Honda Insight Forum banner

Replaced Battery Sticks and IMA Will no Longer Engage

2K views 10 replies 4 participants last post by  eq1 
#1 · (Edited by Moderator)
Hello everyone. I am new to the Insight community. I hope this is going in the right place. I purchased a local 2001 Insight, and it has been a slow and bumpy process getting it back up and running again.

The previous owners told me they had the hybrid battery completely replaced in 2015 by Hybrid Revolt. These batteries lasted two years, and stopped working in 2017. The owners tried to reach out to Hybrid Revolt, several times and apparently were ignored... by this point, the owners no longer wished to deal with it, so it sat and eventually they put it up cheap, and I scooped it up in October of 2019.

I brought it to a Honda dealership and they dragged me through a series of telling me to change out both control modules on top of the battery (because they were wrong for the year of the car) as well as the DC-DC, and then scratched their heads when none of that worked and said I needed a new battery. I was not pleased with their pretending to know what was going on. The car has been stationary in my garage since. I attempted to grid charge the batteries because I read that this was a common issue with failing packs.

With no knowledge of the insides, I went ahead and hooked up a grid charger to see what i could get. The pack never charged past 110 volts. I continued to try charging and discharge cycles, and eventually I saw smoke and my fan shorted... so I walked away for a few months. I returned to the car in May, and decided to tear into the pack to see what was going on.

When I got inside, I was greeted by corrosion corrosion EVERYWHERE. Every single cell was corroded and most if not all were leaking :). I removed the Hybrid Revolt sticks (their stickers were all over the inside of that pack), and swapped them out with ones i found on eBay. All tested good, and came from one provider.

I cleaned away all corrosion, and replaced a few parts that looked beyond saving. I hooked everything back up. I installed the grid charger so I could have it ready if needed, and it was a good way to see that the pack was reading out good numbers and working. The grid charger showed 160 volts. I thought everything was looking up from here.... then I went to start the car, and I was greeted by a check engine and IMA light after hearing a pop sound come from what sounded like in or near the DC-DC 5 seconds after turning the key.

I attached a google drive link, so you guys could take a look at the video of the key turn and the evidence of a healthy pack. If I need to upload more things to that link, I will. Insight

I took an OBD reading and got P1568. I also read blink codes of 67 and 70.

I think the thermistors may be the culprit... but I can't seem to find anyone selling them, so I never was able to troubleshoot, so the car sits yet again...

I am hoping one of you could point me in the right direction/tell me what you think may be occurring, and perhaps where I could find a dang thermistor harness!

Thanks in advance!

Chris
 
See less See more
#2 · (Edited)
Your 1568 codes are for the PTC circuit and the 'thermistors'. The PTCs are the strips attached to each stick and to the orange end boards; the thermistors are attached to a single cell on each of 4 sticks. Sounds like you might have some botched wiring, loose screws and the like - or maybe the connector on the back of the BCM isn't seated well, check that first.

Otherwise you should take the pack apart and make sure those items are installed properly... The 'PTC' stuff can be bypassed, it's not critical, you just install ~20 ohm resistor in the circuit, remove all the strips from the sticks. As I recall the thermistors can be replaced by off-the-shelf 10K units, can't remember if they're NTCs or PTCs. You'll have to search around here a bit, all this stuff is here, somewhere...

Here's a link to a post that describes replacement thermistors: Battery NTC thermistor specs
I'd make sure the actual thermistors are the problem though, rather than some dodgy install, before buying anything.

Here's a link to a post that describes how to bypass the PTC circuit: IMA battery rebuild help (PTC Strips)
Personally, I would bypass the circuit, including removing the strips from the sticks, regardless of what your problem is.
 
#4 ·
All connections and wires seem to be secure. I just don't have the whole pack bolted down because I figured I would have to pull it out again. I screwed in the ground ones. As stated before, this car was running Hybrid Revolt batteries. There were no PTC strips on the ones I pulled from the pack. Based on that, I assume the bypass must already have been completed? The new sticks I put in have the PTC strips attached. Does this help narrow down the possibilities? I know the wiring isn't botched because the grid charger communicates perfectly with the battery pack. It seems to be the thermistors or something else outside the pack. Is there any instruction on how to install the new thermistors? I have no idea how to approach those. I too want to be sure it is the thermistors.
 
#5 ·
^ So, are you saying that the old Revolt pack didn't come with PTC strips, your new sticks did, and that you have or have not attached the ends of the new stick PTCs to the orange end boards?

Same computers as before?

I recall Hybrid Revolt might've done a stick PTC system bypass inside the BCM, so if that were the case and you were trying to install sticks with PTC strips and attach them, it wouldn't work, you'd get a trouble code, one of those 1568 ones.
 
#7 ·
^ So, are you saying that the old Revolt pack didn't come with PTC strips, your new sticks did, and that you have or have not attached the ends of the new stick PTCs to the orange end boards?

Same computers as before?

I recall Hybrid Revolt might've done a stick PTC system bypass inside the BCM, so if that were the case and you were trying to install sticks with PTC strips and attach them, it wouldn't work, you'd get a trouble code, one of those 1568 ones.
Yes, the old Revolt pack came without the PTC strips. I have attached the new PTCs to the end boards. The Honda dealer had me replace the two computer modules on top of the battery pack with the ones that are supposed to go in my car. I remember them saying the ones that were in there were not compatible with my model year (probably what was needed to run the Revolt system), so i ordered the ones they were calling for off eBay. So the computers should be restored to factory. I can provide the part numbers. I will update my google drive so you can see the Revolt sticks and the part numbers on the backs of the computer modules.
 
#6 ·
Hi @HamiltonCJ,

Welcome to the forum. Your avatar picture is cool. Where are you from?

Awesome that you are digging into the batteries. It looks like you got yourself a pack of original cells to replace the bad ones. That should help. They seem to be designed to tolerate overcharging (somewhat) which I've caught my car doing when it encounters an imbalanced stick pair (for example, after sitting many days and self-discharging at slightly different rates). This seems to be the way the car uses to correct imbalance, but it subjects the cells which fill up first to heat and pressure, and it seems that the aftermarket manufacturers didn't get that message (my personal theory, this is.)

One thing to consider once you have this sorted out, is that some of those cells may have significantly less capacity and/or higher self-discharge rates which after a few days sitting could cause the car to detect imbalance and force the balancing I spoke of above which I believe insults even the good cells and reduces their life. To address this with my car, I am building a Raspberry Pi-based smart charger which can condition two sticks at a time (charge and discharge) while monitoring the voltage of every cell and able to drop a one ohm resistor across any cell for tests and to make sure no cell gets overcharged or undercharged during conditioning, without having to disassemble a stick. (I am using darts on springs to pierce the insulation.) At the end of the conditioning I will know the capacity of each cell and the charge/discharge curve and will be able to select good, matching sticks to make one good pack. Crossing my fingers.

You should be aware of the Github repositories that @mudder created (thank you for doing this, mudder) which have the schematics you need to buzz out the circuit and confirm that your wiring is in fact correct and the service info you need to understand what you are looking at. This will get you a long way not only in sorting out this issue, but in dealing with future ones and engaging in the many projects you will find here.

Please, please please read the warnings in the service manual and wear insulating gloves and eye protection around these high voltages. DC at these voltages is a heck of a lot more dangerous than AC especially at these currents. A tool shorting out 100 amps at 180 volts is essentially going to be welding and throwing hot metal (and it won't quench until it is melted because DC has no zero crossings.)

Looking forward to seeing how things progress, and welcome again.

Sean
 
#9 ·
...It looks like you got yourself a pack of original cells to replace the bad ones. That should help. They seem to be designed to tolerate overcharging (somewhat) which I've caught my car doing when it encounters an imbalanced stick pair.... This seems to be the way the car uses to correct imbalance, but it subjects the cells which fill up first to heat and pressure, and it seems that the aftermarket manufacturers didn't get that message (my personal theory, this is.)
You wrote something similar some days ago and I responded by saying that's very likely not the way it works, and that the voltages you posted don't correspond to full, let alone overcharging... If a cell isn't full until about 1.53V at 6.5A, on the bench, then in general it's not gonna be full at higher currents, in a string/pack of 120 cells with more resistance, at the equivalent of only 1.50V, in the car.

This is definitely true with balanced cells - so the second part is to understand how the car manages the pack with stick pairs, coupled with an understanding of how single cells behave and what those behaviors will look like at tap-level. You'd need to either monitor cells or pull sticks and check cell capacity, etc., to determine if cells are being overcharged or not. And/or you'd need to pull the computers and work with them on the bench, simulate different conditions...

I've done a lot of work with cells and sticks on the bench, with packs in the car, measuring voltage taps consistently, and I've done as much as stomach-able doing checks on sticks and cell states of charge, etc., after different car-usage scenarios. From these I've gleaned as much data and insights as I can to bear on the question of, 'What's going on?'. But it hasn't quite been enough. I'm pretty sure I know most of what's going on, now, but some of it falls short of certainty.

You seem to have the energy and interest to push things farther, so, I hope you do. That's why I'm telling you, if you want to understand what's going on, you really need to develop pointed tests and measures of your ideas...
 
#8 · (Edited)
^hmm, OK then. If you restored the computers to original then you wouldn't have the problem I described - unless something else were wrong.

Now, here's the question: What does the 'temp sensor/PTC harness' look like?

If Revolt bypassed the PTC system inside the BCM, what did they do to/with the harness connection that would normally attach to the electronics-side orange end board? I gather you haven't swapped to a different harness from Revolt to the new sticks/computers, right?

There should be a small bundle of wires each clad in a plastic shroud in bright colors - yellow, green, blue, white, and red. The red-clad wires, two of them, attach in two places on the electronics-side orange end board. This is your stick PTC circuit. These along with the others weave up the 'left side' of the electronics board and go to the back of the BCM.

Do you have two red-clad wires (the wires in the cladding are black and should have ring terminals at ends)? If so, where do they go/attach? If not, that's likely your problem, or at least one of them...

Here's an image of what I'm talking about, some random image I found here at IC:


https://www.insightcentral.net/thre...ltage-on-one-stack.116458/page-8#post-1380242


Oh yeah, also, anticipating future steps, we do need to see the computer numbers and the serial numbers off the sticks you bought (just one of them if they're all from a single pack). I think only the battery computer ('BCM'), the left one, is needed. List them in your next post. The sticks need to be generally matched to the computer.
 
#11 · (Edited)
^ I don't think our problem is a definition of "overcharging," though there can be slightly different definitions of that...

On the other hand, your personal definition might not jibe with real-world application and observations - Insight NiMH cells can be charged at a constant current and take a long time to increase in voltage, yet they're still charging. It can be hard to distinguish even if watched over a long-ish period and graphed. So, if say you're watching your tap voltage monitor and you see voltage seemingly plateau over fairly steady input current, you could pretty much forget trying to distinguish a full cell from that info alone...

Overcharging can be hard on cells, there's no argument there...

Practically speaking, in the car, I'm almost certain there's no danger of overcharging if you let the BCM do its thing. The only way to come close is if you manipulate the system, say with an OBDIIC&C, resets, and purposefully try to stuff the pack. But even then, I'm pretty sure (which is slightly less than 'almost certain') any overcharging, using the definition you posted above - charging a cell that's already full - is still minimized by the BCM. It will use both max tap voltage and slope detection to determine when a tap is full. It will throttle regen and then disable it entirely once a tap reaches I think around 17.4V at rest or if tap voltage during charge starts to increase fast (i.e. slope detection). The same thing it does at the bottom I'm pretty sure it does at the top...

Just the last few days or so I've been working with my pack 'at the top'. Tracking amp-hour input, I know about how charged the cells can possibly be. At the same time I'm trying to charge as much as possible in the car (within the time constraints I have in my life, that is). The BCM makes it very difficult, it's just programmed to throttle/disable charging before the cells are totally full.

Oh, and I know they aren't full because, as I said, I've tracked amp-hour input, but more importantly, because I know how balanced the cells are, so tap voltage can be used as a proxy for cell voltage - and the voltages I see when the BCM tries to throttle and disable are not full voltages... IF you have balanced taps/cells, and you see about 184V at pack level, or 18.4V at tap level, during about a 6 amp charge, then your cells are very nearly full. Try it. Tell me if you ever get close to that... I think this is probably a bit conservative, too, as this assumes you have decent cells; most used OEM cells would probably be even farther from full at that voltage, as they are deteriorated and have higher resistance, so their voltages peak higher at a lower charge state.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top