here i go again. IMA acting up

Discussion in 'Honda Hybrids' started by laurieaw, Oct 23, 2007.

  1. Right Lane Cruiser

    Right Lane Cruiser Penguin of Notagascar

    For what it is worth, my understanding tallies with Wayne's description. Essentially, deep discharge cycles tend to create compounds which do not break apart and add internal resistance to the battery. This is what lowers energy density. Keeping the ions separated as much as possible (battery fully charged) slows the process enormously -- though age will still cause a slow accumulation of such compounds.

    Once those are at a sufficient level, regardless of memory effect, your battery becomes effectively useless because its internal chemistry is "clogged."

    If the BCM is out of whack and not correctly determining how much charge is remaining, or how to effectively condition the battery, this might explain the "runaway" behavior we are seeing?

    With several hours of travel time on the highway coming up, if I see such an event I'll have the time to see if it will eventually recover, then act normally. The inconsistency is what makes me suspect the electronic components -- though battery aging could easily be a factor.
  2. laurieaw

    laurieaw Sorceress of the North

    gentlemen, you have gone way past my knowledge base. my eyes are glazing over as i read this, but i know that between you all, you can help figure out what's going on. i can be very mechanically proficient in some areas, but not this. the left brain kicks in and i get lost......

    thanks for all the input :)
  3. Chuck

    Chuck just the messenger


    How far down would you consider safe on the Insight battery pack? My guess would be the safe SoC would be 20-15 bars.

    I'm glad the temperatures have cooled - no longer have to worry about my battery pack cooking in the parking lot. :)
  4. xcel

    xcel PZEV, there's nothing like it :) Staff Member

    Hi Chuck:

    ___On your Insight, minimizing pack use (assist or regen) throughout your normal drive while leaving the SoC high other then the few times a year you may be climbing mountains would be best. I think you already know the best way to climb a mountain with a 5-speed Insight but that is for another thread.

    ___For Laurie, I just want to see your HCH-I’s recovery after a Forced recal and how she acts afterwards is all.

    ___Good Luck

  5. Mike Dabrowski 2000

    Mike Dabrowski 2000 Well-Known Member

    "___The real world evidence is in with 100’s of Insight packs being replaced that were cycled from one end to the other. This has lead to shorter life spans as has been seen in just every battery study ever written."

    I would really like to know where you got that information.

    Ask any EV guy about battery cycling and pack balancing, I am not making this up.

    For one thing the packs are not cycled from end to end, the BCM does its best to keep the SOC between 20% and 80%. good for 100K cycles
    When you don't use the pack, it develops a memory just like a power drill battery that has been used sparingly, the 6.5AH can become 1 AH. Cycle them a bit, and they will come right back.

    Recals are when the SOC estimate gets skewed, and it keeps hitting a lower voltage limit prematurely since it never gets properly charged.

    I realize how hard it is for some hypermilers to exercise their motors and electronics.

    It could ruin their MPG. I am only offering my opinion for what it is worth.
    Send me those links.
  6. laurieaw

    laurieaw Sorceress of the North

    am in correct in interpreting this as it needs to be pushed to the limit at times? i am not asking this in a sarcastic manner (see my last post) but asking if this is what you are saying?
  7. xcel

    xcel PZEV, there's nothing like it :) Staff Member

    Hi Mike:

    ___You know where all the packs are failing unless you have not been reading the boards the past few years? Feel free to look them up for yourself whenever you find the time. Feel free to let us all know after you stopped counting at 100.

    ___Cel balancing is a good thing and why the cel’s in an Insight pack were cap matched to begin with which helps improve longevity and reliability of the entire pack. Any time a cel or pack gets “whacked”, a bit of its life disappears “forever”. The deeper the cycle or higher the C-Rate, the worse the damage.

    ___As for memory, Ni-MH’s have far less memory then a Ni-CAD IIRC and a Prius is proof in practice. Honda opened up the throttles on the DoD and C-Rates and Insight owners have been paying the price for years. An Insight has the most delicate pack of all the hybrids sold in the US given the accounts of replacement packs. Hammering away at them only reduces their cap and their life all that much sooner.

    ___Every touch of the pack takes life out of it. Every hard hit of one takes a lot more life out of it. Thus the reason to be careful on any accel to minimize assist and the long term damage caused by continuously going to the well. We all know the beast way to drive a hybrid for FE and most reading this now know the best way to destroy one. You know which hypermilers to contact wrt recals that finally brought in the IMA code of death. In fact, many actually hope for it so as to get the new pack.

    ___In Laurie’s HCH-I case, it sounds like the BCM is FUBAR and with that, the pack is probably beat up somewhat. We can only hope to nurse it back to health w/ a forced recal and pray it does not happen again?

    ___Laurie, a battery does not need to be pushed to its limits with a SoC draw down test for overall cel longevity. It is a bad thing to be playing that game given deep discharges are killers.

    ___Good Luck

  8. Chuck

    Chuck just the messenger

    If I can summarize:

    • Mike says drain the battery pack every several months
    • Wayne says never drain the battery pack - keep the SoC near the max at all times.
    Hopefully that accurately states the two positions.
  9. xcel

    xcel PZEV, there's nothing like it :) Staff Member

    Hi Chuck:

    ___I am saying never, just when the opportunity forces it. If never, good but if you travel in heavy traffic and are not really careful or any hilly area, it is going to happen.

    ___Mike loves running high C-Rates including deep DoD’s at any time and we all know what that does to a cel’s chemistry let alone a pack. Minimize the hit and the longer the life.

    ___Good Luck

  10. Chuck

    Chuck just the messenger

    I was trying to get some clarification.

    The last few weeks, I've gotten down to the last 1/3 SoC more than Mike would probably care for...
  11. xcel

    xcel PZEV, there's nothing like it :) Staff Member

    Hi Chuck:

    ___As we have talked previously, if it were my Insight, I would want the pack to last the life of the vehicle and the one way I know how to make that happen is by minimizing both assist and regen events and cracking the windows in any type of open exposure sunlight so as to keep the cabin temp too not much more then a few degrees above ambient. If you were to drive down SoC for P&G, get it down there and leave it down there for the entire period you would be playing in that realm just as we did at the Insight Marathon Attempt. For a Tour De Sol type rally, you may burn 10 bars over 5 – 7 miles of a mountain climb but that happens once and she will come back and hold at 19 of 20 for all the easy driving highway and descent stuff afterwards. Every day driving, stay out of it and leave it high. You have already lost one pack and let’s try and keep it to one because I doubt the next one will be free.

    ___Laurie, sorry this thread is not going in the direction you originally intended :(

    ___Good Luck

  12. MnFocus

    MnFocus hanging member :)

    Missed this post Jeff . Fortunately I do not work in a 'float' plant ( Hades is cool comparatively). We did have a tempering oven as well as a chem temp tank- super heated sodium chloride til it turned liquid- at one time though . Plenty warm .I've worked in this automotive replacement glass manufacturing facility since 1986. That will be changing in the next week or so . Cullet ...millions of tons of it .Yup indeed. Kind of hard to dismiss 20+ years as a 'mistake' - but it was. ;)

    Sorry to hijack laurie :eek: back to your regularly scheduled IMA infomercial (I'm learning ! :flag: )
  13. Mike Dabrowski 2000

    Mike Dabrowski 2000 Well-Known Member

    Yes I am suggesting that giving the pack a good exercise by driving it hard may improve things. Drain it by heavy assist till it limits at the bottom, and let it regen to its hearts content to refill the pack. You cannot push it beyond the built in safe limits.

    If hitting the pack hard has not caused my pack to misbehave after at least 45K of MIMA, and a total of 122K, and many other people using their pack lightly, are having problems much sooner than that, would at least be one case where Waynes assumptions about battery use have proven incorrect.

    Remember the design of the Insight and Civic included the IMA, and the design of the IMA assumed that people would use it. When you do not use it, your behavior brings you outside of the design envelope. We can only make assumptions about how the car dealwith that.
    M Santos confirmed in another thread that the Civic will dump charge and boost assist when the pack is too full, just like the Prius, and unlike the Insight. It would seem logical that they would also have worked in an occasional battery extended cycle like the Prius to erase any developing memory effect, which while being less than NICAD's is still a problem.
    I don't want to get in an opinion contest with Wayne over this, we have disagreed on this since before MIMA, and are unlikely to agree now.

    Yes many Insight packs are being replaced, but what is really wrong with them?
    Did the way they were used have any effect on their premature failure?
    Can a simple conditioning cycle or two make the pack behave better?
    How long will the re balancing correct a flaky pack?
    Not enough information for a definitive answer in my book.

    That is what I want to finally determine, rather than make assumptions.

    Again good luck with your visit to the service department, and I would not suggest that you try the hard driving until you see if they can find anything.
  14. xcel

    xcel PZEV, there's nothing like it :) Staff Member

    Hi Mike:

    ___I am glad you directed your response towards Laurie and no, beating the hell out of the pack is the wrong thing to do. Force a recal via the 12V and watch the results closely should help or at least temporarily fix this problem.

    Yes many Insight packs are being replaced, but what is really wrong with them?

    Ask Honda, given they are replacing them to the tune of $1,500 - $2,000 OEM with the Inverter, BCM and pack.

    Did the way they were used have any effect on their premature failure?

    It was either heat or use. Driving Bradlee’s Insight across the country and back while managing the pack properly (minimal to no assist and regen’s accel’s/decel’s) saw no recals over 2,500 miles including climbing the mountains of New York in the Tour De Sol even though his Insight was prone to Recal’s.

    Can a simple conditioning cycle or two make the pack behave better?

    No. Do you think Honda would not try and limit the $’s and reputation fall out with a procedure so simple. New pack rebuilds with some new cap matched cel’s (Honda’s method) don’t recal right off the lot that I have read until they are beat up a little again.

    How long will the re balancing correct a flaky pack?

    Not long if at all. “There was a member using your whack pack procedure who said he has “fewer” recals.”

    Not enough information for a definitive answer in my book.

    ___Fortunately, Laurie owns an HCH-I and they have had far few instances of this problem then the Insight’s. I doubt her HCH-I needs a pack due to damage given the Make and model hybrid, year, low miles and she hasn’t been racing it other then snow tires and a few miles of gravel road travel but hopefully we can help find out what the story is this coming week.

    ___Good Luck

  15. Chuck

    Chuck just the messenger

    I have been cracking the windows while in the parking lot in hot weather.

    Again, I think both Wayne and Mike would agree in the last month I have visited the last 1/3 of my SoC too often - I'm getting better at keeping the decent to about 1/2 of SoC.

    Funny aside - I'll be hypermiling in the 45-55mph range in the rightmost freeway lane and lately, I've enountered more drivers on the entrance ramps trying to use the shoulder while doing 60-70mph. {sigh} Tonight I managed to thwart such a driver and the SoC only went down from 20 to 18. :D

    One thing I've noted is charging/discharging tends to heat up the battery pack...therefore I have used light Assist, although I will be heavy on the Charge.
  16. Mike Dabrowski 2000

    Mike Dabrowski 2000 Well-Known Member

    Honda does not allow dealer techs to touch the pack innards, so replacement is their only option.
    The subpack by subpack re balancing is not easy, the pack must be pulled, and disassembled, and the cycling takes several hours per subpack, or more than 50 hours total. Replacement by the dealers is their only option, given the price of labor. The modules are changed to get new MCM and BCM firmware which many have indicated is much different in IMA usage.
    When Honda takes the old pack back and rebuilds it, we really do not know what they do with them do we?

    A Pack whack test while being the first procedure in the history of the honda Hybrids that did essentially erase a repeating recal condition, did not fully charge any lower capacity subpacks since they were in series.
    We need more people to try rebalancing to get more experience.

    The individual subpack conditioning is the next best thing to rebalancing each cell with a fixture (we could do that as well).
    You are making sweeping assumptions and stating them like they are facts.

    Always better to look at this type of problem with an open mind, and do the necessary test and analysis to make a definitive diagnoses.

    The results of our testing will be made available to all.

    I am finished on this subject, and again hope that laurieaw has good results at the dealer.

  17. msantos

    msantos Eco Accelerometrist

    Wow, nice discussion.

    I can see some opportunities for confusion particularly regarding what is best for our NiMH packs. As I read through the posts I do see some pretty good info that is pretty valid and maybe we can put it all together in a way that is neither complex nor does it trivialize what is going on in terms of battery management patterns.

    Based on some of the info I have access to, I then will make a few cold statements first:
    - To ensure long life, A battery pack must not be subjected to frequent large oscillations in its charge/discharge curves.
    - A consistently topped off battery is just as undesirable as a consistently depleted battery.
    - The ideal voltage for a typical cell is usually determined by the combination of the nominal charge and discharge curves. So in a series arrangement like our battery packs are, this maps into a now typical SoC range of 60% to 80%.
    - Contrary to popular perception NiMH chemistries have been found to still suffer from some memory effect. Although not as severe as that of other/older chemistries (NiCad) these are still observable especially when the deltas in charge/discharge rates are small and persistent.
    - Battery management routines and patterns are nothing to sneeze at. They can be very complex and depending on how well a manufacturer/designer understands the profiled field/test data these can take far more parameters into account than what is commonly considered. As we know, temperature, SoC, Charge and Discharge rates are both inputs as well as conditioned outputs. (I believe Toyota has understood and mastered these ealier than Honda).

    Now, if I understand Mike correctly, what he's saying is that incurring "strategic" deep discharges may have a quantifiable benefit. Am I right Mike?
    Well, based on what I see as part of the current pack management routines & patterns in addition to the many profiling results released by the industry, I believe there's a great deal of merit to this.
    I would say that the key to this function lies in the "strategic" nature of process. It cannot occur often at all and is must be evoked strictly to contain the cell's afinity to slowly and viciously degenerate to a level below its original nominal capacity. Although I have yet to see such a deep cycle discharge occurrence in either of my two HCH-II's or in the 3 Prius-II's I bought for my company, I believe this process is now a critical part of the battery management/conditioning routines.

    It is conceivable that the HCH-1 and Insight may implement a far less aggressive set of management/conditioning routines simply because Honda had yet to learn from the data it is now still collecting. Things get even worse for models equipped with Manual Transmissions as the computerized management control is subject to the input provided by the driver as well as the driving conditions, thus denying the possibility of a more controllable battery conditioning strategy. So maybe carrying out Mike's suggestion on these models may be a key? Sadly, I cannot find any info regarding this issue for older hybrid models. I'll keep looking. Until then experimentation may hold the answer for folks like Laurie.

    More on Manual Transmissions:
    It is on this basis that I agree, that equipping a hybrid with a manual transmission is a bad design choice and I am happy Honda will make MT equipped hybrids a thing of the past (sorry I do not mean to offend anyone here as the MT models have been the FE champions).
    Honda now knows and admits that it was a "mistake" of their part and that is why they will usually offer little resistance to a pack+ BCM replacement for MT models. If I were to own such a model I would certainly exercise that option ASAP as soon as I notice behaviors similar to what Laurie reported. After the replacement though, I would baby that pack as if it was gold. ;)

    I should also add, that even the CVT equipped models are not totally free of ealier battery failures especially if they are driven in harsher topographies. Honda has acknowledged this too and has stepped to the plate on this as well by offering pack replacements for these owners too. Although the HCH-II's pack is managed far more agressively than the prior (IMA1, IMA2 and IMA3) platforms, Honda knows that some people living in tougher topographies may be the ones checking in first with battery failures for these models too. The question now seems to be how quiclky that will happen. It remains to be seen if the current management/conditioning formula will accommodate the std warranty period or not.
    Even for CVT models, the problem when operating in these topographies seems to be the frequent Top offs (descents) and Discharges (climbs) that mountaineous drivers seem to experience on a regular basis, again violating one of the rules listed above.

    Now, I also see a great deal of merit in Wayne's assertions. Honda has a knack for implementing and deploying these assist routines that appear poorly balanced with the regen patterns.
    In other words, we have the high afinity assist kicking in for even the slightest gas pedal input. In a regular driving session this creates a barrage of micro hits on the battery pack (which although not harmful by themselves since they are of low amplitude) happen to not be balanced by a reciprocal and more frequent regen pattern. And when a managed regen does occur (inevitably), it occurs a little too late IMO which causes a greater deflection in the state of charge. Maybe I am missing something here, but most techs I talk to seem to think this is still an area for improvement as well. It remains to be seem what additional future tweaks Honda still has in store for us HCH-II owners.

    So I agree with Wayne that minimizing the frequency of the hits on the pack has much merit and if the assist patterns are too agressive then perhaps the owner should take control or balance the process through whatever means possible.
    For instance, on my wife's car I often try to place the car into hidden regen as a means of minimizing the amplitude of the swings in SoC. I have almost given up controlling the assist part since Honda's latest software updates seem to have more control over the system than what I assumed originally.

    Anyway, I hope I have not fumbled things even more. But if I did please point it out and I'll do my best to explain.


    Last edited: Oct 25, 2007
  18. Mike Dabrowski 2000

    Mike Dabrowski 2000 Well-Known Member

    Nice overview.
    Only a couple of points.
    The information I have seen on SOC range for the Honda Insight was 20% to 80%.
    Is the civic range different?

    Since hitting the pack cannot go beyond those limits, and the limits have been determined to give the best compromise of available capacity and long life. It is not really a hit at all, rather it is a normal short cycle.

    Looks like a good data acquisition system to log battery cycle depth and frequency would shed more light on the normal behavior than any of our speculations and assumptions will.

    We should have more real world feedback on the effect of pack rebalancing and the effect on recals over the next month or so.
    I will wait to make my final determination until after we get the results.
  19. msantos

    msantos Eco Accelerometrist

    Indeed it is. Occasionally, the system will let the charge go down to 2 out of 8 bars but those incursions occur rarely and at rather strategic times (for instance, when it is too hot) - it almost amounts to hitting two birds with one shot (apologies to bird lovers go here).
    While the typical owner will freak out and wonder what's wrong, the event is definitely managed and should be regarded as part of the normal pack maintenance routine.

    However, the "normal" operating charge range is 60% to 80% for the newer IMA platforms. I believe IMA3 is also a little more stringent in complying to these thresholds as well.

    I suspect the 20% to 80% range for the Insight was derived from an early and rather academic interpretation of the charge/discharge curves for the NiMH chemistry. It is my belief that those specifications were precocious and did not (and could not) benefit from the experience that Honda would eventually acquire.
    Again from an academic view, the 20% to 80% basically clips off the "danger zone" ranges of the curve (at the top and bottom) but leaves very little head-room for deviations and other high tolerance scenarios.
    Since heat is still a monumental problem, "particularly" during the charge process and high discharge rate scenarios it makes sense to shrink and slide the operational window further up the curve, just for the sake of ensuring a longer life.

    The info I got concerning the battery recycling program is that used "battery samples" are forwarded both to the client (Honda & Toyota) and to its manufacturer (Panasonic & Sanyo) for subsequent analysis. I do not know what the statistical facts are in regards to the sample sizes and eventual fate but I have been assured that they offer much value to the client as they do to the manufacturer. I do not know if after the pack and module disassembly the complying cells are fed back into the system as part of replacement packs or simply recycled.
    As engineers we often frown upon re-feeds, but I guess we'll never know what additional data they have and what kind of financial pressures exist that will make that acceptable. ;)


  20. Mike Dabrowski 2000

    Mike Dabrowski 2000 Well-Known Member

    I am curious as to where you got the 80%-60% SOC information for the newer IMA systems.
    Thats only 1.3AH, compared to the 3.9AH that the Insight may use from the same batteries.
    Boy have we deviated from the original thread subject, sorry to have shifted the focus.

Share This Page