Honda Insight Forum banner

61 - 80 of 106 Posts

Registered
Joined
164 Posts
I deep discharged yesterday down to 96v after the first 21 hour grid charge that topped out at 167v.

I plugged it back in to grid charge last night around 7:30-8pm, and this morning currently it鈥檚 reading 163v steady. 馃槙 I wonder if my battery is getting hot? It鈥檚 been in the low 70鈥檚 here, no sun, very cloudy and I have the carpet cover and cargo box removed for more ventilation, along with making sure the battery vent is sucking in air for cooling.

Update: Well, it鈥檚 almost 1:30 and other than me turning off the grid charger for about 25 minutes earlier, it鈥檚 been charging from last nights 96v discharge for 17 1/2 hours.

Currently the volt meter on the charger is now a steady 163v again instead of bouncing from 162-163 back and fourth like that every 3-4 seconds when I checked at 8:30 this morning.

I鈥檓 wondering if this is some sort of a voltage depression, or if a cell is overheating? Let it keep going?

P.s. I apologize for posting this here and tried to delete it. I see this thread is more for data. My apologies.
 

Registered
Joined
6,326 Posts
Discussion Starter #62
If you trust your grid charger and it puts out 300-350mA, letting it go longer should be fine, even called for. Staying put at 163V seems odd...

Let's assume you got most of the cells completely discharged. It'd take at minimum about 7300mAh to get the pack fully charged again: 7300mAh/350mA=21 hours... Some people have said it can take as much as 10,000mAh or so to charge a completely, deeply discharged pack - and so that's 10,000/350mA=28 hours...

On the other hand, it is possible your cells were badly unbalanced and the deep discharge (what rate did you do?) pushed some cells over the edge - they got badly, deeply reversed and are failing to recover...
 

Registered
Joined
148 Posts
Will grid charge discharge help

So, I grid charged and discharged (3 times... 96v, 6 and 12). I drove it around and it was great. I had never seen how nice it starts up before, or how the auto off worked when I stop! I have been driving it with the BCM disconnected. But it only lasted for 4 or 5 miles, then I got the P1449 with a 78.
So I checked my tap voltages.
Here is the list:

15.87
13.14 :mad:
15.85
15.44
15.68
15.73
15.61
15.94
15.87
15.41

The previous owner had the battery changed 5 years and 19,086 miles ago.
Should I try to do some more grid charging discharging? Is it possible that it will level those numbers out? Or should I not even bother? :dunno:
Thank you,
John
 

Premium Member
Joined
3,415 Posts
You have 2 dead cells on the second tap.

15.87, .85, .94 and .87 are likely good. All others likely have issues with at least one stick on the tap.

Generally, after the first discharge to 96V, the pack has recovered the bulk of the capacity you will gain from the process and the 2nd and 3rd discharges recover far less. Personally, I NEVER recommend a discharge to 12V.

Nothing you can do to this battery with pack level charging/discharging can fix it. It's done.

20K miles in 5 years isn't enough. IMHO, you NEVER want a low mileage hybrid unless you're totally cool with immediately replacing the battery.
 

Registered
Joined
81 Posts
Same with an all-battery vehicle. The one nice thing about an ICE. You can leave it in a barn for 20 years and then bring a new 12v battery and start it right up with a can of starter fluid.

Perhaps now that Hybrids and Electric cars are starting to get old a market will open for solar interconnects that you use to keep the battery from losing charge while it sits.

-Randy
 

Registered
Joined
6,326 Posts
Discussion Starter #66
If you're going to leave the battery sitting for extended periods, you should drain it by at least driving and using assist as much as possible, preferably leave it in auto stop and drain it. This will minimize the impact of uneven self discharge... In general, on extended pack sitting, it's not my understanding that packs degrade a lot just from sitting, i.e. that the cell chemistry gets out of whack. You can have uneven self discharge and that screws things up, but I don't think sitting in itself harms the cells much if at all...

....The previous owner had the battery changed 5 years and 19,086 miles ago. Should I try to do some more grid charging discharging? Is it possible that it will level those numbers out? Or should I not even bother?
[edit: One thing I forgot to ask, though: Is your pack OEM or aftermarket? If OEM then yeah, short the taps, if aftermarket, don't bother.]

You should try shorting your taps, five at a time, doing an ultra-deep discharge. The PTC resistors in the voltage tap lines act as a very gentle load. Your situation isn't ideal for this, but you don't appear to have anything to lose... It'd take quite a while since it looks like you've got cells that are totally discharged (or seemingly dysfunctional) while others are probably close to full. But, whatever.

Short the taps, let it go for a week or so for each set of 5, then grid charge as you normally would for a completely discharged pack (i.e. it will take longer than 'normal', but it's a normal, long grid charge)... You can check tap voltages by pulling the shorting pin - you want them to be close to zero, certainly below 1V, before you quit. You can also check pack voltage at your grid charge harness, make sure it's about half total pack voltage before you quit (i.e. if it starts at 144V, take it down to 72V for the first set of 5 taps, and then probably about 54V for the second set of 5)...

Here's a graphic and a couple picts that depict the process:


DIY shorting pins:

First set of 5 taps, every other tap

Next set of five taps, every other tap
 

Premium Member
Joined
3,415 Posts
Why on earth would you recommend that? He's reported a tap at 2.X less than the others AFTER 3X grid charge/discharge cycles. You can't possibly expect something meaningful to occur, can you?
 

Registered
Joined
6,326 Posts
Discussion Starter #68
Here's a bit of voltage musing I just posted in another thread, thought it might be good here, too: https://www.insightcentral.net/forums/problems-troubleshooting/123347-increasing-spread-subpacks-after-storage.html#post1425691

"^ I've never really figured out what different peak/charge voltages mean. The old, ubiquitous IC wisdom was that higher voltages were bad, that they meant 'high IR', for instance. But there's more to it than that, for sure... If anything, there's a 'good range' (after adjusting for differences caused by temperature), and anything below and anything above suggests some funk, undercharging or overcharging, etc... 177V seems about right.

My own hunch, kind of based on what I've seen off and on, is that packs that have been used just as the OEM BCM programming would have you use it, will tend to have lower peak voltages when grid charged - say around 172V. I kind of think that 'crud' develops and it tends to pull the peak voltage down, a bit counterintuitive if thinking along the lines of that old IC wisdom...

In my head, it's like there's portions of the cells that aren't contributing to the normal reactions any longer, or rather, there's a different chemistry that's now pulling the voltage downward... I guess you can say it's the same thing or similar to 'voltage depression'... Picture each cell as having 10 compartments, 9 compartments with a mix of two chemicals that produce a relative voltage of say 1.48V when being charged, the normal process, and 1 compartment with a slightly different mix that only produces 1.43V when being charged. The sum of the voltages of the 10 compartments ends up lower...

Having your pack sit for a year - I imagine slow, gradual self discharge could have burned through some of the larger 'crystals' and then, upon charging, you end up with a higher peak voltage... The resistance is lower - which you'd think might produce a lower peak voltage - but I don't think it works quite like that, not so cut and dried. Rather, the lower resistance makes it easier for more active material to get charged, and then you end up with a higher peak voltage... The strange thing is, though, it seems like voltage for 'good' cells tends to drop faster and farther than voltage for cells that are in some way funky. Voltage for my 'bad' cells tends to stay higher, longer. Voltage for my 'good' cells tends to drop to just below about 1.40V fairly quickly, like within half a day or so... Not totally positive on this, but something like it seems the case... "
 

Registered
Joined
6,326 Posts
Discussion Starter #69
Here's another snip from a voltage-related post I did in that same thread: https://www.insightcentral.net/forums/problems-troubleshooting/123347-increasing-spread-subpacks-after-storage.html#post1425819

After another 24h after the 2 gridcharge the battery settled at 164.9 V.
Min to Max subpack reduced to 0.04 V (16.46 - 16.50 V) but this time another subpack is the lowest. It's only at the extreme upper end where they separate to a 0.20 V difference. I accept that for a 9 years old battery pack as very good.
Sounds good to me. I see the same kind of spread difference between resting and peak charging. Makes sense, typical (though most used packs are probably way worse)... I think 1.37V per cell resting, after a full charge, is probably what it should be. Above I mentioned that my 'good' cells tend to fall just below about 1.40V, while the 'bad' ones tend to linger a lot longer above 1.40V. I think that value below 1.40V is about 1.37V... I think it's possible, probable, that some kind of 'crud' or whatever can end up blocking what's likely to be normal self discharge 'off-the-top' of fully charged cells.

Here's a little table of my tap voltages last time I did a grid charge. It starts after the rebound of a tap-level ultra-deep discharge, moves to grid charging - I recorded intitial/early GC voltages and then about every 10 minutes toward the end, then after about an hour of rest. Unfortunately I didn't record the resting voltage after a day of rest. The bottom rows are a few measurements after some driving, such at about 50% SOC and after a pos recal. Values omit the decimal point - so "1657" means 16.57V (UDD voltages are just mV, so "803" means 0.803V)...

I know that stick 54 in the 54,59 stick pair has at least one funky cell - always has voltages about 0.10V above the others... My 'spreads' are even bigger than TRI's, yet this pack functions great - and I really mean great, it can do full assist at 40% state of charge, for instance, performs almost as well when the pack is cool, only 45 degrees F, etc, etc...

 

Registered
Joined
6,326 Posts
Discussion Starter #70 (Edited)
Kind of expanding on what I wrote in this post above, https://www.insightcentral.net/forums/modifications-technical-issues/89298-quintessential-insight-nimh-voltage-thread-2.html#post1389627, thought I'd add some interesting tap voltage values I'm looking at...

I've noticed a bit of slippage in my pack recently and decided to take a look at the tap voltages. I'm mildly surprised and I guess glad to see the 'slippage' reflected in the values: most are about 14.95V while one is 14.87V. Most likely one cell in this 14.87V tap is essentially empty while others are not so empty, i.e. I've got some imbalance, which is what I was thinking it'd be based on the behavior I was observing while driving...

What's interesting - and rather fortuitous - is that I'm actually able to diagnose/confirm by reading the voltage taps after a neg recal, at rest, without any other methods, like subjecting the pack to a load, yet the spread is only about -0.08V.

I was last using the pack at a very low charge state - though not as low as I had been able to go before - and I was seeing super aggressive assist throttling. It was actually rather odd. I was at about 40% nominal, a level at which at other times I've been able to exercise the pack freely. But this time I was seeing the super aggressive throttling - I'd get a second or two of high assist, but then it'd be throttled to almost nothing. I was doing this between about 35% and 40% charge state multiple times, charging a bit, using the pack, charging again, etc...

Typically I'd get some throttling around 36% nominal, but it's more even-handed, like an across the board, one-and-done setting, after which you just can't get high assist levels. This time as I say I was getting high assist then immediate throttling, high assist then immediate throttling, repeatedly. Seems odd that I didn't get a neg recal during any of this - which seems to suggest that you can have a near empty cell/tap yet NOT get a neg recal, only the aggressive throttling.

When I parked the car I left it in autostop, at 35.5% nominal and about a 1.2 amp discharge rate. It only took -80mAh for the pack to neg recal. About an hour later I took the tap voltages. Here's all the voltage readings, divide by 100:

1496,1493,1495,1495,1495,1491,1493,1493,1493,1487

I've gone a long time without much if any noticeable 'slippage'. I took a trip a few weeks ago during which my pack saw heavy use; in a single day I used 10.6 pack's-worth of charge, for instance (69Ah throughput). I'm pretty sure this usage alone, something about it, is what has thrown my pack off. A few days before the trip I was able to drag charge state down to 28% nominal with no troubles, for example. So, I don't know, I'm thinking the multiple cycles up and down, especially the high charge state charging, maybe more heat than usual, maybe starting charges and discharges in 'transient' states a lot (like doing a charge immediately after a discharge, and vice versa), was enough to drag one of the cells in that last tap out of line... Normally, perhaps, there's the same weakness in the one cell, but the usage is lighter and there's more opportunities for it to 'catch up' with the others. On this trip, the usage was like 'bam bam bam', constantly hammering on the cell - seems like that might exacerbate any weakness, make it worse...

Alternatively, maybe all that usage in a tight window leaves some 'unreacted crud' laying around and it causes faster self discharge?? Who knows, just some random idea...

It's also possible that there was some slippage before the trip - that after about a half year of usage since my last tap ultra-deep discharge, some voltage depression was setting in: I had grid charged the pack only a little more than a month ago because I was seeing some slippage then - like a neg recal at 33% instead of being able to take it down to a 28% hang. Only the grid charge didn't last. I'm kind of leaning toward this. I don't know if it happens to all packs or what, but my pack seems to get depressed like in the 1/2 year to year time frame, even if I use the pack at low charge states. Using the pack at low charge states props up the performance and slows the onset of depression, a lot. But it doesn't appear to fix everything always...

* * *

Just thought of something else I should say: It's important to keep in mind that the quote "slippage" I'm talking about is way way less than your typical IC accounts of 'IMA problems'. At most I have about a 10% imbalance in my pack - I'll get a neg recal at say 35% instead of 25%. Most people would never notice this level of slippage/degradation. If I let the car do its thing, it'd charge the pack and I'd have at bare minimum 30% of usable capacity (instead of like 50-60%). 30% is plenty of charge to get your IMA on.

But, the point is, once a pack degrades or slips as much as it takes for most people to notice, such as getting recurring premature neg recal or IMA lights, it's probably long past time that something should have been done. Imbalance means a cell is out of line, and though I don't know for sure, I presume that a cell at the margins being pushed as hard as others will likely get pushed farther out of whack. Not to mention performance reductions, power reduction, due to the BCM and MCM's ways of handling 'slippage'... So, it's probably pretty important to nip problems in the bud when it comes to ensuring the longevity and performance of our packs...
 

Registered
Joined
6,326 Posts
Discussion Starter #71
Over quite some time using my pack/s at very low charge states, I've honed-in on what appears to be rather cut and dried Insight NiMH voltage behavior, as well as the car's response to it. If one were to watch for this behavior it could reveal quite a bit about the condition/working state of the pack. All you have to do is take the pack down low and watch pack voltage, such as with a voltmeter or OBDIIC&C. It's easier and more revealing with an OBDIIC&C - because that allows you to reset state of charge artificially high, which allows you to more easily take the pack low. But even without one, you should be able to glean some useful insights.

Simply put, the pack is empty or very near empty if after an assist event pack voltage doesn't rebound above 144V. I've consistently seen neg recals or the equivalent forced-charging behavior, bar gauge drop, and nominal state of charge drop that happens when the pack is empty - when pack voltage doesn't rebound above 144V.

But the actual cutoff happens at the tap-level: if a tap's voltage doesn't rebound above 14.4V after an assist event, you get the neg recal/forced-charge behavior. So it follows that, if you can see a rebound voltage of around 144V after an assist event - without getting a neg recal/forced-charge - then your pack is balanced, i.e. at or near empty all taps are at 14.4V (14.4V x 10 taps = 144V).

On the flip side, if you can't, it means your pack is imbalanced - because at least one tap is lower than the others, triggering the 'pack empty' behavior... For example, this tap spread would cause neg recal/forced-charging/pack empty behavior:
14.9V,14.9,14.9,14.9,14.9,14.3,14.9,14.9,14.9,14.8
One tap doesn't rebound above 14.4V, triggering the forced-charge/bar gauge to plummet, yet total pack voltage is 148.3V.

The higher the total pack voltage, the greater the imbalance. My guesstimate is that there's many packs out there that would be around 150V+ when neg recal/forced-charging happens, most taps at around 15.1V or more, and one at 14.3V (again, after an assist event).
 

Administrator
Joined
10,960 Posts

Registered
Joined
6,326 Posts
Discussion Starter #73
One other thing I should probably add here. One of the long-lived factoids at IC says the BCM won't let you discharge the pack all the way. The common thing to say is that 'the BCM allows pack usage between about 20% and 80%.' I won't dig into the weeds with this (I could), but rather, let me merely point out that, with an OBDIIC&C, using the SoC reset function, you can drain the pack to pretty much truly empty. In all these years I never quite realized that was possible, until now. The '144V rebound thing' seems to be a tell-tale sign.

Even without the OBDIIC&C/resetting SoC, you can drain the pack completely, just that it takes a lot more work because you have to fight the nominal-SoC-triggered throttling behavior and either use the Calpod IMA disable switch repeatedly, to disable regen and background charge, or do some fancy shifting into neutral. Also, I should point out that it appears some BCMs do allow full discharge as a matter of course, such as the A03 I might've mentioned some posts up.
 

Super Moderator
Joined
7,516 Posts
I hope my post won't upset the larger scheme of things in the thread, but I have a battery phenom which is causing me to scratch my head a bit. I am cycling a friend's car with the Genesis suite.

Before I started, just for ref I measured the tap voltages, at:
15.32
15.27
15.31
15.27
15.29
15.16
15.29
15.28
15.19
15.28
15.18

BUT, we were undertaking this because the IMA warning light was lit.

So, I'm a bit puzzled. The taps are really close, yet the battery threw codes. I'm thinking that there must be a cell somewhere in the 120 cells which has high IR causing the taps to go wonky. Does that make any sense???

Unfortunately, I don't have the codes :(
 

Registered
Joined
6,326 Posts
Discussion Starter #75 (Edited)
I hope my post won't upset the larger scheme of things in the thread....
Not a problem at all. Happy someone's actually visiting it...

You should take a close look at this earlier post in this thread, and maybe the one above it, too: https://www.insightcentral.net/threads/the-quintessential-insight-nimh-voltage-thread.89298/page-3#post-1389627

It details just how difficult it can be to get a good read on tap voltages, and how close they really should be. 15.16V to 15.32V, or a spread of 0.16V, is actually kind of large. From what I can tell, anything more than plus or minus about 0.1V suggests something's off. I mean, it's amazing just how close the voltages of these cells track when they're operating as they should.

Anyway, variation in the voltages you've presented could be due to a number of things, such as longer term imbalance, where some cells are more charged than others, self-discharge, 'high IR', voltage depression/memory effect, 'crud', etc. These are imbalanced taps... My guess would be that the lowest voltage tap caused premature neg recal, that it has one or more lesser-charged cells...

fyi, for comparison, here are my tap voltages. I've been taking measurements every day for the past few weeks, on a new rebuild of an old failed rebuild:
16.04,04,04,07,04,03,04,04, 15.96, 00
Here's some random voltages from my last pack, shortly before I did the swap:
16.38,38,38,38,38,38,38,38,37,38
Here's some older, lower charge state (~actual 20%) tap voltages from my previous pack:
15.16,16,16,16,16,16,17,18,16,16
 

Super Moderator
Joined
7,516 Posts
That ref post was interesting. The technique of measuring the taps under even modest load looks to be a very promising diagnostic tool. Unfortunately, I'm about 2/3 way through my current 3 cycle routine, so probably too late on this cycle effort. It is on the second discharge right now, so I'm at least gonna check the currrent voltages. I can see though that this is a very promising/proven technique. Thanks.

LATER: The discharge is about 2/3 through. I measured the taps under the light load and they are separated by .08V still. I got my fingers crossed that this is going to do the pack some good. Still one more full cycle to go.
 

Administrator
Joined
10,960 Posts
As stated tap variation of greater than 100mv (0.1V) usually indicates trouble brewing..

Personally I start sweating when they deviate by > 50mv

New sticks generally vary by only +/- 10mv..
 

Registered
Joined
6,326 Posts
Discussion Starter #80
As stated tap variation of greater than 100mv (0.1V) usually indicates trouble brewing. Personally I start sweating when they deviate by > 50mv....
I think I lean toward the 50mV value as well. I wouldn't say I'm "sweating," but it suggests a real difference (opposed to just measurement error, for instance). Anything in the 50mV to 90mV range and my brows are raised. At 100mV I know there's either a bonafide electro-chemical difference or a gross charge imbalance -- typically the former, which gives way to the latter, eventually...
 
61 - 80 of 106 Posts
Top