New-to-me IMA pack
 

Scored a 115K mile IMA pack from an '03 HCH on eBay last night. You could have knocked me over with a feather when I found out I had won it with just $250. I hope I haven't bought a pig in a poke.

I have reason to believe it has its own problems, but since my IMA light has been on for about a year - the grid charger isn't helping anymore - it may allow me to kill that idiot light for a while.

And then, while I have a spare pack, I can work on more fully refurbing my existing IMA pack to get it to a much more serviceable condition. Because for all its faults, I strongly prefer driving the Civic over the other vehicles I have to choose from.

Nice find. Planning on sharing the process with us?

You could always test the individual sticks, and make one pack from the best ones.

Absolutely. I'm anticipating it arriving sometime next week, and I don't think I have any volunteers scheduled until after Thanksgiving. The swap is a simple process, and switching the grid charger over will be a doddle.

That's the plan. I've seen the occasional box of sticks - untested - for sale but for some reason the complete IMA felt like a better deal, especially at this price.

Okay, some updates:

The IMA pack finally arrived from the seller in Ohio. He packed it very thoroughly and everything looked in excellent shape. No pictures of the box or the unpacking, though I made my Dad who is visiting open the box. Dad custom-makes boxes specifically for when he ships things, and while his stuff always arrives in good condition they're notoriously difficult to open. It kept him busy and happy for a few minutes.

Son #2, I and Dad all spent a few minutes last night swapping the packs. Mostly it's just cumbersome, heavy parts in cramped confines.

Hooked up the grid charger and left it on overnight. Pack started at 155v.

Unplugged grid charger this morning showing 175v on the input. That's a bit lower than I hoped.

Started the car. Fired up on the IMA motor, so that's good - it means I remembered to turn the pack's breaker ON.

12v charging warning light was ON. Hmm.

IMA SOC array bottomed, but climbed to full after a couple of minutes.

IMA light was still ON.

CEL light was still ON.

I'm not surprised about the CEL light. I have a cooked O2 sensor somewhere and insufficient kung fu to figure out which one. I'll take it to a shop eventually. Maybe ream out the EGR valve too. I can do that at home though.

I had hoped the IMA light would go out after a few minutes. No such luck.

The 12v warning goes out above 1400 rpm. This gives me some hope, it suggests the IMA pack is simply too out-of-balance to allow the DC-DC converter to do its thing. Another night of grid charging may wrench it back to where it belongs.

I'm reading everything S Keith has to say about it, him and Snikrepretep. I hope I haven't bought a pig in a poke.

You likely have some bad bacon there. It's 12 year old IMA pack. Pretty much a certainty that it's a dog, but the price was right for the risk. I've kept my mouth shut to prevent a jinx... :)

You need to read the IMA codes. For your DC-Dc to only work in the 1400-3500 (or so) range, this pack is probably in pretty bad shape. There could be something else wrong with that pack. Getting the codes read will help diagnosis.

As always, make 100% sure your 12V is healthy (Get it load tested) and your grounds are good. Probably not an issue in this case, but it's always good to confirm it when chasing IMA issues.

Your first grid charge should have been for 24 hours or peak (175V) for 8 hours. If you did this or close to it, a second grid charge very unlikely to fix it.

If you're not discharging your packs, I recommend you do so to this and your original pack. Your original pack is likely suffering from massive voltage depression due to overcharging and may have suffered capacity damage simply from the excessive number of trips to 100% SoC via grid charging.

I recommend the following for both packs:

1. Grid charge to peak voltage for 8 hours, not to exceed 24 hr.
2. Discharge to 135V with 2X 250W bulbs in series.
3. Discharge to 60V with 2X 60W bulbs in series.
4. Grid charge for 2 hours.
5. Attempt to Car charge at idle. If the IMA light remains lit and it refuses to charge, abandon effort and revert to grid charging. If the car will charge the battery, attempt to charge for a total of 20 minutes or so. It may take multiple IMA resets to force charging for that long. You know you're done when the charge cycle terminates very quickly and the SoC gauge goes to full.
6. Grid charge for 8 hours if you were able to charge the battery with the car or for 24 hours if not.
7. Discharge to 120V with 2X 250W bulbs in series.
8. Discharge to <12V with 2X 60W bulbs in series.
9. Repeat charge cycles above (5 & 6).

I recommend dual bulbs because I have popped a lot of single bulbs, and it's really annoying.

The car charging is optional; however, there is evidence to suggest that higher current charging from a deep discharge can help reactivate the cell materials further improving capacity and lowering internal resistance.

I normally recommend a single deep discharge to <2V but given the age and history of your original pack with its existing failure to respond, I am reluctant to do so.

Given my personal experience with Insight and HCH1 sticks, it's unlikely that 20 of the 40 sticks you own will perform well enough to make a functional pack. Stick level work is very time consuming, and doing it right requires high load equipment and test method consistency.

The good news is that the pack deep discharge process essentially does exactly the same thing as most stick refurbishment techniques, so you know that your sticks will be as healthy as possible after these deep discharges, so no refurbishment will be necessary, and you will merely need to devise a means of sorting.

If these deep discharges of the pack don't work, I recommend the following stick sorting method:

Make yourself a nice table to record all the data.

Buy something like this

• Grid charge the pack to peak voltage for 24 hours.
• Disassemble and let the sticks rest for 24 hours after grid charge.
• Measure and record resting voltages. Sort by voltage descending.

--------

• Starting with the highest voltage stick, attach the tester
• Attach a separate voltmeter
• Activate the tester until one of the following is reached:

1) 30 seconds is achieved, record voltmeter voltage and a time of :30.
2) 5.4V is achieved on the voltmeter, record time and 5.4V

Repeat the above load test for all sticks.
NOTE: This will be hard on the tester. You may want to wait a few minutes between tests and apply forced air for cooling between tests.

Repeat all of the above for the other pack.

Select the sticks that went for :30 without hitting 5.4V and then select the sticks that went the longest before hitting 5.4V. The best 20 are the ones you want to install in your pack. Sort them from best to worst (highest to lowest V at :30 and longest to shortest time to 5.4V). When assembling the pack, pair the strongest with the weakest within a given pair as that will minimize the voltage drop for each tap and report the most balanced voltage to the BCM. The voltage taps are on the orange plate to which the big orange cable is attached. Pairs are the adjacent sticks between those wires.

This gives you a means of quickly and cheaply testing all your sticks to determine those capable of delivering high current. The chance of success is low, but this method has you with only another $25 invested and about 4 total hours of bench time.

Good luck,

Steve

EDIT: REMEMBER, READ THE CODES FIRST, AND CHECK THE 12V/GROUNDS BEFORE ANY OF THE ABOVE! :)

This all sounds very achievable and I VERY much appreciate the guidance. I now only have to contend with:

1) Dad's in town. He lives 10 hr. away and is gettin' up there in years. He wants to go to the mountains.

2) My wife is saddled with a large event for the homeless shelter we work for, and her 40+ volunteers who signed up all buggered off, leaving her with only 3 for tomorrow and none for Sunday.

Yeah, "buggered off." I could say what I'm really thinking but asterisks just aren't sufficient. I'm not feeling very charitable toward the vols right now.

3) I get pretty depressed when it doesn't work the first time.

It barely got nine hours on the GC total in the first place; I will give it 20+ and see how it does. If that craps out, I'll dig deeper into it. I have a car to fall back on, and Son #2 can drive the truck.

Nevertheless, I will keep trying. Again and very sincerely, thanks for the help.

Given #3, in retrospect, I should have spoken up sooner to keep your expectations and disappointment in check.

DEFINITELY go longer on the GC!

On the bright side, you got a killer deal. Since you already have your old pack out, build yourself a discharger:
https://dl.dropboxusercontent.com/u/...120_164528.jpg
and discharge it. The above is about $4 from Hope Depot. The board is one of those screened fascia pieces for ventilation. Yank the screens out, and you don't have to drill the 2" holes in it. They're $0.98 each.

If you measure voltage and current, you can estimate your pack capacity.

Good luck!

Steve

Okay, the longer GC didn't work. I left it on the charge for 23.5 hours. GC voltage output topped out at 180 for a while, but then dropped back down to 175-178. Output current never started to taper off. Looks like the pack's a bad 'un. I knew it was a gamble.

I have other motors to drive while the Honda is benched, so I'm going to:

Build a discharger and sock down the original pack. For future clarity, Pack 1.

The Honda is mobile with Pack 2 but the 12v light coming on is a little unnerving. I think Sweetie is happier driving a different car rather than rolling the dice. I tell her it's not much of a roll but frankly I can't blame her.

On the upside Dad and I went to Cades Cove in the Smokies and got to see the sorghum sugaring demonstration. The last time we saw that, he was younger than I am now, and I was...hmm...seven?

Yeah, seven.

Also consider that there are many uses for the battery sticks that don't work in a pack. For instance, I have an old battery-powered hedge trimmer that would run maybe a minute on its original pack. Soldered in a couple of leads, duct taped a couple of sticks to the handle (to keep it nicely balanced), and it's run all summer on one charge. Not that I use it that much, but there were 3 or 4 half-hour or so sessions...

Sorry to hear it didn't work.

That's a very high peak voltage, but low temperatures can get you there. The drop is a little surprising. Were you running the fan? Are you using a Hybrid Automotive or a home-built? If it's home-built, and you're not running the fan, be very careful. Even the typical 350mA of a grid charger can cause a battery to cook if the ambient temp is high enough.

Output current doesn't generally taper unless you hit the power supply set voltage.

Just make sure you to a good solid grid charge before you deep discharge. That will ensure that the fewest number of cells will spend the least amount of time in the reversed condition.

Indeed. I have 20A charger/dischargers that can discharge 300W regeneratively. That means they drain the test battery into the battery that is powering them. I have 4 strapped together (2 sets in parallel, then in series) that I use as a regenerative sink for testing other batteries at 20A discharges.

Quick update
 

Went to Ace and spent $20. Picked up a couple of lampholders and incandescent bulbs.

There are no incandescent bulbs in my house except two places: the fridge and the oven. And when the bulb in the fridge dies, it'll get an LED.

I got 130v rough service bulbs to provide a bit of overhead. I hate when bulbs blow. Wired them in series as recommended.

Checked starting voltage: 154.8v

Hooked it up, fired it up. Very gratifying: they lit right up.

I saw a couple of VERY involved setups that incorporated built-in meters and automatic shutoff. I suppose if I concentrated very hard I could figure out how to do that, but what the heck, it's Sunday.

So I set a timer for 10 minutes and fired up Halo. Every ten minutes I hit PAUSE and check the voltage.

Finally, at 70 minutes I got down to 143.8v DC on the terminals.

I am pausing here while I try to figure things out. I have seen mention of minimum cell voltage of .78v, beyond which you risk reversing the polarity of cells. In my pack would mean a total voltage of 93.6 - not having disassembled the pack, I have no way of knowing whether any cell has gone below this threshold, of course, so I'm giving myself plenty of wiggle room before I do something irretrievable.

UPDATE AGAIN: it's another hour since I wrote the above and the voltage climbed back to 145. I put the discharger on it for another 10 minutes and pulled it back to 143.5. I'll see what it's relaxed back to in the morning.

If you follow my suggestions, you should be fine. Your pack is broken. You're not going to make it any worse.

You WILL reverse cells. Period. The key is to do it at low current and in stages to minimize time and duration at reversed polarity. This is why I suggested 250W to 135V, then 60W to 60V. Of course, this is after a grid charge, and your starting voltage doesn't look very grid-chargy.

0.78V is the threshold at which voltage depressed cells will yield their remaining capacity. Powering through this level at low current generally restores high-current capacity.

Steve

EDIT: I've run IMA and Prius packs down to < 2V TOTAL voltage at very low current (<200mA). You're not going to hurt anything. The folks over at Insight Central are doing this routinely. Reversing 1-2A consumer grade cells will kill them. Reversing 100A industrial cells at very low current does no measurable damage.

BTW... The concept of X Volts per cell goes out the window below 1V/cell. When you go <1V per cell, you can more accurately describe it as "Pack Voltage = # of cells that haven't reversed".

Okay, that's very different from what I had anticipated. I thought when I saw the 2v description...

Okay, that was a headslap moment. I just realized that if it meant 2V per cell, the pack voltage would be above 200. I can be monumentally stupid sometimes. D'oh!

All right, back to the store for smaller bulbs. This evening (hopefully, Son #1 has a super-late class on Mondays, we carpool, and I get home completely ragged out) I'll pull Pack 2 from the car so I can remove the GC preparatory to beginning the recharge.

Don't sweat it. The concept of taking a 144V pack to 2V is hard to wrap your head around if you haven't been exposed to the concept and practice in the Insight community. It's particularly concerning when

Any progress?

Thanks,

Steve

Progress Report
 

I wired up two 100w bulbs in series to load Pack 1. The car is benched while I discharge the pack, even though I could probably run it safely on Pack 2. It makes my gut clench a little every time I see the 12v light come on, so I'm doing without my sweet sweet economy goodness for the time being.

In other news I'm using Torque to chase my Forester into the 27.x MPG range, which is crazy great for that thing.

I will stop by a store this evening and pick up a couple of larger wattage bulbs. I may die of old age and/or boredom waiting for the 100's to get the job done. I pull it down to 143.x and in the morning it's at 144.x again, and I can only do a couple of hours of draining per evening, and the voltage rebound between drains is mocking me.

I was going to have Son #2 screw in the smaller bulbs and just let it be going while I'm at work, but he appears to have left for class already, so that's an opportunity lost.

If I'm still at this on Thanksgiving Day, I'll be able to give it a lot more time and attention while helping in the kitchen. The rig is set up on the kitchen floor (though if I'm still at it on Thanksgiving I'll move it to the sitting room).

That's the downside to dual bulbs in series. They keep bulbs from popping, but 2X 100W are like a single 50W. Since you're already at nominal and not terribly higher than 120V, you're mostly out of the bulb popping danger zone. You could switch to a single 100W, which would be like 2X 200W down to 135V. Then switch to 60W to 120V, then dual 60W to 60V.

Don't fret the voltage bounce. It's not an indication of lost progress. It's the robust NiMH chemistry getting back to the nominal 1.2V/cell after the load is removed. Voltage bounce peels away pretty quickly once you resume discharging, and it adds a minimal amount of total time to the overall discharge. Note that when you switch bulbs to lighter loads, you will see a notable voltage bounce even after the load is applied.

Good luck!

Steve

Another update
 

Happy Thanksgiving everyone!

And to our non-American friends, happy, um...November 26. Thursday.

Finally pulled the pack down to about 55v. Took FOREVER. Wrestled Pack 2 out, wrestled Pack 1 in. Hooked it all up.

Removed the pack bolts, wrestled it halfway out, hooked it all up. Looked at everything two more times. One more time for good measure. Reinstalled the pack bolts.

Plugged it in. 181v, 0mA? Whiskey Tango Foxtrot?

Son #2: "Hey, I turned the switch back on." 125v, 348mA. Man, I'm fatigued. It's the little stuff that gets you and the damned IMA is ALL little stuff. And Son #2 builds computers because it's fun, he's really good at catching details.

S Keith says give it a couple of hours on the charger, then fire up the car and go for a spin. So after Thanksgiving dinner, I'll take a little cruise and see what happens.

No, S Keith said to charge for two hours, start and let it idle charge for 15-20 minutes, resetting the 12V battery as needed to get there.

Spin comes after the car terminates idle charge shortly after charging.

And happy t-day to you too.

Steve

Another update, another oversight
 

S Keith keeps giving advice and I keep not following it, or not all of it. My Dad insists I'm ADHD but I think he's full of hey, squirrels!

After letting Tiptoe sit on the charger for 2 hours, Son #2 and I took it for a quick spin. LOTS of depth of charge has returned, even with such a short charge aboard. Regen and assist behaved properly.

A quick note about assist: under the old paradigm before the discharge, I had about ten seconds of full assist before the IMA would go into recal. It would recal multiple times per trip. I got a grand total of about 20 seconds of full assist (and lots of lesser assists) on this little trip and it hasn't recalled at all. So on that basis alone I would say this has been a success.

But the 12v light came on. That was worrying. And it stayed on at all speeds, which worried me even more.

I brought it home and did a few 12v resets as described, pulling the negative, waiting a couple of minutes, reconnecting and starting to let it run for a few minutes. No luck on the 12v.

In fact on the 5th 12v reset process the car was struggling, wouldn't rev above 1200 or so, and the instrument panel was flickering. Shut it down and fetched the meter to check the battery. Metered at 11.45v. It was still daylight then.

Fast forward to ten minutes ago. Read about another tinkerer wandering through the maze of Honda hybrids, he mentioned the 100a fuse on the connections side of the IMA pack. Well, I just happen to have a spare here, let's tip that pack up on its side and look at it, shove that plug out of the way...

hmm...

that plug...

I remember what a challenge it is to get that out, but do I remember what a challenge it is to put it back? Unholster the flashlight and pop the panel out of the way. NOTE: knowing myself a little too well, only the battery pack is bolted down. Every other cover is still loose.

No, as it turns out I do not remember plugging it back in, and for good reason. Plug it in.

And the car fired right up, 12v light is out. So that's one hassle less. If nothing else I have my car operating in condition similar (actually markedly improved) to before I started all this.

Good to hear there's been some improvement.

I recommend another 12V reset and charge to full SoC.

Key off.

Charge the 12V battery and grid charge IMA battery for 8 hours.

Do not reset the 12V again.

Grid charge and discharge every oil change.

Steve

One last update...
 

My IMA light has gone out. The code has cleared.

It had been staying off for as long as 20 seconds after first turning on the engine, but on Saturday after Christmas, it stayed off.

This may be because the battery has equalized a bit more, but I suspect it probably has more to do with me having left the grid charger plugged into the battery. When I went to throw some luggage in the trunk for my annual post-Christmas cabin vacation, I unplugged the GC and stuffed it deeper into the corner, out of the way. Whatever the case, the light's out and my IMA is assisting like it did when I first bought the car.

In other news, Bumblebee has very graciously agreed to purchase the replacement IMA I had bought on eBay. If they ever wanted to cement my business in the future, that was the way to do it.

Eli is always looking for core packs, and he'll pay a fair price for them.

Thank you for the follow-up. Glad to hear you're seeing improvement.

Keep a close eye on the SoC gauge. Look for recalibrations. When they start to get frequent, do another charge/discharge cycle OR do one every oil change (1-2 per year).

Steve