[gobuchul74]

OK, Well I recently swapped out the IMA battery pack with a used one from HAPS recycling. It worked fine for about 2 months, but yesterday the IMA failed again. This time it happened much more suddenly. There was no gradual battery problems like before, the IMA and CEL lights were on immediately and I have no IMA assist.

I still have my old IMA battery pack and was thinking about swapping cells bewteen the packs in order to get one working. I'm wondering if anyone has any suggestions. I do forsee a couple problems that I'm not sure how to work out.

1. As I understand it, the computer will fail the IMA if it detects different voltage levels between cells. I think any swapped cells would cause this to happen immediatley.

2. How to detect the bad cell "Stick"? I think I would have to rig up a way to charge/discharge the sticks individually to check them out.

Any suggestions? Thanks.

BTW, Anyone interested in a used Insight?

 

[flunkysama]

When the IMA light comes on, it indicates a problem in the IMA system, not just the IMA battery. Since there were no symptoms like before, the problem could be somewhere else in the IMA besides the battery. If you have a service manual, try reading the IMA error code like I described here.

 

[gobuchul74]

Thanks, I'll give that a try. I have the service manuals so the code will help. Also, I haven't tried it yet, but I'll try clearing it to see if it resets or not.

 

[Wayne M]

dissasemble and test

If you get past finding any of the modules bad, or other items, I'd guess the best try would be to pull out the latest battery pack and open up both. Then test each 'stick' for indicated voltage, mark each one, then assemble a pack with the sticks that come closest to the manuals minimum stated voltage variations. That's the only way I could see to get one out of two. That would also weed out any 'dead' sticks if it came to that.

What year car was the 'donor organs' removed from, and your car as well?
Just curious on that count, relating to my own recent IMA swap. (2000 mine with '04 donor)

 

[Armin]

matching battery modules

For the purpose of this post: module=one stick of six cells welded together.

I don't think a simple voltage measurement is the best idea to find matching modules to make a good battery pack out of two bad ones. Since you don't know the state of charge of either pack, much less each of the modules, the voltage is meaningless.

To find matching modules, you really should compare their capacities first, and resistance second. This is not trivial, but since you seem to be comfortable swapping battery packs on your own, you can probably do it.

I would determine the capacity of each module by charging it with I20 (325mA) for 24 hours. Then discharge with I5 (1.3A) until the module voltage drops to 7V. You could save time by charging an entire pack at once, but you have to discharge each module on it's own. Otherwise you can only run the discharge cycle until the first module goes empty and you won't be able to determine the capacity of the other ones.

To measure resistance, I would charge the module as above, then allow it to rest (maybe over night). Apply a load of I1 (6.5A), maybe even more, for - say - 1 minute and measure voltage again. Disconnect the load, wait for another minute and measure voltage again. The difference between the two voltage measurements, divided by the current is your resistance.

Note that the resistance determined will depend heavily on the method used (timing and load current), so be consistent.

It would be very exciting to do a full characterization of two battery packs this way. I would be very curious about what you find! When you disassemble the pack, make sure you mark the location of each module. Heat is distributed unevenly in the pack and it would be good to know if the weaker cells are all located in one area.

Good Luck! I hope you find the time to do this, it would be a great service to the Insight community! I'll be happy to help with advice as much as I can!


PS: how much you want for the used Insight?
(you are not serous, are you?)

 

[Rick]

Call up HAP and ask if they offer a warranty on their parts. Most places like that have at least some sort of warranty on their parts.

As far as swapping out cells I'd see if any of the "sticks" have lower voltage than the others as a start.

 

[gobuchul74]

Well, I've got the pack apart tonight and removed the sticks. So far all I've done is measure the voltage of each stick after removing them.

Unfortunately, I didn't check this board before I ripped into it, so I didn't take all of your advice Armin. I did not mark each sticks location in the pack as they were being removed, but I did notice that the placement of the sticks had a definate effect on their voltage reading.

All told, the voltages ranged from a low of 6.10 to a high of 7.98 (If I remember correctly). Right away I noticed that the cells at the immediate front and rear of the pack had high readings. The bottom layer of cells also had some of the higher readings as well. Any stick in the top and middle row that was not at the extreme front or rear of the pack was much lower.

I must say I have been an electronics tech for 13 years, but taking this pack apart made me nervous. I was pretty happy with how it came apart though, I think the pack was well engineered for safe maintenance (by cautious techs). It was a pretty logical disassembly, no tricks to it.

I think I'm at a temporary stopping point until I figure out some way to charge and load test individual sticks. Any suggestions? Maybe a variable power supply, something current limiting... It seems like this is going to be very tedious, 20 sticks....

 

[Yves M.]

I know that ther are a lot of them but if you have a standard D cell NiMh charger, you can use it to charge the batteries.

Then you can check the resulting voltages. Maybe use some kind of discharging appliance like an electric motor with a speecific amount of time to get the capacity

I figure you will need to extend the batteries with a wire to get them connected into a charger. I have a 4 D batteries Nimh charger used to charge a lamp. (120 cells at 4 at a time mean to do it 30 times (outch)
Something like this that I have (no intent on advertizing, only a google search):
http://www.nimhbattery.com/ap-accupower20.htm

 

[IamIan]

battery

gobuchul74, good work ... thanks for your efforts....

If I may be so bold when you put the pack back together and put it back into the insight could you try it charged even if only partially charged.... to see how the IMA electronics deal with a Battery SOC different from a previous SOC??? Either way thanks...

You asked for stick charging ideas.... I may not know nearly as much as many people out there but I'll give my 2 bits.....

You could use a home charger Designed for 6 D cell NiMH batteries.... connect a wire / dummy to each cell possition .... Wire the charger output in series since the stick is already in series and use that to charge the 6 cell stick.... You might also be able to test / Measure the amount of energy that the Charger is putting out to the stick as a total series voltage and AMPs if you have an AMP counter or the charger may have a rated power Input etc....etc.... this would let you possibly use the deltaV / deltaT already built into the battery charger....

You could just put together a regulated voltage & amp power supply for a known period of time baised on the charge rate the battery capacity.

Or Built a Smart charger from the ground up.....

Or By a Smart NiMH Battery Charger Deisgned for 7.2V / 6 cell and 6.5 AH battery packs... like those used in robots and or remote control cars...

The Insight batteries can do 50+ AMPs so charge rate should not be an issue....

Best of luck...

my 2 bits

 

[gobuchul74]

I picked up a programmable charger at a RC Hobby shop and I'm putting the sticks through a discharge/charge routine. Then I imagine I'll do some kind of test to see how the cells handle a load.

Here's some pictures... The Auto Battery charger is to keep the 12V car battery charged. The car battery is powering the programmable charger... just in case you were wondering...

 

[Dougie]

Nice pics. Can you find where the cooling air for the battery comes out? I'd like to put an external, non-modification-requiring temperature sensor on the output air from the battery compartment...

 

[gobuchul74]

Sure, You see the fan on the battery module? Thats where the air comes out. It is ducted in through the forward side of the module and is pulled out the rear by that fan. From there it's no longer ducted so you'd have to have the sensor on the interior side of that fan.

 

[Armin]

Cool battery charger! The picture is a little fuzzy, can you tell us make and model? How much was it?

 

[gobuchul74]

It's a Great Planes® Triton DC Computerized Peak Charger (Thats a direct Cut N Paste off the web)

It cost me $120 at the hobby store, but I think its worth it.

Let me tell you what I've got it doing and tell me if I should be doing it different.

The charger is putting each stick through a Disharge @ 3.0A/Charge @ 5.0A for two cycles.

After each stick is complete, I am recording the mAH's disharged and mAH's charged, as well as the average voltage level. All this is recorded by the charger.

I've done two sticks so far. It actually takes about 6 hours. And it seems that the initial disharge cycle drains thee stick after about 1300 mAH, then charges the stick for about 90 minutes totaling 63-6600 mAH. Then next disarge cycle drains the stick of about 5500 mAH. Then it charges again.

I started with what I think are the good sticks. I think what I'm looking for is a stick that has a much lower discharge rate on the second cycle.

 

[Rick]

Nice pictures! Hope things go well. While you've got everything cracked apart, could you possibly jot down the specs on the panaflo battery cooling fan for me? I'm interested in the model number to find out whether they used the high of low speed fan.

 

[Armin]

Let me tell you what I've got it doing and tell me if I should be doing it different.

You are doing great! I need to look for one of those chargers. What a great tool!

I guess I'd be more conservative on the charge current (0.65 A maybe). But then, it would take a month to go through 20 modules! Anyway, it doesn't really matter, as long as you are consistent (use the same method for all 40 modules). Then just pick the 20 with the highest capacity and assemble them into a new pack... voila new Insight battery!

Good Luck!

 

[gobuchul74]

Panaflo DC Brushless
0K06AN-2C
MODEL FBL12G12H
DC12V 0.6V

Thats the exact markings on the cooling fan.

Well, My progress is going slowly. I figure I can only check 3 sticks a day, as it takes about 3-4 hours to do each one. So far I have had one unusual reading. The stick I hooked up before I went to work appearently peaked at 12-1700 maH charging and only discharged 10-1100 maH. This was one of the sticks with the highest initial voltage when removed from the pack. Compared to the others so far the discharge capacity is extremely low. I'll try it again when I finish the rest just to make sure, but I don't see how I could have screwed it up. The charger runs the charge/discharge cycles by itself.

I guess this is going to take a couple weeks, but I'm really starting to feel confident about getting this back together. We'll see. I really appreciate all your advice and support. Thanks.

 

[Rick]

Thanks for that. Now this is kind of wierd, I can't really find that fan listed with panasonic. There are fans similar, but they have no designation for the 3rd letter of the model number being L. Originally I was thinking maybe they used a lower speed fan since I can't ever hear it, and I was going to replace mine with a faster speed one to cool better. But even though I can't seem to find that part info on the fan, it's a high speed version of the fan. They do have an ultra high speed available too. Hmm the hunt continues.

 

[Rick Reece]

Rick,
I think the Honda computer must have a very high set point for the fan speed to switch to high as in 5 years I never heard mine until I got MIMA. Mike put extra temp sensors to monitor the battery and DC-DC converter and they trigger the fan speed. Now if it is hot or I am using the battery a good bit I can hear the fan especially in autostop when there is no tire noise. The other item to note is that the fan housing doesn't fit perfectly (There is a gap). Mike uses some electrical tape to seal the housing so that the intake air is not drawn from the output side. Have fun, Rick

 

[Mike Dabrowski 2000]

The fan cranks pretty good when turned fully on, but what I found that was wierd was that the fan has slotted sides, rather than the smoothly ducted fan sides that are pretty much standard with every other fan I have seen. Since the fan is set up to exhaust, and the air path is so long and obstructed, I would have expected a sealed blower like the MPI has instead. The fact that the fan shuts of in autostop tells us that there was a decision made that having silence during autostop was more important than the battery cooling, which I think is a mistake.
Sealing the sides of the fan to fan enclosure stops air from being sucked in the sides and forces all the exhausted air to be drawn through the battery case.
It will be interesting to see what Highwater measures when He adds his booster fans to the front inlet vent, that should help increase the air flow a lot.
I still like the AC outlet vent to battery cooling vent duct as the best turbo cooler for the pack. Now where did I put that aluminum flex duct?? :wink: