voltage/capacity cutoffs with safety timer instead of deltaV? Minimalist yet effective approach to the kind of results you have achieved?

 

Tracy seems to have got bored with Prius Chat as his last activity was 31w 2d ago. Whether your post attracts his attention remains to be seen.

This thread was the most in-depth he posted, so while waiting for a reply, maybe some rereading of the whole thread will aid with answering your questions.

 

Since I have some bad modules and a couple of marginally good ones to play with, I took just the sequential 6.0 discharge part of the process as an experiment.

One small point was that the last discharge was 5.9 and the reasoning being that perhaps the equipment might not be precise and so this was a bit of a fail-safe. On the discharge rate, this was clear on my CQ3 charger, where the max discharge was about 1.7a regardless of a "rated" ability of 5a. (nothing new). On the later discharges, a .1 difference was pretty standard - specify X, you get X-.1a except at the very end. (.1a = .1a) Checking your equipment is always good.

What's interesting is that there is nothing special about the timing of the additional discharges, they are very quick and not much lower capacity change until you get to the last one (5.9). That one takes significantly longer at .1A than any other step. Wondering about the accuracy of my chargers (CQ3) and to test my cables, I started measuring volts at the cell and at the banana plug (no issues during charge nor discharge). Seeing some significant differences when the cycle stopped, I put the meter on during discharge and tried to capture immediate termination voltage.

The voltage on meter at immediate termination (reaching 5.92V) matches the charger, but quickly ramps up to 6v+ once the cycle stops, in a smooth quick voltage increase (not an immediate jump). So far, has grown to 6.44v on a marginally good module.

Thinking earlier that my charger was inaccurate, I had set a bad module to discharge to 5.5v, .1a just to see what happened. During discharge, everything matched, as expected. At termination, the change in voltage was a visibly immediate jump from 5.6v to 5.8v, followed by a similar ramp-up to 6+ as with the marginally good module. Stable at 6.1-6.15, which is pretty standard for these bad modules (the math works to be a blown cell in each more or less).

I know this is where electricity meets chemistry, but the slow discharge (.1a) at 6v seems to be eating up volts within the in-spec voltage unlike any previous sequential discharge. The better of my marginally good modules took 96 minutes to get to 5.92 (203 capacity "points"), the other one took 21 min, 46 capacity "points".

On a slow discharge, I don't see how you could get below 6v although it seems possible, perhaps at a higher discharge rate.

With this experiment, I wonder if the extreme low end is where the improvement in capacity was achieved. Charging/Discharging cycle now with typical settings: 5a in, 1.8a out to 6v, DeltaV set to 8mv/cell, capacity cutoff at 7500, 5 min rest on the two moderately good modules.

Next test series will be to see where deltaV kicks in - no doubt it has kicked in at 5mv/cell before, but extremely often very early. Seemed to be less reliable at 2a and spec says that it is unreliable at rates below c/4 (1.6254), pretty close to 2 but the intention is to see how the deltaV on the CQ3 really behaves since it is so critical to prevent overcharge.

 

What gauge cables are you currently using?

 

18awg, rated at being able to carry 14 amps.

Shouldn't have been a problem, but I wanted to check just in case.

 

Any additional tests or experiments worth posting?

 

I just put the battery in the car today - worked on the first try.

I did find the process of discharging repeatedly to 5.8v did help on capacity, breaking past the deltaV so I could overcharge.

Will post my findings, methodologies, kudo's etc. a bit later. Headlines: 2A works like a charm, heat is the enemy, water balloons are invaluable, and the process is indeed repeatable. Buying modules is hit and miss, whether you buy them from eBay or an "experienced" battery rejuvenator (will not name names to protect the guilty...but there may not be any guilt at all, really).

But as a trivia question - I've been playing with my batteries for about 2 months. (12v also disconnected). Discharged to 7.8v before putting it in the car (module by module), then ran a test run to charge the battery to 75% SOC. Kept the A/C off to provide maximum charging, but when I put the A/C on, no cold air - no codes, etc. just no cold air. Since the AC compressor is powered by the HV battery and I had to replace the ECU as well, I'm guessing I missed something. Unless it's just a coincidence and I need a recharge...

Any helpful experiences to share?

 

Congratulations Computer Jeff. No small feat to get the HV bat to work on the first try.

I’ll be interested to read your methodology. As to how long the refurb will last, time will tell.

I’ve found it fairly easy to get entire HV bats in LA from wrecking yards. Not cheap, but at least you have a pretty good idea of the age and mileage on the modules.