Honda Insight Forum banner

21 - 40 of 52 Posts

·
Premium Member
2001 5S "Turbo"
Joined
·
10,925 Posts

·
Super Moderator
Joined
·
8,493 Posts
Here is a screen capture from a road test. The top row is the peak voltage seen for each stick pair.
What sort of device are you using to gather this data? The display is unfamiliar to me. Looks like it shows max., min., and rest voltage for each stick pair?
 

·
Registered
Joined
·
883 Posts
It does sound like you have a cell in there with a capacity problem.

The Prolong test should be interesting. Keep us posted.

Let us know when you start looking at things at the cell level. This is where I'm currently building tools.
What sort of device are you using to gather this data? The display is unfamiliar to me. Looks like it shows max., min., and rest voltage for each stick pair?
It's a homebrew stick pair voltmeter that uses Bluetooth to ensure that dangerous voltages never leave the IMA enclosure and an Android app for a modern display.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #24
I used the method supplied by Willie Williford , I shorted the two pins and got a blink code of 78. OBDII showing on the Scangauge is P1107. 78 is battery deterioration on the blink code sheet. Seems appropriate.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #25
An update on my driving experience here. I hadn't driven since my last post. Went out today to run to town, about 8 miles up a long hill for the first 4. I did not clear any codes nor reset the IMA light, just went for a ride. I had absolutely no assist, nor did any charge register during the entire ride, so I'm assuming the IMA was not functioning at all. At the top of the hill on the way to town my car battery light illuminated as well as the brake warning light on the dash, never observed that before. Kept driving and those went out in a mile or so. On the way back I was thinking I would get some regen happening coming down the big hill, but nope, nada, zilch. I'm doing a reset and grid charge tomorrow as I am planning on driving it to Waimea Thursday to the dentist. About 50 miles one way.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #26
Forgot to add that I put the Maxx Volt grid charger on the pack last night and read 157 V. I've reread this entire thread, now with a bit more in my noggin to work on. With no decrease in the BATT gage and no power at all it would seem to me to be one or more cells are really weak, so even grid charging appropriately will only go so far if the discharge curve or the low cell is effecting the low side capacity. I'm thinking I've got to "stretch" the low side of the pack, and if I understand what eq1 suggested by finding the individual cells that are low and discharging the others slowly down to the low level maybe I can achieve some sort of balance. I have received the Prolong system, but not the OBDIIC&C from Peter, so I will just grid charge today for my trip to Waimea tomorrow, and when I have received the OBDIIC&C I will get into the pack and install both. I know this is not the cell level process that eq1 recommended, but maybe with the capability to discharge the pack strategically I may achieve some success. If I am totally off the rails here with my perspective someone please reply and set me straight! I have no objections to being straightened out with all of this.
 

·
Registered
Joined
·
7,061 Posts
^ I wasn't recommending a cell-level regime, exactly. One thing I suggested, the last thing, is based on stick pairs, or 'taps'. The pack is made up of 10 stick pairs. The battery computer manages the pack based primarily on voltages it reads at those taps. You can access those at the rear of the battery computer (BCM). It's the silver box 'on the left', on top of the pack.

The tap harness/wires are protected by little orange nubs on the orange end board - they're positive temp. coefficient resistor things that act like fuses if they heat up. When they cool back down, things go back to normal. The upside of this is that these tap PTCs can be used as a gentle discharge load for stick pairs: You can short the terminals at the tap connector on the back of the BCM, 5 taps at a time, every other tap, and the resistor/PTC in each tap will discharge the stick-pair it's connected to (there's actually two PTCs per tap).

On the PTCs I've measured, the resistance has been about 190 ohms, x2 = about 380 ohms. When you short the terminals, a given stick pair will discharge through that 380 ohm load; assuming a discharge voltage of about 15V on average, the discharge current would be 15V/380ohms=0.039 amp, or 39 mA. A full stick (or stick pair) might be about 6000mAh, so discharging at a 39mA rate will empty the pair in about: 6000mAh/39mA=154 hours, or about 6 days...

In the car, sticks aren't charged to completely full, so it'd be less time. If, on the other hand, you've grid charged, some cells could indeed be full... I usually estimate a discharge of about 14% per 24 hour period.


Given your situation - you've grid charged, so some cells could be near full, your 78 blink code, which is a P1449-78, which means less than 10% capacity, your short-lived success, which suggests fast self discharge in at least one cell - you basically almost certainly have some near full cells and some near empty cells. That's not good, that's really bad and hard to work with at the pack level. If you try to grid charge - which at this point I advise against - you end up seriously overcharging probably most of the cells. If you try to do a full-pack discharge, you end up seriously reversing the most empty cells. Neither of these are good.

So I would suggest you look into the tap-level discharge. You'd need to measure voltages at BCM connector C, note the high and the low, and start discharging the high, perhaps in increments of a day or two at a time and see how it goes. This should 'free-up' some room at the top so you can get some charge into the least-charged cells. On the other hand, if you do have a fast self discharge cell or more, it probably won't help much if at all. For that, the only thing I've found that can reduce self discharge is a super-deep discharge, which means something like every cell discharged to about 0.5V at very very low rate, 39mA is probably fine...

But, we might be getting ahead of ourselves. I think I'd try an incremental approach and see where it gets you.

Here's a link to a post or two that talks a bit about tap-level discharge, with a diagram of what needs to be shorted. Keep in mind that only a few people have done this, so you'd be kind of a guinea pig if you did it. I've hocked the method here and there but it has never really caught on. I've done it probably about hundred times or more though, usually not super deep, and initially I did some tests (like measuring the temp of the PTC resistor things during discharge), never had problems.

https://www.insightcentral.net/threads/p1449.125324/#post-1449272
^^ NOTE: The text in that post talks about an 'ultra-deep' discharge. Kind of ignore that. We're not talking about doing ultra-deep here, rather, I'm saying use the tap discharge to try to get some balance in your taps, first. Bring the highest taps down, probably by at least 30%, so a couple days of shorting per set of 5 taps.

Also, I mention measuring voltages during the process and using them as a guide to how long you should work. I wouldn't bother with that. I mean, you can, but at this point your voltages are probably wacky anyway, so they wouldn't be very strong indicators of anything important.

Just think big-scale and conceptually: You've got some taps that are charged way higher than others, and that's one main reason why the car won't charge the whole pack enough to function. So, you need to bring those down.

I shorted the two pins and got a blink code of 78. OBDII showing on the Scangauge is P1107. 78 is battery deterioration on the blink code sheet. Seems appropriate.
That's interesting, an unusual trouble code. Note that if your barometric sensor is indeed faulty, I'm pretty sure it will manifest in unusual IMA performance, as in it could take a lot more throttle to invoke assist. I've experienced changes in IMA behavior such as this at high altitude, going back to normal at sea level.

Here's some text describing that code, from the pdf service manual:

DTC P1107: Barometric Pressure (BARO) Sensor Circuit Low Voltage

General Description

The barometric pressure (BARO) sensor is built into the engine control module (ECM) and monitors atmospheric pressure. The ECM estimates appropriate intake airflow from the manifold absolute pressure (MAP) sensor output voltage and BARO sensor output voltage. If the BARO sensor output voltage is a specified value or less, the ECM detects a malfunction and a DTC is stored.

Malfunction Threshold
The output voltage from the BARO sensor is 1.58 V or less for at least 2 seconds.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #28
OK. Actually I mis-wrote about the cell discharge regime. I did mean the pair discharge. So you're saying that if I continue to grid charge now, with seriously depleted pairs in series, that I risk overcharging the good pairs before I ever get any meaningful charge from the weak pairs, correct? And this is even using the Prolong system with the OBDIIC&C to obtain a greater discharge. Given this I will discontinue the grid charge I am doing right now, and wait to get the IMA opened up and the Prolong and OBDIIC&C connected so I can at least get some more meaningful data. I think I get what you are suggesting doing the discharge at the tap level at the BCM. Thank you for all your assistance here! Guess I won't be getting 60 mpg to Waimea tomorrow!
 

·
Registered
Joined
·
7,061 Posts
^ That's about it, but when you charge a pack it's just a string of 120 cells, so the "pairs" concept doesn't come into play.

You've already tried grid charging a couple or few times, I don't think you're gonna get anywhere else doing it more. And, yes, you risk seriously overcharging already charged cells in an attempt to get more charge into the depleted cells. If you were to discharge the whole pack, you'd risk seriously reversing the depleted cells in an attempt to...get whatever you can from that discharge process.

You don't need the Prolong and OBDIIC&C to do the tap-level balancing discharge thing. You need some paper clips and a multimeter/voltmeter. Measure voltages at the taps once you get the back open and report back. Here's a link to a post on how to do those measurements: The quintessential Insight NiMH voltage thread
 

·
Registered
Joined
·
36 Posts
Discussion Starter #30
I removed the cover on the module this afternoon. Looking around and getting familiar with things and one of the first things I noticed was this...I'm assuming that the "CV" is referring to the transmission. I have a MT. Obviously they are different or they wouldn't need to be labeled. Can anyone tell me what the difference might be and if this difference would have an effect on the issues I have noted above? I did get the OBDIIC&C PnP equipment from Peter this afternoon and the MCM is showing a "MT", so that issue if it is one should be moot.


78DCD546-42D9-48D8-A97E-5CEA61FD6B42_1_105_c.jpeg
 

·
Administrator
Joined
·
12,317 Posts
Is your car a CVT converted to a manual many moons ago?
More likely someone has been fiddling and put in a CVT MCM for some weird reason.

Luckily they are basically interchangeable.
I've used both types of MCM in both types of car for long periods with no apparent ill/adverse effects.

IIRC The CVT Module limits assist when starting from standstill to spare the start clutch.
Not sure if I ever tested or proved that. It may have other nuanced or subtle altered behaviour.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #32
I am fairly certain this car has always been a manual, and more likely has had the module replaced, probably as a complete swap of the entire battery module. I have a second battery module sans the MCM, so maybe that is the original, but really who cares, eh? I will be messing with this new fangled project some this weekend hopefully.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #33
I wanted to update the forum on my progress with my car. I purchased the OBDIIC&C from Peter, and also a Prolong charger/discharger kit. Got everything installed over the weekend. Here's what I am intending to do: first to use the OBDIIC&C to limit the SOC to 75%, which I have done in order to minimize the possibility of overcharging of the "good" cells/sticks/pairs. Then when I have the pack charged (and at this point I may have to stop the charge and lower the SOC% setting as even 75% may overcharge something), and let the pack sit for some time, I will measure the tap voltages. IOW eq1's suggestions, I will short and discharge the highest reading pairs, and measure again. Then discharge the pack to 92V, and remeasure the voltages at the taps to see what's what, particularly interested to see what the shorted pairs read. Recharge the pack to the appropriate SOC% and remeasure and repeat. At some point I'm hoping that I will start to see some balance and if so lower the discharge voltage limit. I figure the tap measuring at the top and bottom should provide something to chew on. Right now the SOC is 24.9% and the battery voltage reading is 160V. I have no idea if these numbers are in line with other packs readings or even if it matters.
 

·
Registered
Joined
·
7,061 Posts
^ You shouldn't need to do anything special to prevent overcharging in the car, the car (the BCM and MCM) handle this on their own.

Is your IMA working at all right now? If so, I'd suggest you:

-drive it (use assist a lot) until the BAT gauge drops to the bottom ('neg recal'),
-immediately reset state of charge to 75% and don't use assist or regen
-go home and measure your tap voltages
-report back

This should give us the best chance of identifying the tap that's causing the empty behavior (neg recal) - and thus, let us know which tap not to discharge, at least not yet.

This could take some iterations and time, so if time, etc. are an issue, you might be better off ... I don't know, saving your pennies and buying a new pack?
 

·
Registered
Joined
·
883 Posts
I kind of like the idea of limiting the voltage excursions in the car because a common theme, it seems, is that shallower cycles return an increased equivalent number of cycles. But it will probably require the cells in the pack to be periodically conditioned - something I'm working on for my car but not reasonable for most people.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #36
eq1, in the other instances where I attempted essentially what you are suggesting above, i.e. using assist until the BAT gauge drops to zero, has resulted in the IMA light illuminated and no assist nor charge, but, the BAT gauge stays fully illuminated. So what does that tell us? I can certainly do the other stuff, well I'm not sure how to deliberately not use assist or regen, but I can drive home and measure tap voltage. I'm grid charging right now, just went to check...the charger says 168V 347mA, OBDIIC&C says 160V and SoC at 24.9%, so the voltage crept up and the SoC hasn't moved. That seems weird to me, the voltage (charge I thought) increasing but steady SoC. Should I continue charging see if the SoC increases, (i think now with the OBDIIC&C I can monitor the SoC to failure), do a hard reset (meaning disconnect 12V, pull fuse and relay), drive it till it runs out of assist and come home and measure taps, or just terminate the charge and drive till no assist? Suggestions?
 

·
Registered
Joined
·
7,061 Posts
in the other instances where I attempted essentially what you are suggesting above, i.e. using assist until the BAT gauge drops to zero, has resulted in the IMA light illuminated and no assist nor charge, but, the BAT gauge stays fully illuminated. So what does that tell us?
Nothing really. It's not typical though. I think we established earlier that you got a P1449-78, that's really all we need to know to move forward, I think.

I can certainly do the other stuff, well I'm not sure how to deliberately not use assist or regen, but I can drive home and measure tap voltage.
If you have a manual transmission, installing a 'calpod' clutch switch would be useful, not necessarily now, but at some point. If MT you can shift to neutral more often. You can try to time your neg recal so it's close to home.

I don't know, I can't tell how much functionality you have at the moment. Normally it'd be easy to do these things. It's like, you just use assist and the pack goes empty, the gauge falls to the bottom, you reset with the OBDIIC&C, maybe charge a little bit to keep it from going totally empty the rest of your drive, that's it...

I'm grid charging right now, just went to check...the charger says 168V 347mA, OBDIIC&C says 160V and SoC at 24.9%, so the voltage crept up and the SoC hasn't moved.
Yeah, this isn't normal, your SoC should be increasing. Sounds like you have some computer warble or something. Perhaps a reset, pulling fuse or neg 12V cable, is in order...

Should I continue charging see if the SoC increases, do a hard reset (meaning disconnect 12V, pull fuse and relay), drive it till it runs out of assist and come home and measure taps, or just terminate the charge and drive till no assist?
Well, I thought I suggested earlier not to grid charge again - and here you are grid charging again? I can't help you if you don't follow directions... My whole 'program/plan' was spelled-out in post #27, and a bit more below that. I can't say any of it now more clearly than I did back there.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #38
Yes you did say not to grid charge, however that was before I had the OBDIIC&C to limit my SoC to prevent overcharge, and because of that tool I was in the process of implementing the procedure I outlined in #33. When you suggested above that I should drive the car I kept the charger on as I need some charge to have any assist, and I was interested in observing the SoC to the BatV relationship. The question I was asking was how much was enough, your reply implying only enough to get it to neg recal. OK that's what I'll do next, I'll reset everything and drive it, now at least I'll have the OBDIIC&C to monitor SoC and other parameters. I'll just stay close and limp home and then measure the tap voltages.
 

·
Registered
Joined
·
7,061 Posts
Sounds like you're misunderstanding where the risk of overcharge exists. It's not when you're driving the car, it's when you're grid charging. If you have near-empty cells mixed with near-full cells, and you try to 'fix' the pack with a long grid charge, you end up overcharging the near-full cells in your attempts to bring up the near-empty cells... I don't know for sure, but my guess is that, say, 24 hours of overcharging say three-quarters of your cells is at least as likely to harm pack operation as it is to help it. I've never overcharged cells that long...

In my opinion it'd be better to try to bring the 'high cells' down first, which we can at least attempt to do with the tap discharge process...

My suggestion that you drive, use assist until neg recal, and then measure tap voltages is meant to put your taps in a weak position - we'd prefer to measure tap voltages when the 'problem' is close at hand, not after, say, you've grid charged and 'artificially' inflated the voltages. So basically, whether it's driving until neg recal or whatever - we want to measure the taps when you're having the problem...

Another alternative if for some reason we can't 'get there', might be to hook up your Prolong discharge load and then measure tap voltages, with the load in place. That will put some stress on the taps and help bring out the weakest... In fact, we might want to do that regardless. Unloaded, 'resting' voltages are hard to interpret, they can all be about the same even if there's major problems...
 

·
Registered
Joined
·
36 Posts
Discussion Starter #40
I understand the basic risks of grid charging, and I thought to minimize those by setting the SoC to 60% using the OBDiiC&C. I just returned from trying to get the pack to the point of neg recal, and try as I may I was unsuccessful for whatever reason. Left home and drove along the coast road, monitoring the SoC , BatA and BatV. SoC started about 70%, moved up to a bit over 80 as the regen cycles bumped it up, which I don't understand as I had the SoC on the OBDIIC&C set to 75%. Then I went up the big hill, and because I was deliberately trying to get the SoC to drop drove as I normally would drive the hill. Watched the SoC drop from say 75% to below 50, also the BAT gauge dropped to around half, so there seemed to be some correlation as I would have expected. I didn't want to go dead far from home so I turned around about 2/3 up, maybe 5 miles. This is a long steady grade. Coming back down the regen brought the SoC back to low 70's. I should add that during the entire excursion the BatV on the OBDIIC&C would fluctuate, 145 to 170 depending on the assist/regen applied. Got to the bottom and since I had 75% or so I went back up, this time driving more aggressively, (at least for an Insight, LOL!). On shift changes I would go up to 3/4 bars on ASST, Amps 30-35, geez the car felt great. I got up to say 60 MPH and at this speed drove right to the top of the hill and beyond, SoC got down to low 50's, not as low as the first hillclimb.. Anyway to shorten this saga up, I basically repeated this 3 times and drove home. It's like there is no problem with the pack at all. The only piece of hardware that was changed was the MCM that I obtained from Peter with the OBDIIC&C. Could that have been the culprit?

Given this it doesn't look as though I'll "get there" as you say, so I guess I'll try measuring the tap voltages with the discharger running... that to me seems a really good idea, as the pairs would be under a constant equal load. But what do I know anyway?
 
21 - 40 of 52 Posts
Top