Honda Insight Forum banner
21 - 38 of 38 Posts

·
Registered
Joined
·
7,746 Posts
Eq1 I believe you are correct with that - I do believe they take more to charge and don't give the same percentage back,,, once charged i believe they would give more back then a low pack though would they not? so I was kinda thinking yeah it takes more - but you do get a bigger punch back when taking off and such...
I tried a bit to answer this, but it got too complicated. There's just a lot of stuff that goes into understanding what's going on. In general, I get 'more punch' out of my pack at low 'charge state' vs. high - the operating voltage tends to stay higher, higher longer, and is more consistent (I can get consistent, repeatable high performance). When I go high, performance deteriorates pretty quickly... There's probably a host of reasons why this happens, some related to the NiMH electro-chemistry, some related to the condition of my pack as a whole, etc...

I was re-reading some stuff and ran into this post, linked below. This and maybe some others in that thread might help. There's also another thread, 'the quintessential NiMH voltage' thread, that goes on and on about this stuff - often not in definitive ways, but it's where I've asked all the questions and have tried to find answers over the years. I think there's maybe two more that I tend to revisit, contribute stuff, all slightly different focus but generally the same overall questions...



edit: Here's a snip from one of the earlier posts from that thread linked above. It's one of my earliest tabulations of OEM battery management efficiency (which is akin to keeping pack at high charge state) versus manual methods, which at that time for me was mostly just avoiding high charge state, rather than purposefully using really low:

Really, when it comes down to it, there's only 3 key data points - the beginning, the middle, and the end. The beginning and the middle form the start and end points for the first management scheme: OEM. The middle and end mark the start and end of the second management scheme: MANUAL.

Over the OEM management interval, the amount of 'excess' current needed to maintain state of charge was 12,346mAh.

Over the 'MANUAL' management interval, the right half of the chart, the amount of 'excess' current needed to maintain state of charge was only 1,440mAh (roughly, both of these figures).

12,346mAh/1,440mAh=8.6 -- the OEM management required 8.6 times more current to maintain state of charge than when I did things manually... That seems really big to me...
 

·
Registered
Joined
·
1,253 Posts
I'm just giving you grief though, I generally keep my battery close to full unless I really need it. Pretty rare it goes below half and then I'm trying to be easy on it to refill. I actually had Scott upgrade the fan and set it to always on to keep the Arizona heat from killing it, seems to be pretty happy most of the time. I only had one issue when it was 100 out and I started getting limited assist and regen. Turns out the intake picked up a plastic bag, removed it and within an hour of driving it was fine again.
 

·
Registered
Joined
·
7,746 Posts
I'm just giving you grief though...
Not sure what you mean...

I generally keep my battery close to full unless I really need it. Pretty rare it goes below half and then I'm trying to be easy on it to refill...
Are you going by the 'BATT' gauge to judge "full" and "half"? I don't think it's able to indicate the true charge state, or it's not designed to. There's a variety of things that make it difficult to judge true charge state, particularly with the BATT gauge, but even with an OBDIIC&C...

Probably the first or main thing is simply cell balance, or maybe tap balance: The BCM will see a cell or tap reach the 'full' threshold and call it quits, even though most of the cells aren't truly full. So full and empty depend on the balance of the cells. Maybe one cell is near full, or near empty - but the BATT gauge implies everything's full or empty.

The second thing is the non-linearity of the BATT gauge: top 3 bars are usually equal to something like 6% of total nominal capacity, or roughly 12% each of usable capacity, the lower bars are something like 2% each. So, deplete the top 3 bars and you've used 3 X 12% = 36% of usable capacity, roughly...

Third thing - rescaling: The BATT gauge has the ability to rescale, so the actual capacity of the bars isn't necessarily a fixed amp-hour amount. Not sure if they're even a fixed percentage of total estimated capacity, I don't think they are, it can be different...

Etc etc. Basically, only way to be reasonably sure you've got a 'full' pack or an 'empty' pack, or close to these, is to really stuff the pack, such as by utilizing an OBDIIC&C's SoC reset feature, or to really drain it, such as by discharging in auto-stop until neg recal. And even then, you're only reaching the limits of the highest or lowest voltage or charged cells.

As they say: BATT gauge is a finger in the wind at best. At worst it's a pernicious, cynical, evil device deployed by Honda - to manipulate dumb-*** Insight drivers.
 

·
Registered
Joined
·
1,253 Posts
Apologizing for giving you grief was in reference to the 5 year warranty and if he dies he dies meme from Rocky 3, it was a joke because you were saying for people to be nice to the batteries because you have to cover them. Mine is relatively new so hopefully it's in balance. I should have my grid charger back in a few months and I'll do the 6 month grid charge. When we do those do we only do a charge and not a discharge? I'm just going off the battery gauge, I generally try and keep it full. I'm pretty excited for Peter's C&C, it's on the way as we speak but probably be a week or two with overseas shipping and I have to solder the one board because my likely green stick pack killed it first grid charge so he couldn't prep my spare for me.

I'm generally not hard on stuff just for the sake of being hard on it, I even added an upgraded fan that's always on to save the battery from Arizona heat and it's doing quite well so far as I can tell but we haven't hit any 110 days yet. I've only ever gotten throttled when it picked up something behind the passenger seat and blocked the air intake.
 

·
Registered
Joined
·
7,746 Posts
^ I think you're mixing me up with Eli - eq1/Eli - it's happened before.
 

·
Registered
Joined
·
36 Posts
Welp im officially confused --- between you and Eli I think I might just split the difference and try to keep my pack between 1/4 and 3/4 lol

Eq1, any advice for charging individual sticks on my I max B-6 charger? I know that's a major question but just to keep it brief what range do you stay for discharge/charge volt wise and also is there any tricks for taking sticks that bleed down fast and reconditioning them so they wont?
I seem to have a pack that does ok (not great but driveable) but just sitting overnight has to go through an entirely new re-calibration process in the morning due to the storage graph reading full when it was parked and it not being full at all...
 

·
Premium Member
Joined
·
640 Posts
Welp im officially confused --- between you and Eli I think I might just split the difference and try to keep my pack between 1/4 and 3/4...
This is exactly how I feel after reading all this. Very confused. But the truth is that is exactly where my pack tends to hover, between 40 and 60%. Which is what it seems Honda intended if that is how the car is maintaining the battery without me interfering.

I want to take the best care of my battery but there seems to be contradictory info here.
 

·
Registered
Joined
·
7,746 Posts
^ Yeah, I know. It's funny how me and Eli seem to have settled into opposite camps... In general Eli should be the one with the superior knowledge and insights, as he has worked with tons of packs and does this for a living. On the other hand, sometimes I think that's part of the 'problem' - that because he's doing this for a living, the goals are different. For example, he might focus on reducing risk and maximizing longevity across a broad spectrum of users and usage scenarios, where, say, 5-6 years of seemingly error-free performance for the most packs is...acceptable, good enough, etc.

I'm not suggesting anything nefarious here, though. He shows the graphs in those earlier posts and generally supplies some rationale, I can see why he came to those conclusions and as far as I know he really believes them...

I still want to know whether he has discharge graphs of the sticks with supposedly the 'weird knee', though. To me that's the key question here - whether his sticks that have been operating at low charge state end up with genuinely stunted performance. That's where my experimentation comes in - because I see the higher charge voltage being a good thing, because voltage is also higher on discharge, i.e. more power, among other things.

I just care about maximum performance from my personal pack, doing as little as possible yet as much as it takes, even if that means thwarting OEM battery management, etc., trying to make the pack last as long as possible... On this latter aspect, actually that's secondary - I care more about being able to thrash the pack day-in and day-out. The longevity thing - who knows, it might happen, it might not - though it seems to be happening.

My pack is an original 2002 (except for 2 sticks), when I got it it was used and error-throwing, and now it's been 4 years since I started reconditioning and testing stuff. It still puts out full assist (about 90 amps at about 125V) and takes in full regen (~40-50 amps) across something like a 30% usage window (i.e. about 2000mAh of capacity at full performance level). The window thing is hard to nail, it may be larger, and it can be smaller, it depends on how the pack is being used, such as whether you charge-up day 1, let it sit, and then use the pack next day, in such a case the window can shrink... etc etc. It really is complicated stuff, just a lot of things to think about.


Eq1, any advice for charging individual sticks on my I max B-6 charger? I know that's a major question but just to keep it brief what range do you stay for discharge/charge volt wise and also is there any tricks for taking sticks that bleed down fast and reconditioning them so they wont?
You should probably copy and paste this into a new thread, I might be able to contribute there... The short answer is, really, I have no advice.
 

·
Registered
Joined
·
411 Posts
My battery seems to respond very well to grid charge, semi deep discharge done in cycles of three every six months. I emphasize, seems to. Cause I’m no expert or even amateur battery specialist. Scott Kulbeck during My last visit said to do it every 4 months. On my last 2000 mi road trip I drove extensively on mountain roads. As per eq1’s suggestion I do not baby my battery. While climbing. I used second and third gears in my 5 speed as well to maintain speed. My IMA battery gauge regularly read three bars and was still putting out assist. Later, on the down slope I would see full charge again. With my limited physical chemistry and electrical understanding and the ‘haystack’ of IC threads on this issue ringing in my ears I believe this: 40% to 60% stock Honda battery management allows chemical crud to build up at the electrode electrolyte interface. This combined with other issues over time takes out some of the cells in the pack. This is a chemical issue. Charging combined with deep discharging and vigorous use of the battery seems to reverse, break up and normalize these chemical issues to a great enough extent to significantly extend the operating life of stock IMA batteries. I’ve gone thru a learning curve on this. I’ve allowed the bulbs on my discharge set up to go dark and full pack voltage go to 0. Still, even my imperfect application of these maintenance cycles seems to be working well
 

·
Registered
Joined
·
7,746 Posts
...As per eq1’s suggestion I do not baby my battery.
hey hey, when did I suggest that?... There's a major caveat here: I think one can thrash their pack -- if it's already up to snuff, like if the cells are 'conditioned' and balanced. Otherwise, who knows, thrashing will probably just hasten what's already a pack in major decline.

Personally, I don't care what happens to my pack; I'm more interested in seeing how much I can throw at it yet still have it fully perform. It's been at least a solid couple years of 'thrashing'. I think just lately I've gotten even more 'abusive' - like I'll let it charge at super high rates when cool, say 2nd gear background charging at around 20 amps. Or, I'll do braking regen when cool, down steep hills, at whatever charge state. And, of course, I'm doing full throttle full assist on a daily basis, at as low as 5% true charge state (or thereabouts, about 5-15%).

That's what's so amazing about low charge state: I can have a nearly empty pack (absolute truly near empty), let it charge up a mere 5-10%, and immediately throw full assist at it - and the pack doesn't bat an eye. Loaded voltage stays rock-solid. Temp stays low, et al. When I cycle high, this behavior just doesn't stick, performance declines, voltages sag, the usable window shrinks...


Off the top of my head, I'd guess that:

-high charge state usage might enable a small fairly high performing usage window (10%?), mediocre to low performance across maybe a 20-40% window?, and maintain such performance for a fairly long time, 5-8 years?

-low charge state usage might allow top performance across a fairly large window - 30-40%? - high power output though possibly not total energy output, perhaps for a long time -- 10 years? 15 years?? But perhaps it's more risky - more risky because if you have unbalanced cells you're pushing the lowest cells very hard and it probably increases imbalance.

That last idea might be key to what Eli sees: it's hard to have perfect balance down low (it's actually not possible without circumventing BCM management), yet if only one or a few cells go low, near empty, they'll start to have different voltage behavior from the other cells.* The other cells will be more like cells that are used at high charge state, with depressed voltages; only the lowest will have the alternative, higher voltage behavior...

* Here's an old graph that illustrates this idea well:


The yellow curve is a discharge of a fast self discharge cell. This would be the lowest-charged cell in the pack, the first to reach empty and causing 'neg recal'. The black curve is just a normal other cell.

As a result of being discharged to empty, the 'yellow cell' has a high discharge voltage curve. The 'black cell' has a saggy, depressed voltage. Since the yellow cell reaches empty before all others, and since the BCM prevents additional discharging, the black cell (and all others) never sees empty and its voltage remains depressed. The yellow cell's voltage doesn't. The yellow cell ends up leading a totally different 'life' in the pack and the imbalance causes problems (of course, the fast self discharge is the real problem here)...

The 'low charge state' idea is to have all cells performing like the yellow one - minus the fast self discharge. The high charge state idea is basically having all cells look like the black curve, or at least that seems to be what you end up with, with high charge state usage.
 

·
Registered
Joined
·
7,746 Posts
I was trying to add this to the previous post but was having problems:

Here's a couple more graphs that can be used to further illustrate the 'low charge state' idea:

As I recall, these were discharges on cells that had been fully discharged - 'deep discharged' per whatever typical regime - then charged half way. The curves are for pairs of cells as that's how my equipment works. Look at how high the voltage curves are - they don't hit 2.4V (1.2V for a single cell) until they're nearly empty. That's high-power output capability. Compare these curves to the black one above. The black one above is below about 1.25V for nearly the whole duration of the discharge (about 30 minutes, same duration as for the graphs below)...

90911
 

·
Registered
Joined
·
411 Posts
Did I forget to say thrash? I guess what might be helpful here for me at least is a set of simple guideline rules for daily use that lead to a few more years of battery service. When I first got my G1 I noticed the instrument display was training me to drive in a way that kept the bars on the battery charge display as high as possible. Basically I drove so I used assist as little as possible. So now after reading all the threads on battery care. I think using assist promotes battery health. I don’t thrash my car. I don’t obsessively hypermile. I do however keep the fun in driving my car. So now at least I drive more by feel and sound while being a little less concerned with the gauges. So eq1 how does your understanding of all your technical work on these batteries play out where the rubber meets the road. How do you drive your car these days?
 

·
Registered
Joined
·
411 Posts
Actually eq1 as i re read you above you make it clear that you work the battery, even when it’s near empty. And further you imply that you think it puts out more power at what you call low SOC and imply that may be better for the long term health of the battery, and you Eli too feel this way. How do you drive your G1s Eli?
 

·
Registered
Joined
·
7,746 Posts
Actually eq1 as i re read you above you make it clear that you work the battery, even when it’s near empty. And further you imply that you think it puts out more power at what you call low SOC and imply that may be better for the long term health of the battery, and you Eli too feel this way. How do you drive your G1s Eli?
^ Based on what Eli's written in this thread - he does NOT feel this way. The whole thread is about how Eli "cringes" when he hears me (or whoever) talk about low charge state usage. His recommendation is the exact opposite. And, frankly, the "simple guidelines" you're looking for are to do exactly what Eli suggests - use as high a charge state as possible.

What I'm talking about would be too difficult, too much trouble, for most people to do. Most people's cells are probably too imbalanced, too crudded-up already, to get any where with the 'low charge state regime', so you first have to get the cells conditioned and balanced by external means, and then you need to thwart BCM management to keep it like that. I haven't figured out an easy way to do all this. I mean, it's relatively easy for me, but I have an OBDIIC&C, a CALPOD, I take notes and log data, and I generally know what I'm trying to achieve.

The way I see it, my 'regime' is more of an extreme enthusiast connoiseur-like one, whereas Eli's regime is kind of the best one-size fits-all among not-so-great options one...

I don't consider thrashing the pack as part of the regime. Rather, I thrash the pack cuz that's the way I like to drive. I like to use a lot of electric, I like the acceleration, I like recouping via regen, I like testing the pack, etc. But none of this has anything to do with what's good for the pack. Maybe it is good, but who knows.

The key elements of my regime are having the cells as perfectly matched as possible (i.e. conditioned the same), charge-balanced, charge-balanced at the bottom, and using the bottom. It probably also involves leaving the pack low when you park, rather than charging high and leaving it; you lose a lot of 'charge state' from sitting even just overnight, and operating voltages will sag, whether using low charge state or not, when you go to discharge the charge state range below where you left it.

For example, if you charge up 20%, park overnight, the next day you try to discharge that 20% and the voltage will be low; if you charge up 10% (so +30% total), you'll get 10% of high operating voltages, but the lower 20% will sag... Pretty sure this happens anywhere across the full charge state range...

The difference is that if that range is absolutely high, voltages will start to sag on a regular basis, whereas if you use low it's akin to continuously rejuvenating 'something' where voltages will always pop back up to the high operating range...

There's other things a low charge state regime helps, too. If you have cells that have different capacities or have different self discharge rates, then keeping the pack low can keep them better balanced - it effectively negates the impact of faster self discharge and uneven capacity. If all cells are near zero charge state there simply isn't much to self discharge - they can't become too out of balance. If you charge up to say 75%, and you have a faster self discharge cell, then you have that huge range of potential imbalance. With capacity, if say you have a cell that has 70% of the original capacity, and a cell with 90%, once the 70% cell reaches 70% it will stop charging, while the other continues to charge. You end up with a 20 point imbalance right there.

But, like I said, it's not easy to implement this program. I'm not really sure how much a half-assed version of it works, if at all. If all you got is the dash gauge and you just drive and bring the BATT gauge low? I don't know, I doubt that will help much. You'd have to start at rock bottom and purposely try to keep it low. Like, maybe a 'simple' low charge state regime would do one of the typical full pack regimes, charge and discharge, but leave it low at the end of it all. Then, when you drive, you'd have to prevent the BCM from background charging all the way up, and, you'd have to bring charge state back down to the bottom, preferably via leaving it in auto-stop and on a fairly regular basis. For most people, this stuff just isn't practical.

As far as longevity goes - who knows. My idea is that whatever regime produces the highest operating voltages, the lowest operating temps, shows the most consistency and stability, results in the highest efficiency (like Coulombic efficiency), etc. - that's the regime that will win in the end, that's the pack that will last the longest. I see all this stuff with low charge state usage vs. high, so I do think my 'low charge state usage regime' will result in a longer lasting pack.

I'm not sure how long I'm willing to carry-on with this, though, it's already been 4 years... And really, it happened by accident - I was on the verge of building an LTO pack at the time, using my NiMH pack in different ways to understand how the prospective LTO would be treated by the BCM - I was giving up on NiMH. But, that's when I started seeing different, promising things, and now I'm stuck with it. I almost wish my stupid pack would die so I can join in all the lithium fun, it's just that, at this point I don't think there's anything a lithium pack can do that my old NiMH can't do already, and I have a hard time throwing away perfectly useful things.
 

·
Registered
Joined
·
411 Posts
Thanks eq1. You have taught me a lot. While I have not and will not go as deeply as you into this. I really appreciate that you have. While as I have said I lack the electrochemical understanding you and others on the forum have I do have just enough to, as they say, get me in trouble. I might guess, and be wrong, that Eli’s thoughts might relate to the aftermarket packs he services and sells more than to original equipment. But you have done your homework and I on some basic level understand and agree with your conclusions. I now drive my Insight the way you do. On my recent road trip, thinking of your ideas, I tried something.
Climbing aggressively on quite a few mountain grades, often in third gear trying to maintain 60 to 70 mph, I called on a lot of assist. The battery charge gauge would fall no further than three bars from the bottom. And yet time ant time again I could get assist that never stopped coming at this apparent low state of charge. I would stop paying attention on the downslope only to glance down ten minutes later and see the pack charged up full. As you have said this may only be because I luckily got a car with a battery that is not too crudded up. And we will see what the future brings. Again thanks.
 

·
Registered
Joined
·
7,746 Posts
^ Just thought of something. One good thing about 'thrashing' the pack, using lots of assist and regen, etc: Then you actually know whether your pack is in decent shape or not. Granted, the BCM/MCM throttles assist and regen, so you can't be sure you're actually getting full assist or not. But, you'll know when you see an error or whatever when it's in a seriously sorry state.

It's ridiculous. The IMA doesn't do a whole lot anyway, so if one is babying the pack or whatever, it's almost pointless to even use it/keep it. You have to max-out assist and regen to make the IMA worthwhile, otherwise you'd probably be better off just ditching the system, saving the weight, gaining some space. Given what I've seen with my pack it doesn't take a whole lot of pack to max-out the system, like, it barely takes anything.

It's almost shameful what Honda did with this system - like the main thing they were worried about was failure, so they hedged their bets at every turn, demanding very little of the pack, futzing with the user-feedback (BATT gauge, etc.) to make it seem like everything's OK -- until things are really really bad...

Packs should be able to handle full throttle loads and high braking regen without trouble, definitely without an IMA light. If you can't do that then you've got a messed up pack. It's as simple as that.

Given what I've seen with my 19 year-old pack, age and usage does not automatically mean degradation in performance. It's totally...non-sensical! That's what's so amazing. When was the last time you heard of batteries lasting 19 years, and not only that, putting out as much as these cells do/can? (mainly thinking power output and input here). Unfortunately that mental baggage makes expectations low, plus, people think the Insight NiMH cells are crap. That's nonsense. Any cell that can do what mine do is phenomenal - last 20 years, charge at about...6C, discharge at ~16C, etc etc.

The cells are fine, it's the battery management that sucks.
 
21 - 38 of 38 Posts
Top