Battery rebuild on a dead 2002 Prius (NHW11)--Mixing Gen 2 with Gen 1

This was started as a private conversation on accident (noob first post blunder), so I'm restarting the thread here.​

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
Back story (skip if you like): My local Toyota dealership (Toyota of Orange) discharged my traction battery during a recall service for mechanical part (pinion nuts). They were offering a free oil change with the recall. My car was coding before this for a few months. I, as always, was out of cash to fix it. I've owned the car outright since 2007, and have enjoyed very low maintenance (this is my Karma?) so I jumped on the recall and free oil change. All coding problems had been ELECTRICAL. ICE would start/stop constantly, plus I was having trouble starting the car (would have to turn off & on several times to avoid angry triangle of death).

Long story short, I told them to ignore codes and just fix my chief complaint (i.e. pinion nuts + oil change--both of which are mechanical services only). I told them NOT to work on the electrical and that I would repair at a later date (when I had money). TofO killed my battery (left it to discharge, not in "READY" position). They then went on to explain they could not recharge the HV battery (not trained or equipped, etc.), and then wanted to hold my car hostage for $3800 quoted for replacement battery or else I could get in a new Prius and trade in my old one! Gee whiz, what a great offer!

As time ticked by with no action, I grew angrier. TofO tried strong-arm tactic of saying they would charge for storage (approaching 1 month on the lot). I wrote an angry demand letter stating I expected my car back in the same working condition as when I dropped it off within 30 days. They tried dropping the price to 3000 for the battery, then to $300 (allegedly for two "factory techs" aka magicians to clean my battery for 4 hours using sodium borate, in order to allow the battery to charge). TofO stated battery corrosion was preventing charging. After declining offer of $300 and waiting against my demand letter time, they finally offered it back up, working order, no charge. I haven't been back since and have spread bad faith about their service to everyone I know. This was in Feb. to March of 2012. I knew back then that individual dead cells could be replaced at significant savings (~$1600 or cheaper for an entire new battery, less for single cells--BTW, I use "cell" and "module" interchangeably. If I say cell, I mean one 7.2v module, and vice versa--I see no need to discuss single 1.2v cells here, and if I do, I will be sure to clarify).

Well, my backup car just crapped out (a 1996 Rav4), so here I am.
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

My equipment:
2002 Prius OEM Toyota Service Manuals (x2 volumes) (~$60--eBay)
BAFX OBD-II scanner ($24--Amazon)
Magid Linesman Glove Protectors ($35--Amazon)
Salisbury Class 2 Linesman Gloves ($30--eBay)
Thunder T6 Multifunction Charger/Balancer (x2) ($85 ea. + shipping--Hobby Partz)
HiTec Power Supply ePower Box 540W ($110--RC Planet)
1 meter charging cables (x10--purchased as a lot) (~$5--Aliexpress)
Quantum Storage Systems plastic cart (~$80+$30 shipping--eBay)
Orange electrical tape ($4.00--Home Depot)
Home Depot "Homer" bucket ($5.00--Home Depot)
Vinegar, baking soda, shop towels, rags, steel and brass bristle brushes, toothpicks, Q-Tips, etc. (on-hand)
Socket set + hand & power drivers (on-hand)
--------------------------Totals---> Roughly $600 so far (a good deal, as now I have the "infrastructure" needed to continue repairs as modules die).
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

My problem: Car has been sitting lifeless since about April or May of 2012. The TofO "repair" did not last long. Lacking money to fix it, I parked it and forgot about it. I didn't even unplug the 12v battery (hangs head in shame).

Well, I am now determined and at least half-way through my repair thanks to this forum and diligent online research. I plugged in my new JNC660 ($130) to rejuvenate my 12v to pull codes on my also new BAFX OBD-II scanner ($24).

Got a P3030. Gives me a starting point. Looking online, I discover a wealth of information from PC and Luscious Garage. An HV sensing line snapped. Thanks to LG, I also know what that looks like. So I ordered a Prius Haynes Manual from Amazon (big waste of money--says nothing about battery). Angry, I wrote a nasty review on Amazon and I found the two-volume 2002 Prius OEM repair manual on eBay for $50 (whew! significant savings!).

Upon disassembly of back seat & gaining access to my battery, I discovered TofO did NOT clean the corrosion off my battery (like, at ALL). I found the source of the P3030 (yellow line severed from corrosion and/or heat). There was also some arc melting of a bus terminal that connected the pack to the ECU. P3030 was all my OBD detected, but examining my TofO service paperwork, I noticed other codes (one of which I know was addressed with the oil change).



After an initial survey of the HV battery (using multimeter leads on the series still strung up in HV configuration), I believed my overall voltage to be around 54 volts. Some quick number crunching told me 54 volts was right around 20% of the nominal 273.6v. Yay, I still had a charge! And I believed it was high enough not to have really seriously damaged the battery (below 20% would have been really bad). However, it gets better.

I called Luscious Garage and spoke with Carolyn (very friendly & knowledgeable owner). I wanted to know how much they charge just for the "#2 Wireframe Assembly." She stated she does not sell those directly to customers, but that I could go online and purchase the part from Toyota for wholesale pricing (I did find it online-->about $116 for the busbar assembly, or $46 just for wireframe assembly #2 with the sensing wires). She asked a few questions, and I told her my initial voltage readings. She said it's basically a brick. Well, that was a downer.

Finally when my Quantum cart arrived, I could pull the battery out, get the other bus bar off, and safely test individual cells. To my delight, the aggregate voltage taken individually measured 163.28 volts! Now I REALLY was in business! That meant my battery was averaging a 60% SOC, right about where it should be. About the same time, I ran across a forum thread that stated the HV battery could be left for 10 years, without being put under working load, and still have enough high voltage to kill. I also discovered two VERY dead cells (adjacent, though in different battery blocks from one another). I guessed the dead cells were providing so much internal resistance, the battery (when connected in series) was not able to communicate all the electricity from one side to the other (akin to taking a five lane freeway, narrowing it to 1 lane, throwing in a speedbump, then widening it back to 5 lanes).

I ordered a replacement for just those two cells (yes, I know I need to match capacities and voltage). I am going to try an experiment. I understand the risks of reverse charging, but I just can't afford to buy 2x 38 cells to have a nice, happy battery. At this point, a moving car is worth it to me. So my plan is this: I will definitely replace the two dead cells. I will continue to survey the health of my other cells as they undergo the discharge/charge process. I may replace a few other cells, if I determine it to be in my battery's best interest. It looks like many of my cells are healthy, while a few are showing some weakness (voltage swings, fast charge or discharge time, low mAh capacity, etc.). I am performing a predictive analysis for which cells would be next to go, so I know how many and how soon I will need to replace them. I have to weigh cost with benefit and can only replace a few more cells (maybe 5 more--7 total).

Finally, the big experiment: since gen 1 cells are harder & harder to come by, I am going to be MIXING at least two gen 2 cells into my 38 cell pack (to replace the two dead cells). I didn't want to do this initially, but as I thought about it, gen 1 cells are mostly garbage by now. I know this is inadvisable due to the reverse charging effect (differences in internal resistance and especially capacities). But my plan is to cycle in the newer gen 2 or even gen 3 cells (if those fit) as my old ones go out. From what I understand, the gen 2 cells are FAR superior to the original gen 1s (far fewer failure rates, much less internal resistance--think 5 lane highway vs. 2 lane). I can afford about $60 to $100 per month on 2-3 batteries at a time. I can't afford $600 or $1600. After 3 to 4 months, I'll have replaced 1/3 of my cells, and I should be sitting happy as a clam.

I know this is long for an initial post, so I apologize. There will be data to follow. To borrow a line from Alien, "I'm still collating."

Also, a special thanks to posts from JeffD, Britprius, and especially Bob Wilson and ryousideways. You guys have been VERY informative and helpful.

Oh, and Chief, I invited you because I was an STG2. Sagire, Classis, Destructum.

-----TheDoof​

 

I ended up reworking the wire frame assembly by cleaning it off in vinegar/baking soda, then rinsing & drying. Then I reattached the copper by stripping the broken ends, twisting them together, using a little dab of solder to improve the connection, and using some heat shrink tubing to seal it back up, then I wrapped it up in a bundle with the other sensing wires using orange electrical tape. What I effectively have now is an "up-armored" cable that should be able to handle a little potash drip here & there.

I've scrubbed the bus bars and rinsed them overnight in vinegar to get rid of the copper sulfate on the bus plates. That icky caramel colored sealant glue from the service campaign (SSC-40G) really does not want to let go. I got most of the offending substance off the busbars. I may need to reseal the battery terminals with something non-conductive. I was thinking of using No-Alox on most of the electrical contacts (esp. the busbars and sensing wires) to help prevent corrosion from leaky acid. Would this be advisable?

There was definitely some kind of arc fire once-upon-a-time, which makes me nervous... Any hints as to what caused that? Was it a short when the sensing wire blew?

 

Oh, and I was getting a P3009 code (HV leak), according to the dealership paperwork, but I think their voodoo priests factory technicians may have corrected that. Probably was from a stream of leaky electrolyte, as I could not find any likely candidates for "mushroomed deposit" grounds (per LG description of P3009 DTC fault, see below). I am going to add a strip of electrical tape along the length of the top of the busbar and battery modules (and other places where a leak may short, like at the ends of modules 1 & 38, where the module is separated from the metal casing by a few millimeters) to reduce the likelihood of this causing a P3009.

 

(From Patrick Wong, 6/27/13 at 6:27 p.m.)

1. You should measure the voltage of each module, which should be at least 7.2V which is the nominal voltage rating. Ideally the modules would show 7.6V or more. If a module shows a lower voltage than 7.2V I question whether the module can be restored to productive service.

2. The difference between the module voltages should be no more than 0.2V. If greater than that, the battery ECU will probably log a DTC.

3. SOC is not based upon a % of the nominal battery voltage. When the SOC is reported at 50-75% per the Classic SOC gauge, the actual battery voltage will be in the 290V - 320V range.

4. It will not be easy to mix 2G modules into the Classic battery case which is partially equipped with some Classic modules because the physical form factor is slightly different and the busbars will not fit.

5. I suggest you should plan on replacing all 38 Classic battery modules, and budget your personal finances accordingly. Good luck.

 

Thanks for the suggestions! Does that mean each module should be measuring at least 7.2 volts at rest? And especially after 1 year of non-operation? It sounds like each module should run 11.22 volts at maximum charge (based on 320v = 75% SOC --> 320/.75 = ~426 peak volts divided by 38 modules -- I know these are rough numbers, not exact attributes. I'm just trying to get a sense of overall health of the pack). What are the max voltages I should expect from a cell? I'm seeing some numbers in the high 9s after a full charge, and some in the low to mid 10s. How long should it take each cell take to discharge to 6.0v under a .7A load?

As far as the voltage differences go, I should be able to match those by running the battery in parallel for 24 hours, followed by a targeted discharge using a 55w halogen for the load, correct? That is what ryousideways did, and it seemed to work for him. Plus I have plenty of speaker wire.

I can (and will probably have to) modify the busbars slightly. I can drill slightly larger holes or some such other stopgap measure... but I'll cross that bridge when I get to it (as soon as my new modules arrive). I'll have to in order to determine exactly how they need modding. Necessity breeds invention.

My survey numbers are showing that I will need to replace at least 10 modules now, and 9 or 10 in the near future. So much for my dream of replacing only 2 now. I'm willing to risk some. I'm not willing to risk all. Plus, time is a factor for me as school starts in August.

I'll probably end up shuffling the modules (matching them) and placing the best NHW11 modules on the ends, with the NHW20 modules in the middle (better heat profile). That will help make my busbar mod easier to accomplish, too.

If, at the end of this experiment, the busbars need to be replaced, I can do that. They are worn and could use replacing anyway, so I don't fear messing them up too much. I already had to jury-rig the sensing wire to battery block 8 (solid yellow). I don't even know if my fix will work and won't know until my pack is balanced & reinstalled in the vehicle. I don't have a megger and can't test resistance in the line (and/or don't know how to use all the functions on my multimeter to properly test for resistance).

But I did have a question about capacities:

I understand it generally takes 3 cycles of discharge then charge to condition the cells. That is fine for a battery that has not sat idle for the last 12-13 months. My numbers are a little disconcerting, and I'm not sure I'm doing everything properly. I don't know whether to believe the numbers. Perhaps my battery is unrecoverable.

When discharging, my smart chargers give one mAh value, and when charging, they give another (typically lower numbers upon discharge, higher numbers after a charge, sometimes radically different). The worst cells discharge (and/or charge) too rapidly (giving low mAh numbers), and I'm sure I'll have to replace those. But how do I find the true Ah capacity? Is it the ending number after the last discharge (or charge)? Is it the average of the three charges? Or is it something else, maybe having to do with some complex formula that figures in the discharge numbers as well? My numbers are not very high (nowhere near 6.5 Ah). Yes, I can Google, but I am not running into anything terribly enlightening, and this is the best place for me to posit my questions. Also, is it better to discharge first, and then charge? Or is it better to charge first, then discharge to the desired voltage? The latter was my initial thinking, but I see others are doing it the first way.

As I have been led to believe, a standard new cell should have a capacity of 6.5 Ah, but a good cell will hold at least 3 Ah (according to "Battery Boy"). Most of my "good" cells are below 3, but above 2 Ah. That is after 3 cycles of charge, then discharge only to 7.3 volts (I don't know why I didn't start out discharging to 6.0v in the first place, but I didn't). Then I did one cycle of a discharge to 6.0 volts, followed by a charge with an mAh limit of 7250 (up from 6750 for the first 3 cycles). I know my HV battery is fairly crippled, but I'm trying to give it a little life for a few months at a time. I'm replacing all the cells below 1 Ah. I know the new cells I ordered will not match the old cells' capacities. They'll just have to do until I can cycle in newer cells as I can afford them. Hopefully they'll hold up that long. As I said, this is an experiment. Others have asked. I might as well do.

 

(also from Patrick Wong):

As I previously stated, you have an incorrect concept: that SOC is directly proportional to battery voltage where 0% equals 0V and 100% equals some maximum voltage spec of the module. Please discard that concept.

After one year of non-operation there is a high probability that the modules have deteriorated to the point that they cannot be restored to proper operation, hence my suggestion that you plan to replace all 38 now.

I believe you are over-charging the modules. The maximum voltage you should charge a module to is 8.4V, which is the voltage that the Prius traction battery ECU will allow the module to charge to.

 

(From usnavystgc):

Doof,
Here's is some advice from a retired STGCS(SW). Slow down!!!. And I know you think its going super slow already. The first thing that needs to be addressed is physical not electrical. As Patrick correctly stated, there is an difference in physical size between Gen I and Gen II modules. Theoretically, you should be able to mix them (although this is not ideal) but, do you want to do some serious mods to the battery casing or, do you want to have a bastardized temporary setup? That's up to you but that is a lot of work. Now as a Senior Chief, I want to help, give good advice and take care of my sailors. Here is my advice to you (and I'm certain many will disagree but, I'm trying to take in your budgetary limitations and get you up and running reliably).

Rebuild your battery by replacing the bad modules with Gen I modules. Here's my logic and reasoning:
1) The Gen I battery has 38 modules (that requires using 2 Gen II/III battery packs to make one. With your budget, replacing all modules isn't feasible right now. Replacing one or two (or ten) with Gen II modules will likely not fit in the case (I'm not 100% positive of this and that is why we need to solve the physical first. If you have some Gen II modules, give it a try). I know Gen I modules are considered "inferior" but, these inferior modules have lasted 12+ years in my Gen I and I'm confident there are many "good" Gen I modules left out there. To get good Gen I modules, I would contact 3prongpaul, or maybe even ReInvolt (or any salvage yard really). A "bad" Gen I battery pack will have good cells in it. Buy 2 or 3 extra and match capacity. If you follow jdenenberg's posts about charging 3x and matching capacity, you will have a reliable battery (even with Gen I cells) that should last you at least a year. During this time, save your money and buy a new Prius or a new/reconditioned battery (or keep replacing bad cells with Gen I modules.

Additionally, Patrick's advice to not charge above 8.4V is good stuff. 11.X is too high. Stick with 8.4 when reconditioning the modules.

I hope this helps you and I wish you the best of luck. We all have the best intentions here.

 

Thanks, Patrick. I was wondering what the upper end should be. I hope pushing my cells that high has not weakened them further. I know SOC is a range of values, 0% not corresponding to 0 volts. But what should they be? Are they definite values, written in stone by Toyota, or does it depend on each individual cell? What would 6 volts be? Is that a 20% SOC? Would the 8.4 volts you mentioned correspond to a 100% charge? Or would that be 80% or some other number? I can't find this information, and I feel knowing that will help me understand SOC better.

Senior Chief, how do I limit the charging voltage? My chargers are automatically pushing that much voltage into my cells. I don't have any way of having a voltage cutoff when charging (only choice is how many amps and a mAh cutoff). Is my 7250 mAh too high, forcing extra voltage into cells with reduced/weakened capacities? Only the discharge function has a voltage limit. I'm charging at 3.7 amps (less than 1C) because my power supply is rated for 30 amps. Since I've got 2 quad chargers, I figure 8 x 3.7 = 29.6 amps (I don't want to overload my power supply).

As far as the physical, I already ordered two gen 2 cells, but held short of ordering more. I do want to see if they'll fit before I put all my eggs in one basket. I can always resell them if they don't work out. From what I've found online, the G1 module dimensions measure 275mm (L) x 19.6mm (W) x 106mm (H), whereas the G2 modules measure 276mm x 20mm x 106mm. I don't anticipate a workaround to be too troublesome, although I found a student paper online that documented an attempt to do a full 38 swap of G1 to G2 cells, titled "Prius High Voltage Battery Diagnosis and Upgrade," by Timothy Janello. He coded a P3009 HV leak after 5 miles. One of the G2 cells popped and leaked electrolyte, probably due to overheating (and/or overcharging). His conclusion: one cannot interchange cells from G2 into G1. Maybe he's right. Maybe not.

What can I say? I'm hard headed. Where there's a will, there's a way. I will post pics and stats if I can ever get this done.

 

(Patrick Wong):

There is no definitive link between a particular NiMH battery voltage and battery SOC. See for example:
State of charge - Wikipedia, the free encyclopedia
http://99mpg.com/blog/batterypacksexpose/stateofchargesocde/

I would limit the charging current to C/10, or 0.65A, to minimize the risk of damaging the batteries by overcharging them.
How to charge Nickel Metal Hydride Batteries.

There is nothing wrong with using 2G modules in a Classic battery case, but as previously suggested, all 38 modules should be 2G.

 

Excellent. That gives me a good reference. So I guess my 3.7A charge rate was too high. I imagined it was low because others had stated a 1C rate (theoretically 6.5A) was sufficient. What I did not consider was that perhaps my older low capacity modules (sub 3Ah) were being pushed to the brink at over double their actual remaining capacity. Now that really makes me nervous... looks like I'm learning these lessons the hard way (like many before me). I'm starting to see I may just need to buy as many modules as I can afford now. But I will do that, I'll reduce my charging current to .7A (I can't set it to the hundredth ampere) and see if my numbers improve.

One problem I note is: if the G1 cells are 19.6mm wide (744.8mm total width when in battery), and the G2 cells are 20mm wide (760mm), I am left with an extra 15.2mm of battery width I need to somehow cram into the case (unless the 20mm measurement I read was rounded up from 19.6mm). I'm wondering how I might accomplish that... spacers maybe? But then there's the matter of the case fitting over the completed assembly. I can pound the metal out to make room if I have to, but if the mounting holes don't line up, then there's gonna be some more metalworking involved, which will complicate the process. I've seen some posts from years ago where you were interested in the logistics of this (pics never seemed to materialize from posters).

Okay. So using a little more "due diligence," I found a good resource (a "case study") on the THS (Toyota Hybrid System).

The file is called: 20110606092430_HEV_Toyota.pdf

It's really cool and makes me love my Prius that much more. Anyway, the source stated my 2002 Prius (Called a "Prius II" in the file), at a 50% SOC, should deliver 20KW ("discharge power capability").

I know 20,000W = A x V

and

20,000W/A = V

The case study says max current of the battery is 80A discharge and 50A charge (which makes sense-->it should be easier/faster to discharge the battery than it is to charge it--that comports with real world experience of all rechargeable batteries). Divided by 38, that gives about a 2.10A max draw per module.

So 20,000W/(assuming) 80A = 250 V (at a 50% SOC)

... am I right so far?

250V/38 modules = 6.5789V (~6.58V) is a reasonable estimated 50% SOC (assuming full capacity, right?). I know you said SOC does not directly correlate with voltage, but it should in some way correlate with voltage, shouldn't it? Or is it referring to capacity? Maybe each figure for SOC:voltage is different for different battery packs, but how could the voltage stay around 7.2 through different states of charge? That doesn't make sense to me. I'm still trying to understand what value SOC is if there are no other numbers to go along with it. If 8.4v is at the upper end of safe module limits, shouldn't that correlate to 80% charge (or similar high %-age)? I'm just not grasping something. I don't see how SOC could have nothing to do with voltage. Is there a graph or chart you could point me towards that shows SOC vs. time?

The normal operating range (limited by ECU) of the pack is 40-80% SOC. If the max allowable voltage is 8.4v, shouldn't that be 80%? How do I know my cells are at a 40-80% SOC? Do I need a scan tool that reports these figures, or is there a workaround using charge/discharge figures?

 

Ah, I see (I think).

The voltage method of determining SOC is unreliable because a stated goal of battery manufacturers is to provide as steady a voltage as possible throughout the discharge cycle. (Paraphrasing the wiki article)

But then there must be some swing... the 8.4v drops rapidly to the nominal operating voltage under load, which holds relatively steady for a period (determined by capacity, as measured by Ah or mAh), then begins a drop curve fairly rapidly at the lower end of charge. Right? So where does that drop curve occur? Is that the 40% cutoff the ECU mandates? Or does the drop curve begin at a lower SOC than the ECU will allow?

I've observed this in my charging cycles. As some of the stronger modules discharge, they begin to hold steady around the 7.2 mark (give or take). Some last a while at this voltage, others begin to drop off. Do I need to find modules that exhibit similar voltage under load? Do I need to find modules that last about as long at the nominal voltage as the other modules?

 

(Patrick Wong):

1. The 38 modules are connected in series. Hence the current outflow of 80A as cited in the case study applies to every module. Each module is producing 80A (for a very short period of time).

2. Since the traction battery ECU will complain if module pair voltage varies by more than 0.2V, you need modules that exhibit similar voltage behavior as they are placed under load.

3. I am not able to find out a reliable reference that can correlate 40% and 80% NiMH battery SOC to any particular voltage. However I agree that 8.4V is probably around 80% SOC. It is not clear what the 40% SOC level is, but my guess is that the voltage would be higher than 7.2V.

4. Since you already own some 2G modules, I would think you can measure the thickness of Classic vs. 2G modules yourself, and form an opinion about whether there will be a problem mounting the 2G modules in the Classic battery case, or not.

 

(usnavystgc):

It is well documented on this site and I've personally read numerous examples of putting 2g modules into a 1g case. They will fit. The only modes required were lengthening (or ovaling) some of the bolt holes in the case.
My concern about the physical is mixing the two and the problems that might cause but, it sounds like it might work.

Just to make sure, are you keeping the modules under compression as you charge them? If not, they can swell (see pic)

 

... they're in the mail. I'll know as soon as they arrive.

But that info really helps, so thanks again for the input! Trying to piece this puzzle together. If that's true about 40% SOC being above 7.2V, I'm probably in trouble. I have only about 17 modules that should be healthy enough for a several-months-long solution (they hold the nominal voltage the longest).

As far as swelling goes, I am charging them under the tension provided by the battery case. I'm not charging every other, as recommended by Seilerts, but simply doing 8 in a line, then moving down the line until I get to the last 6. I observed slight swelling (small separation at the tops, between cells). I saw this in only a few cells (#s 4 & 6, and #34 & 35), and figure it's due to higher resistance (higher heat) in the affected cells. One cell at the end of the block kept farting, so I know that one's toast (to coin another's phrase, "it lost it's sealed-in goodness"). I think that one was 35.

I have not disassembled the battery fully yet, but will do so when the new modules arrive. Those pics (and ones like them), I have seen before, and they are a constant reminder for me to do this right and "slow down." Hahaha... point taken, Senior Chief!

 

I was considering that. I have some very clean copper pipe on hand (1" variety) that I was planning to use to fabricate the parts needed. I can saw it up, pound it flat, and file it smooth, then drill pro re nata. I also have some tin snips on hand that will make short work of the shaping. I was also considering maybe adding washers to the shorter NHW11 cells, if that would be easier or better.

Those discharge capacities are far above what I'm seeing. Most of my "good" cells discharge from full to around 6 volts in 30 minutes or more, yielding very low mAh rates (~350 and up). My garbage cells discharge in under 10 minutes. I may just be screwed here. My two best cells are #1 & #38, with discharge times of 95:55 minutes (1.5 hours) and 149:15 (2.5 hours), respectively. This was under a .7A load. I could almost cry. I'm gonna cross my fingers and see what happens. Looks like I'm gonna need to bum some money from friends/family to complete this with 38 G2 cells before school starts. That's an extra $700 bucks I just don't have.

Yes, as I understand it, the slight overcharge is to prevent the worst case scenario of a cell falling below the others and reverse charging (thereby destroying that cell/module). What I don't understand is why the chargers say my cells are taking a much higher charge than it says they're discharging. Is that an indicator in-and-of-itself of a bad module? My highest discharge was 1874 mAh (#38).

At the risk of confirming my discouragement:

This is my latest discharge cycle (settings: 1.0A discharge to 6.0v --> each cell only discharged up to a .7A rate, however):

1 - 6.00v / 95:55 / 1213 mAh
2 - 5.99 / 10:05 / 124
3 - 5.99 / 54:20 / 672
4 - 5.91 / 1:18 / 15
5 - 5.95 / 31:37 / 375
6 - 6.00 / 11:53 / 154
7 - 5.96 / 5.52 / 69
8 - 6.00 / 25:16 / 297
9 - 7.44 / 9:57 / 115 +++ (interrupted? restarted-->) 5.97 / 41:09 / 482 (totals: 5.97 / 51:06 / 597)
10 - 5.70 / 00:06 / 00000 (this was 1 of 2 completely toasted cells, registering only millivolts upon initial survey)
#10 self-discharged before I could start the cycle, as did #11 and #20.
11 - 3.81 / 00:00 / 00000 (this was the other completely toasted cell)
12 - 5.91 / 00:24 / 4
13 - 5.99 / 46:58 / 569
14 - 5.97 / 40:47 / 480
15 - 5.99 / 8:08 / 101
16 - 5.99 / 29:24 / 346
17 - 6.00 / 75:56 / 903
18 - 5.97 / 3:21 / 42
19 - 5.96 / 1:13 / 15
20 - 4.51 / 00:00 / 00000
21 - 5.95 / 29:25 / 345
22 - 6.00 / 37:43 / 442
23 - 5.99 / 38:24 / 452
24 - 5.97 / 46:04 / 543
25 - 5.73 / 6:23 / 75
26 - 5.99 / 25:10 / 327
27 - 5.97 / 1:56 / 25
28 - 5.96 / 28:46 / 344
29 - 5.95 / 30:01 / 357
30 - 5.99 / 42:32 / 506
31 - 5.95 / 26:48 / 318
32 - 7.24 / 1:15 / 13 +++ (another interruption?) 6.00 / 18:07 / 218 (totals: 6.00 / 19:22 / 231)
33 - 6.00 / 48:17 / 571
34 - 5.98 / 2:31 / 31
35 - 5.96 / 1:18 / 16
36 - 6.00 / 14:38 / 176
37 - 6.00 / 16:14 / 192
38 - 5.99 / 149:15 / 1874

Those 12 modules highlighted in red are on the "get rid of immediately" list, and there are a few others close behind (like 2, 6, 15, 32, 36, & 37). But they're all pretty much garbage. Look, all I need is a pack that can start the ICE and get moving. I don't need Prius MPG at the moment. I need MPG period. If I can get enough voltage out of the pack to get the ICE started, then the battery can stay in charge mode, with the engine kicked on for the duration of my trips, for all I care. If the system rapidly discharges the lower cells, wouldn't the ICE kick on automatically, and try and juice up those cells before they reach a deep discharge? Sure I may code like hell, but wouldn't that at least get me moving? And are we talking about days of operation here, or maybe a handful of weeks (note my reaching optimism)?

 

Chapman, I really appreciate that thoughtful response and the time you took to write it. I do think I may have "jumped the gun," so to speak. There's quite a bit I still don't understand about electricity and charging (and quite a bit I'm sure I never will). That is the value of a formal education (at least to me): you begin to realize just how little you know. I went to college and came out knowing less than when I went in (though I did gain some wisdom along the way). As an autodidact (like many on this forum), I am learning as I go. Thanks for the advice; it's far more useful than just saying, "go read a book" (like I've heard others on the forum say to other posters).

And John, my numbers are all over. After the first 3 cycles, some went up, then down, others went down, then up, and some stayed about the same. None, as far as I could tell, really got any worse, but there were just inconsistencies that did not add up to any observable rehabilitation. That's why I wasn't really ready to believe my numbers. I'm really not convinced as to the utility of continuing the charge/discharge process, at least for the entire pack. I think I'm burning time there. I may just try and keep conditioning the best cells and order the other modules (still awaiting the first two). My setup is by the beach, in the shade (with a breeze most days), so overheating is less likely, though I may take my Hampton Bay hurricane fan to that.

I ordered two from "Battery Boy" (Ted, in Healdsburg) from eBay ($45 + $8 shipping), but I found some cheaper modules ($33 + $6) closer to my location (Lancaster), which I think I will snatch up after I get my hands on the first two modules. I know the BB cells are expertly conditioned & balanced, but I'm willing to risk the lesser cells, as we're talking a difference of about $10 per cell (which adds up quickly). And Senior Chief, I'm still considering reverting to a G1 setup if I have to. Perhaps a salvage yard, but I have very limited experience with finding parts via this route (did it once a long time ago at a Pull-Your-Part in Sand Dog for my old '88 Accord).

The car drove until it didn't anymore. I didn't anticipate multiple, system-wide failure of the battery, but then, it is 11 years old, with a full year of non-operation. I took such sweet, loving care of this car, I know it will serve me well as long as I can deal with the battery issues (at this point, looking like a DIY full G2 swap). I don't think the cost of a shop-rebuilt replacement is unreasonable (~$1600 to $1800), but I am both a cheap and a skate, and I have the necessary obstacle of limited funds each month (and with bills, pretty much a net zero income/savings). I can't afford to get double the modules and cherry pick the best ones to match to the pack, as advised on the forum. Some day, that should all change, as I'm entering law school soon, but that's another story. Right now, it's top ramen, macky cheese, and hot dogs.