Honda Insight Forum banner

21 - 40 of 49 Posts

·
Administrator
Joined
·
12,434 Posts
This is an interesting discussion and I read most stuff on the forum.
EQ1 has a great handle on a lot of things and stick/tap level stuff/chemistry etc

I just can't be bothered with much of that anymore. I'm stupidly busy working on gadgets/projects etc etc
I did stick stuff ten years ago with hobby chargers like the rest before we knew better and realised it's limited value and exorbitant consumption of time.

I built and still use a tap voltage monitor based on a hacked BCM, and I also have an earlier one made out of ten Chinese voltmeters. They are interesting to use, mainly to demonstrate to a customer that their pack is garbage.

I personally do and recommend pack level stuff as the first port of call, it's relatively easy, simple and quick.
Plug in battery cycler for a week let it do it's thing, measure taps, let it rest for two weeks, measure taps again.

Or simply cycle pack, is car/pack any better Y/N? If Y great.. If N then time for a new one, bypass or a conversion.
Almost all packs will respond positively in some degree to cycling, even if only for a relatively short period.

I always charge for 24hrs before discharging any packs.

I still have in my mind an automated bench stick tester that you load like a magazine or machine gun belt and it works its way through a batch of sticks, evaluating and categorizing them in an autonomous robotic fashion.

At the end of a week it's done 20/50/200 sticks and you get a simple report saying bin or keep for SD test.
I have hundreds of sticks but time is the enemy so it may never get done.
 

·
Registered
Joined
·
7,165 Posts
Sounds like a good idea. Take some unloaded and loaded tap voltages and report back. Hopefully we'll be able to spot something...
Was re-reading some of this thread and realized that, at this point, taking tap voltages almost certainly won't add any knowledge to inform our decisions. You already took unloaded and loaded stick voltages - and the loaded current was even around 7 amps - so there's nothing tap voltages will add to this.

Don't know why that didn't click with me earlier, I had even re-posted that chart. I think I was thinking there had been steps, car usage, etc. in between those measurements and now. Re-reading, I don't see that that's the case - so your sticks should be basically the same now as they were in that chart - unless you have some super-duper fast self-discharge cells...

So... Looking at the chart I made of your unloaded and loaded stick measurements, sticks 4,2,3 are lowest, 14,15,19,20 highest, the rest pretty even.



Here's your stick order from that other thread you posted:
My numbering is below, I have no idea why I went left to right...

“BLANK”654321
13121110987
20191817161514
Based on the ordering of your sticks in the table above, the Taps as depicted in the diagrams I've posted, are comprised of these sticks:

Tap 1= sticks 14 and 15
Tap 2= 16,17... 3=18,19, 4=20,13, 5=12,11, 6=10,9, 7=8,7, 8=1,2, 9=3,4, and 10=5,6
(in the future it'd probably be best to renumber your sticks so Tap 1 has sticks 1 and 2, Tap 2 sticks 3 and 4, etc.)

hmm... Though it's hard/nearly impossible to say, I think we have to assume that stick 4 is the most likely to have a near empty cell. And since stick 3 is also among the lowest, as well as stick 4's pair, that's a tap we need to exclude in the tap discharge procedure... Sticks 3 and 4 belong to Tap 9 - so we exclude Tap 9 from the discharge.

Stick 2 could have a near empty cell too. I think we need to exclude the tap that contains stick 2, even though it's pair (stick 1) isn't low. Stick 2 belongs to Tap 8 - exclude Tap 8 from the discharge.

The second step is to consider how much we should discharge, at this point...

If we assume Tap 9 is the least charged, we also have to assume it's no more than about 10% (I've seen 20%, but 10% is the P1449-78 'spec'). And the highest tap theoretically should be around 75%, though I wouldn't put too much stock in that value.

IF we knew exactly what charge state all your sticks/taps were at, it'd be pretty straight forward. For example, if you had one low tap at 10% and all the rest were at 75%, you'd simply discharge all but the low tap for the duration that would discharge 75% minus 10% = 65%. Since the PTC load discharges at about 14% per 24 hour period, we'd have to short the taps for 65%/14%=4.6 days. And that's for each set of 5 taps...

So what do we do not knowing exactly the charge state of your taps? We guess, go half way, or simply do what we can in the time we have... That's about how we get to my earlier suggestion of 30%. Make it 28%, for 2 days, 48 hours, each set.

* Discharge Taps 1,3,5,7 for 48 hours if you can fit that in, then 2,4,6,10 for 48 hours if you can fit that in. However much you end up doing with one set, do the same time on the other set.

After that - not sure how we'll have to deal with the P1449-78 code pending. Seems like that doesn't always erase if you pull the 12V neg or #18 fuse. But try those, you should then get some charging from the ICE... I think I'd do all this before your grid charge, when it arrives, because then you'd be able to get a read on how imbalanced your pack remains, i.e. if you can actually charge it in-car and use it for a good stretch of assist without getting a neg recal, you get a sense of how much usable capacity there is...

* There's other ways to do this, too. You might rather try discharging just the taps with the highest voltage sticks in the chart, maybe for a day, just to see if it helps. So, taps with sticks 14,15,19,20. That'd be taps 1,3, and 4 - but remember, you can't discharge taps that are next to each other at the same time, so no 3 and 4 at the same time, it'd have to be 1 and 3 for one session, 4 in another session. Etc etc...
 

·
Registered
Joined
·
90 Posts
Discussion Starter #23 (Edited)
OK, time for some tap discharging while I wait on my charger to be returned, or the rest of my DIY parts.

I really appreciate that you've distilled the theory down to actionable tasks for me. That's great. I'm looking for other documentation on this because I'm trying not to ask dumb questions but I'm not quite understanding it. The image you've posted in other threads that reference the BCM connector where the taps are, you've got them labeled with letters and polarities; but here you're referring to tap 'numbers,' unless I'm misinterpreting something.

Is it like this?
(A+, B-) = Tap 1
(C+, D-)= Tap 2
(E+, F-)= Tap 3
etc.
This seems to make sense since you also mention taps can be called "AB", "CD" etc. But I'm not able to correlate that with your discharge instructions.

Can you clarify?
 

·
Registered
Joined
·
7,165 Posts
^ That's exactly what it is. The 'AB' 'CD' nomenclature helps to capture the fact that each 'tap' is made up of 2 sticks, where letters correspond to spaces in the case, not the actual sticks. That way you can also, say, get replacement sticks and number them sequentially, put them anywhere in the pack - yet retain the 'AB' 'CD' etc.-to-tap relationship... For example, I currently have sticks 64 and 70 in spaces E and F; sticks 64 and 70 make up Tap 3...

But that's all too confusing for lay folk, and there really isn't anyone messing with sticks and taps etc. anymore, so it kind of outlived its usefulness (in reality, I think it was only ever useful to me).

My brain has a hard time converting between letters and numbers in a speedy fashion though. Yet, when you're working with taps you need to keep the numbers straight since you can't work with taps that are next to each other, i.e. you can do even, or odd, or some combination as long as they're not next to each other. Easier to keep track of that when the taps are numbered.

One thing I want to mention now, that I'm not sure I mentioned in the other threads/locations: You should turn the pack switch OFF when you're working with the taps. It's not necessary for the process to work, but it should make it safer... I'm not really sure where the danger lies, but Peter P. commented to someone who was measuring tap voltages the other day, with wires hanging out of the taps, that it was dangerous.

If the pack switch weren't OFF, the maximum DC potential between the end terminals (Tap 1 and Tap 10) would be whatever voltage the pack were at, so maybe around 160V. I guess, if one were to have a wire dangling out of each of the end terminals, and grabbed hold of each wire, there's a risk of shock? I don't quite understand that, because there would also be a ~380 ohm resistance in that circuit, so the maximum current that could flow would be about 160V/380ohms= 421mA... I think I read one time that it's voltage that matters, though, not current - that you can have a serious shock even if the current is very small...

IF the pack switch is OFF, the maximum voltage at the terminals is about 100V, and you'd have to have a wire dangling out of I think Tap 5 and Tap 10, and grab hold of those...

The risk seems pretty limited, in either case, to me... The shorting pins you make are small and can't bridge between Tap 1 and 10 or Tap 5 and 10. And if you have the pack switch OFF the max potential is, I think, too low to be seriously dangerous... I've been poking around my taps a lot over the years and haven't had a problem. I've even accidentally shorted the wrong terminals (a couple times), so putting an approximately 28V potential through the ~380 ohm resistance - which bumps the discharge current up to 78mA, not to mention discharging the wrong sticks... But, no lasting problem, no failures...

I think the main risk would be if a PTC failed, then there would be a short with high current potential - like taking your 12V battery cables and touching them together. The 2 sticks would discharge as fast as they could through the small shorting pin and small ~18 AWG wire, probably get really hot, melt stuff, etc... I think that's a risk we just have to accept if we're doing this sort of work... Wait, actually, there's two PTCs per tap, so both PTCs would have to fail to get this kind of 'uncontrolled' discharge. If one failed there'd still be a ~190 ohm resistance in place, so the discharge current would just double. Both failing at the same time seems like a very very small probability.

* * *

When you make your shorting pins, make sure you use paper clips that fit nicely inside the BCM metal terminals - they should slide easily yet snuggly into the terminal proper, not just jammed into the connector space that holds the terminal. The pins should be completely surrounded by the metal terminal, not held in place between the plastic connector and the outside of the metal terminal. Typical medium-sized paper clips work.

Have good light, like a head lamp, when you're inserting the shorts. Double-triple check, work methodically.

* * *

Here's an updated tap-shorting diagram, as well as the images of tap shorts in place from those other threads/posts:

88282


DIY shorting pins made from paper clips:


First set of 5 taps, AB, EF, etc. or 1,3,5,7,9


Second set of 5 taps, CD, GH, etc. or 2,4,6,8,10:
 

·
Registered
Joined
·
90 Posts
Discussion Starter #25 (Edited)
This is great, maybe I'm not as dumb as I look. :) Great discussion on the safety factor, I'll definitely turn the pack switch off. Maybe Peter will come back and chime in on that. Thanks for the updated images, this is perfect...much more clear.

I actually planned to do this yesterday but I wanted to take the tap voltages before I did (even though you said it probably doesn't matter), partially because I think the downhill trend continues... Yesterday afternoon on my commute, I made it about 18 miles (commute is maybe a little over 20) and the "battery" light on the dash lit up. Meaning, the DC-DC converter disengaged? The pack is still on and BCM connected. When I pulled into my driveway, my cigarette lighter-voltmeter read 11.5v at this time. I turned the car off, let it sit just a moment and tried to start it again just to see what it would do. The battery light didn't come back on, the 12v starter fired off and the engine started and my voltmeter read the usual, around 14.1v or so. This morning, no problem starting on the IMA motor, and the commute was fine again up until the final stretch. Battery light came on again, maybe the last mile. Hopefully it bounces back enough to get me home today, at least it's Friday.

All that is partially to say, when I got home and wanted to take the tap voltages...I realized I had left my DVM at my office. And I got distracted, so I did neither.

Disconnecting the BCM wouldn't prevent the DC-DC shutting down, right? It's ultimately all because of the capacity limitation of the pack due to those couple cells, right? I imagine not being able to put a grid charge on it is a big factor as well.
 

·
Registered
Joined
·
7,165 Posts
I actually planned to do this yesterday but I wanted to take the tap voltages before I did (even though you said it probably doesn't matter), partially because I think the downhill trend continues... Yesterday afternoon on my commute, I made it about 18 miles (commute is maybe a little over 20) and the "battery" light on the dash lit up. Meaning, the DC-DC converter disengaged? The pack is still on and BCM connected...
OK, I'm getting bogged-down in details, timeline. I was under the impression your IMA is disabled and that you have a pending P1449-78 preventing IMA usage. Doesn't sound like that's the case...

Has your IMA been operating since you last measured stick voltages, in that chart I posted?

Yes, the 12V battery warning light means the 12V battery isn't getting charged, DCDC is disabled whenever that light is ON... I think the typical cause of that is pack voltage falling too low under load (assist); second common cause is probably a single tap falling below 14V unloaded for some duration... The former you'd get without necessarily getting an IMA light or a 'neg recal', the latter you'd almost always get at least a neg recal (i.e. empty pack), and if the capacity imbalance between taps were too great, you'd get the P1449-78, after the car tries to charge the pack...


When I pulled into my driveway, my cigarette lighter-voltmeter read 11.5v. I turned the car off, let it sit just a moment and tried to start it again just to see what it would do. The battery light didn't come back on, the 12v starter fired off and the engine started and my voltmeter read the usual, around 14.1v or so.
Is your IMA light ON at this point??

If you don't have an IMA light, then it sounds like your pack voltage must be really low - and it has to be really low to get 12V auxiliary starter. When you tried to re-start the car in the driveway, pack voltage was so low it switched to 12V starter, and...

This morning, no problem starting on the IMA motor, and the commute was fine again up until the final stretch. Battery light came on again, maybe the last mile. Hopefully it bounces back enough to get me home today, at least it's Friday.
...the next day, voltage probably rebounded enough to avoid triggering the 12V start... I don't think you necessarily get an IMA light from this behavior, rather, you'd get one if, after the car goes to charge the pack it can't charge very much without a tap hitting the upper voltage limit...

On your commute, what's the BAT gauge doing?? Was it empty and then re-populated, then jump to the top, or what?? Were you seeing 4 green CHRG bars?

All that is partially to say, when I got home and wanted to take the tap voltages...I realized I had left my DVM at my office. And I got distracted, so I did neither.
Well, given that it seems like you've been driving and using the IMA, it seems like we need to reassess 'the plan' - based on what you're actually seeing, now, like during your commute. Basically, if your pack is 'going empty', but you're able to get enough charge in it to avoid the P1449-78, the question is how much of a charge are you getting and how long does it last?

I guess I'm trying to figure out if you're more likely to have a generally putzed-up pack, with a combination of 'stuff', or at least one fast self discharge cell that's throwing everything off. I'm leaning toward the former. I think earlier I was leaning toward the latter. On the other hand, I'm not sure it matters too much in terms of 'prescriptions'... If it were the latter I'd be more worried about serious reversal during whatever kind of discharge - because typically you'd have a cell at rock-bottom charge state. If it were the former, there's less worry. The problem is, I don't really know how to distinguish between the two types of problems, just by looking at a few measurements of tap voltages, for instance...

Disconnecting the BCM wouldn't prevent the DC-DC shutting down, right? It's ultimately all because of the capacity limitation of the pack due to those couple cells, right? I imagine not being able to put a grid charge on it is a big factor as well.
Turning the pack switch OFF and disconnecting the BCM would prevent the DCDC from disabling.
We don't know for a fact what it is - could be a couple cells, might not be. It's probably what I described some paragraphs above...
Grid charging should help, but its application is limited (more limited than most people around IC seem to believe, for instance). One of the basic forms of degradation leads to the ineffectiveness of charging - the cells just don't charge much. You have to discharge to 'free-up active material' that then can be charged properly. One of the common misconceptions among people who grid charge is that, when you first charge the pack all the cells reach 100% full, and then you can discharge without risk of reversing cells or what-not. But that's not the way it works. The initial grid charge can maximize the charge possible, bring near empty cells off the bottom, and reduce charge imbalance, but it doesn't charge all cells to 100%, as-in 100% 6500 mAh capacity from cell to cell...

I don't know, it can get complicated fast.
 

·
Registered
Joined
·
90 Posts
Discussion Starter #27 (Edited)
OK, I'm getting bogged-down in details, timeline. I was under the impression your IMA is disabled and that you have a pending P1449-78 preventing IMA usage. Doesn't sound like that's the case...

Has your IMA been operating since you last measured stick voltages, in that chart I posted?
You are correct. The IMA has been disabled for a while now, due to the P1449 I believe.
Is your IMA light ON at this point??

If you don't have an IMA light, then it sounds like your pack voltage must be really low - and it has to be really low to get 12V auxiliary starter. When you tried to re-start the car in the driveway, pack voltage was so low it switched to 12V starter, and...
Yes, the IMA light is on and has been on for a while, since the P1449 was set.

...the next day, voltage probably rebounded enough to avoid triggering the 12V start... I don't think you necessarily get an IMA light from this behavior, rather, you'd get one if, after the car goes to charge the pack it can't charge very much without a tap hitting the upper voltage limit...

On your commute, what's the BAT gauge doing?? Was it empty and then re-populated, then jump to the top, or what?? Were you seeing 4 green CHRG bars?
I agree...the voltage likely rebounded enough to get rolling for a little while. I imagined it's similar to a flashlight or something where you can get a little more juice out of it if you turn it off for a while. Now I only get a few miles before the battery light comes on. I'm still able to make the commute though, apparently. Looking at my cig-voltmeter, it held 11.7 almost the entire commute. One thing I have noticed and have not remembered to mention...is that when I start the car in the morning, I do get what looks like 4 bars of regen...but it goes away quickly after a few seconds. Another thing I have been meaning to mention, and I think what you asked about, is that the SOC indicator has been empty for some time as well. When I first threw the P1449, it hung out for a few days wherever it was when the code hit, dropped a bar here and there and seemed after a few days it emptied out entirely. So SOC is showing zero bars, and has been since maybe a few days after the P1449 set.

I would concur with the idea (not based on much technical knowledge) that the pack is VERY low. It hasn't had a grid charge since well before I pulled the sticks out, and at that time it needed one and had been accumulating a number of weeks of self discharge. It's been about a month since then. That's part of the reason I decided I needed to build a grid charger and not wait for HA to send mine back. But supposedly that's on its way right now.

I think I need to start putting the car on a 12v charger at night, if I'm going to keep limping the car along on my commute. I have a Battery Tender I bought for my motorcycle that I think I'm going to start hooking up.

Turning the pack switch OFF and disconnecting the BCM would prevent the DCDC from disabling.
I guess I'm confused again about 12v charging in relation to the DC/DC. First I thought that the DC/DC was the only thing that charged the 12v, and if the DC/DC is disabled then basically you can drive on whatever capacity the 12v battery has, and then you're on the side of the road. Earlier, it seemed you explained that wasn't the case and I would still get some charging with the pack off and BCM disconnected. I suppose I assumed with the pack breaker off, the 12v won't get charged because the DC/DC isn't getting any juice (because the pack is off). I know there's always AC potential at the IMA motor, like an alternator, I just don't know how it gets back to the 12v battery..I guess I assumed it would only get there by way of charging the IMA battery and THEN charging the 12v. But I guess that also requires that the car DECIDE to charge via the IMA motor and I don't know how that decision is made or prevented. My lack of clear understanding on this I think is clouding my confidence on disconnecting the BCM despite multiple directions to do so.

Is turning off the pack breaker and disconnecting the BCM still a prudent move at this point?

It seems like performing the tap level discharge (while still trying to commute with the car) would make the DC-DC / 12v charging problem worse? Or that's WHY I need to disconnect the BCM, so the DC/DC STAYS engaged to charge the 12v battery and be able to continue driving the car?

I've also been thinking about a while back when someone mentioned interrupting that one green wire so the charge voltage stays high...I don't know if that would help or not, but I will probably pass on that unless absolutely required, and only that one person has mentioned it.

One of the common misconceptions among people who grid charge is that, when you first charge the pack all the cells reach 100% full, and then you can discharge without risk of reversing cells or what-not. But that's not the way it works.
My understanding (which certainly has proven incomplete more than once), was that grid charging brings lower cells into balance by allowing normal cells to overcharge, shedding energy as heat, while the lower cells (hopefully) to come up to full charge...by simply charging for a longer period of time with a current low enough not to damage the cells being overcharged. Imprecise, yes but also easy. Also, I thought I read a while back that Peter P. did some extended testing on cell reversal and determined that it was far from the catastrophe once believed? Not to get too far off topic of course.
 

·
Registered
Joined
·
7,165 Posts
You are correct. The IMA has been disabled for a while now, due to the P1449 I believe.... Yes, the IMA light is on and has been on for a while, since the P1449 was set...
OK, so we're more or less on track. But you should take tap measurements again before you do anything, let's take a look at those again. Given your DCDC disable situation, it seems like we should be seeing a lower tap voltage than the voltages you posted before...


I think I need to start putting the car on a 12v charger at night, if I'm going to keep limping the car along on my commute. I have a Battery Tender I bought for my motorcycle that I think I'm going to start hooking up. I guess I'm confused again about 12v charging in relation to the DC/DC.
That probably wouldn't be a bad idea, at least for one night. But if you turn the pack switch OFF and disconnect the BCM you shouldn't have 12V problems. The DCDC can be powered directly (more or less) off the engine and charge the 12V battery, it doesn't need the pack. IF you intend to do the tap discharge while still using the car, then you'll need to bypass the pack in this manner.

Your plug-in voltmeter value sounds too low (11.7V), it might be wrong. If you have a reliable DMM you might want to cross check... If it were truly only 11.7V then the DCDC wouldn't be charging (and your 12V warning light should be on). It should be about 13.8V or higher when DCDC is charging, once in a while dropping to about 12.2V, though that wouldn't happen if you have the headlights ON or possibly in your situation with IMA not functioning...



Is turning off the pack breaker and disconnecting the BCM still a prudent move at this point?
Yes.

It seems like performing the tap level discharge (while still trying to commute with the car) would make the DC-DC / 12v charging problem worse? Or that's WHY I need to disconnect the BCM, so the DC/DC STAYS engaged to charge the 12v battery and be able to continue driving the car?
No on first question, yes on second.

I've also been thinking about a while back when someone mentioned interrupting that one green wire so the charge voltage stays high...I don't know if that would help or not, but I will probably pass on that unless absolutely required, and only that one person has mentioned it.
Pass on that. Not necessary.

My understanding (which certainly has proven incomplete more than once), was that grid charging brings lower cells into balance by allowing normal cells to overcharge, shedding energy as heat, while the lower cells (hopefully) to come up to full charge...by simply charging for a longer period of time with a current low enough not to damage the cells being overcharged. Imprecise, yes but also easy.
Yes, that's the general logic. But degraded cells often, usually, don't charge fully until you discharge them to at least around 1V (at very low current, probably no more than 500mA). And I think overcharge, even at low grid charge rates, can depress subsequent operational voltages... So particularly if one's trying to bring near empty cells into balance with near full cells, via grid charging, the 'cure' can be - not necessarily worse than the 'disease', but at least it can be far from optimal...

Here's an old post/thread I made to try to explain some of this, with graphics. Some of the details/info are wrong, but the general sweep of it is OK:


Also, I thought I read a while back that Peter P. did some extended testing on cell reversal and determined that it was far from the catastrophe once believed? Not to get too far off topic of course.
It was probably Mike D. - 'some' testing, nothing elaborate. Cell reversal isn't a catastrophe, it's not instant death. I think when people hear "reversal" they think - boom, the cell's done. That's not the case, that is, with stock Insight and Civic cells (I think aftermarket cells aren't good with deep discharge or reversal). Typically cells have to be held and discharged with reversed polarity for a while to end up with noticeable damage. But, it's a total crap shoot how reversed a cell can get during a full pack, even only deeper-than-normal discharge, particularly with problem packs that likely have massive charge state imbalance (i.e. a cell at say 75%, a cell at say 10%). Better not to tempt fate when there's other options, like tap-level discharges...

In the early days we started doing full pack deep discharges - I was into really deep, like down to near zero, and I didn't even do grid charges before. I probably did that dozens of times to my packs. I don't recall ever seeing instant death or even noticeable problems. But I had also worked with my sticks on the bench, etc., so my cells were likely already relatively balanced - compared to the typical user with pack problems. Seems like some people tried full pack super-deep discharge and were ending up with failed packs, and that's when the recommendation to grid charge first started to take hold... Personally, I went my own way. Once you're able to grasp just the raw...dynamics of charging and discharging a string of 120 cells, with a sprinkle of knowledge about how these cells behave, it's easy to see that full pack work is a blunt instrument. You don't need a scalpel, rather, a utility knife would do and that's a lot better than a hammer (in a most cases). Once one gets their pack up to snuff, pays attention enough to be reasonably informed about its state/condition/balance etc., then full pack stuff can become a viable option, such as as light maintenance...
 

·
Registered
Joined
·
90 Posts
Discussion Starter #29
Alright, it's good to know I was wrong about how the car would/wouldn't function after disconnecting the BCM. That shouldn't be a surprise (and really isn't) given that a lot of folks have done it. I disconnected it and turned off the pack breaker this morning and drove to work. That seems to be a success, got to work with no problems or additional idiot lights. True, the cheapy voltmeter may not be accurate. Much above idle RPM it reported 14.2v throughout the commute and dropped to 12.7-13.0 at low RPM.

I am a bit frustrated by my grid charging situation; one of the parts I'm waiting on for the DIY charger has been in USPS limbo for a week now. My HA charger, which HA says is OK, is being sent back and both it and the discharger should be here Thursday. The whole point of building it was that I assumed HA would tell me it was bad (I have seen the fan stop working) and I'd be able to build one faster than they would ship mine back. Whatever, I still need to spend some time discharging anyway. Just venting I suppose.

I'll get the tap voltages after work today, report back, and begin the tap level discharge.

Should be OK to put the battery tender on the 12v battery while tap discharging, right?
 

·
Premium Member
2001 5S "Turbo"
Joined
·
10,965 Posts
Yes.
 

·
Registered
Joined
·
90 Posts
Discussion Starter #31
So this was unexpected...with the BCM disconnected and pack breaker off, I still got the battery light on this afternoon’s commute. It seemed to occur after moderate acceleration and went off after a minute or two of cruising at speed.

Tap voltages recorded today, post commute, post disconnecting BCM and IMA pack:

  1. 15.42
  2. 14.77
  3. 15.34
  4. 15.38
  5. 15.12
  6. 14.79
  7. 15.28
  8. 14.98
  9. 14.58
  10. 15.27
 

·
Administrator
Joined
·
12,434 Posts
The OEM DC-DC stops working at over 220V

If the battery is not in circuit the IMA motor generates very high voltages at high rpm and the DC-DC stops until the voltage/rpm drops back down again for 30 seconds or so.

You can use a MEANWELL psu instead which can tolerate higher voltages.
 

·
Registered
Joined
·
90 Posts
Discussion Starter #33
Thanks Peter. I figured it was a well understood and probably expected phenomenon...just not to myself since eq1 explained that disconnecting the BCM would prevent the DC-DC from disengaging...although it’s clear now that he meant that in the specific context that was being discussed.
 

·
Registered
Joined
·
7,165 Posts
^ Yeah, forgot to mention that you don't get any charging at idle (engine speed too slow) and that DCDC temporarily disables over about 4000 RPM, with pack OFF/bypassed...

Tap voltages recorded today, post commute, post disconnecting BCM and IMA pack...
I've re-arranged your earlier (~2 months ago) stick voltage measurements into taps, and created charts comparing those to your contemporary measurements. My first impression was you got massive fast and somewhat uneven self discharge - but now I'm not sure. This thread started with a P1444 HV short problem, and at one point I thought for sure it was a problem with your 'snubber cap' thing. So, I wonder if that was indeed a problem, and if it could still be causing problems, that it's possible a faulty snubber unit can pose a drain on the pack?... I think it could, especially since you've had your pack switched ON, but I'm not positive...

Degraded sticks/cells can self discharge pretty fast, but given the voltage drops across all your taps, from 2 months ago to now - they seem too even and large to me. Usually, it seems, you'd have some cells that simply don't self discharge much, and others that can do it quite rapidly...

In any event, here's those charts:
The top panel compares older resting voltage measurements to the loaded measurements (about -7.5A after 2 minutes), the bottom panel shows older resting voltage to contemporary resting voltage. Grey bars are old resting voltage, outline bars in top panel are loaded, in bottom panel new resting voltage. Numerical labels are the difference.

88324



The top panel seems to establish that the taps were generally charged, with resting voltages at about 15.6V - which is a little on the low side of normal, but still charged - and loaded voltages relatively even and above 14.4V. The lowest tap is at about 14.6V loaded. Even that's pretty typical. You've got some imbalance or mismatch, of one sort or another, but nothing majorly jumps out here - particularly compared to the lower panel.

About two months later, you measure tap voltages again, and now you've got at least three that are virtually empty (2,6,9). These were all among the lowest 5 or so taps in the previous chart. Now, look at the voltage drops for most of the other taps: Tap 1 -0.22V, Tap 3 -0.25, 4 -0.25, 7 -0.30, 10 -0.29... To my eye that just doesn't look right.

The other thing that doesn't look 'right' is that your resting voltages at time 1 are all a little on the low side - they probably should be closer to 15.8V or a bit higher, but instead they're down around 15.6V, and then, by time 2 they're all fairly evenly low again, at around 15.4V... I would normally expect most tap voltages to be around, say, 15.8-16V, dropping to around 15.6V after sitting for a couple months, with a couple miscreants being low due to faster self discharge. You do have a couple (a few) 'miscreants', but my hunch is that you might have a drain on the whole pack, too...

Not sure. I combed through some of my notes and data and really don't have anything to establish what's 'normal'. My expectation is that cells normally get charged to around 1.37V, or 16.4V at tap, in the car, and then voltage will drop no lower than about 1.32V, 15.84V at tap, over a long time period if they just sit, as long as they don't have issues. 1.318V is the 'equilibrium' voltage for the chemistry, so as far as I can tell, they usually don't self discharge below that unless it's a really long time or there's problems. The problem with your data is that I don't know just how charged the cells were to start, at time 1, the 'old measurements'.

Anyway, I think we should probably just try to get the pack going/operational and see if the P1444 comes back.

I think the new plan is this: discharge high taps in the lower chart for 48 hours, do Taps 1,3,7,10, then 4 alone. Afterwards, grid charge to full. Keep your pack disabled until you get it grid charged.

If you want to take it a step further, you can measure tap voltages after the first 48 hour sessions, and then do more, such as a little bit to tap 5 and tap 8 (maybe 24 hours), and more to the previous if voltage is still high. In general, 'taking it further' means bringing the voltage of those taps down to the level of the low taps... FYI, you can't measure a tap's voltage with a shorting pin in place on that tap or on adjacent (overlapping) taps.
 

·
Registered
Joined
·
90 Posts
Discussion Starter #35
Alright...that's what we're gonna do next.
  • Discharge taps 1,3,7 and 10 for 48hrs
  • Discharge tap 4 for 48hrs
  • Measure tap voltages, additional discharge if necessary to get high voltages closer to lower ones
  • Full grid charge
Then...?
  • Measure tap voltages again? Does this tell us anything about any progress?
  • Flip the switch on, reconnect BCM and see how she drives? Something more nuanced for testing?
 

·
Registered
Joined
·
7,165 Posts
Then...?
  • Measure tap voltages again? Does this tell us anything about any progress?
  • Flip the switch on, reconnect BCM and see how she drives? Something more nuanced for testing?
Then? Good question... After the tap short stuff you'd hook everything back up and grid charge. You need the pack switch ON when you grid charge... After the grid charge, in general you'd just remove the charger and go. Or actually, you should probably pull the 12V neg cable to reset all computers, try to clear your DTCs.

If you wanted to be more sure about things you'd monitor the tail-end of the charge, such as by taking tap voltage measurements every 10-15 minutes. That way you'd be able to see how tap voltages change and be more sure that you got all the cells charged. But that can be too much of a hassle for most people... One problem is that it can be hard to tell just when you're nearing the "tail-end," and so hard to know when you should start taking measurements.

Given your situation, where you have at least 3 taps that are near empty, the minimum time it would take to charge to full is about 6800mAh/350mA (grid charger output) = 19 1/2 hours... I think I'd take tap voltage measurements around that time, and if at least one were 17.4V I'd consider that the tail-end of the charge and take measurements starting at every 15 minutes and decreasing the time interval as the charge progresses... Once all tap voltages start to drop, or at least peak and hold for maybe an hour, I'd call that done... This "tail-end" period can take a while, it's not convenient at all to do this... You can be standing there for a few hours, measuring tap voltages...

If things are 'good', you should be seeing fairly even voltages, peaking at about the same time, etc... Some of this depends on just how even you get your taps during the tap short discharges, and even then, it can be hard to say because voltages can be misleading...

I think the minimum I would advise is take a pack voltage measurement at about 19-20 hours and if it's at least 174V, go another 4 hours then quit. If it's below 174V, let it keep charging, come back a bit later and take another measurement: Is it at least 174V? If so, go another 4 hours... Something like that... I don't think I'd charge more than 28 hours total. If the pack voltage were above 174V at first check? I don't know, I'd stick around and take some measurements to try to spot peak voltage, and once voltage peaked I'd let it go for an hour, maybe 2...

Incidentally, when you do the grid charge make sure it's not too cold - ideally ambient would be up around 60F degrees. I wouldn't do it unless temp were above 45F... Use a space heater if you have to. [I now see you're in Florida, probably won't have a problem with temp. I was thinking you were in the NE.]

After all this, it's hard to say where we'll be. There's some things you could do to gauge the condition of your pack, but it all takes time and effort. You could get the P1444 again and you'd need to figure that out... If you don't get the P1444, I imagine you're gonna get the P1449 again sooner or later, maybe within a month or two? Maybe sooner? Hard to say.

Personally, if the IMA actually works after all this, I think one of the first things I'd do is find a good hill and use assist until the pack is empty - neg recal. How many 'bars' did you get before the neg recal? Did the BAT gauge plummet from say the middle, or did it get to 3 bars from bottom? If you can get it to 3 bars from bottom, at least we'd know that the pack/all cells can be charged to near full. I'd then let the car charge up the pack and I guess drive as normal, see how long you can go without seeing any problems... When 'letting the car charge up the pack,' try to note where the BAT gauge bars are when they 'decide' to pop-up to 19 bars - if the pack were decently functional, in most respects, you'd be able to charge for quite a while and the BAT gauge wouldn't pop-up from the bottom or middle to the top, rather, you'd see a steady charge all the way to 19 bars...

I don't know, lots of possibilities...
 

·
Registered
Joined
·
90 Posts
Discussion Starter #37
Alright, first set of discharge is done, now discharging tap 4. Here is where we started and where we are today, taps having been discharged are bold and underlined:

Tap11-18-202048hrs disch. (11-20)Difference on 11-20
115.4214.57-0.85
214.7714.79-0.02
315.3413.08-2.26
415.3815.380.00
515.1214.79-0.33
614.7914.70-0.09
715.2813.08-2.20
814.9814.60-0.38
914.5814.51-0.07
1015.2712.87-2.40

It looks like the targeted taps for round 1 are going down, I was a little concerned that I’d find one of my clips wasn’t making contact or something silly like that. Looking good though, I think. Except, seeing somewhat consistent discharge on taps 3, 7 and 10 about 2.3v, I am surprised that tap 1 only lost less than half that. I am not sure that means anything significant.

Is 48hrs enough to look at the taps NOT being discharged to say anything positive or negative about their rates of self discharge? Tap 4 (now discharging) may be a good baseline to compare others to, since it didn't discharge significantly at all over 48hrs?

In other news, my HA charger and discharger just showed up yesterday. They said they had not received my harness, I informed them they were mistaken and should put a new one in the return package. They complied. They said nothing was wrong with my grid charger, but it appears they may have replaced the output connector. I do recall it was kind of crusty. Even they were suspecting the harness as a possible cause of the P1444, and I have to say although I could find nothing wrong with the harness, the code hasn't come back since the harness was removed. And now I have a new one. Bad harness could explain zero volts on the charger, discharger and a HV short...potentially. Maybe that's wishful thinking. In any case, I supposedly now have a working charger, discharger and a new harness. I think I'll get all the tap level discharging done before I do anything with those things though.

Anyway, now discharging tap 4... after 48hrs I'll measure all the taps again and begin discharging taps 5 and 8 for 24hrs unless the above numbers suggest a different course of action.
 

·
Registered
Joined
·
7,165 Posts
...seeing somewhat consistent discharge on taps 3, 7 and 10 about -2.3v, I am surprised that tap 1 only lost less than half that. I am not sure that means anything significant.
Nothing really definitive we can get from that. It's hard to know the charge state of individual NiMH cells in a string of 12 based on total voltage. Your low voltages could be due to all cells being, say, 13.08 / 12= 1.09V, or some other combination, say 11 cells at 1.18V and 1 at zero, or perhaps 10 at 1.17V and 2 at 0.7V... My guess is it'd be something closer to the latter.

The 14.57V tap - all the cells are likely just around 1.2V...

I'm not really sure how to deduce charge state from total voltages on possibly putzed-up cells. If they were my cells I'd be reasonably sure that they were all close to empty. But degraded cells can have a lot of 'capacity' locked-up at low voltages, I think within the ~1V to 1.2V range...

Here's a graph that kind of illustrates this idea. Some days ago I pulled my 12V stick-based NiMH battery and tested it, worked with it. Although it was charged to a fairly high charge state - about 5200mAh out of say 6300mAh, so ~83% - the cells were so degraded that most of the energy could only be pulled out at relatively low current AND lower-than-normal voltages.

88395



The colored curves are for pairs of cells, blue curve is total capacity, black is current. The discharge starts at 'only' 6.5 amps - that's relatively low for our cells, considering they're supposed to handle like up to 90 amp discharges. But these cells can't handle it. They only 'put out' about 1250mAh at only 6.5 amps, i.e. about 20% of total capacity. I step the current down from thereon out, to 3.2 amps, 1.6 amps, 600mA, 300mA, and finally I think 100mA. At the end of it all I indeed 'pull out' a lot of capacity - but it's all between about 1V and 1.2V (2V to 2.4V in the graph) and at these low currents; most of it should be pulled above 1.2V, and at higher currents...

So basically, it's really hard for me to tell these days whether low voltages reflect low charge state, or whether they reflect degraded cells. In general, I think I'd say you can't tell how discharged a given cell is until you get down to maybe 1V at super low current, like, if you can drag your taps down to 12V with a tap discharge, I think we could be...at least more confident that all the cells have nearly completely discharged. Between 12V and 14.4V I'm just not sure...


Is 48hrs enough to look at the taps NOT being discharged to say anything positive or negative about their rates of self discharge? Tap 4 (now discharging) may be a good baseline to compare others to, since it didn't discharge significantly at all over 48hrs?
Not sure I understand the question. One thing that occurred to me is that, IF there were a problem with your 'snubber caps' (the HV short problem) and turning the pack switch OFF effects that potential short, then since you've turned your pack switch OFF you might not be having that problem. In other words, turning the pack switch OFF might bypass a short or problem with your snubber caps, so you might not see the discharge that you have - IF that were the problem. It might be worth testing later - turn the pack switch back ON for 24 hours and see if your voltages drop more than they have with the pack switch OFF...

...Even they were suspecting the harness as a possible cause of the P1444, and I have to say although I could find nothing wrong with the harness, the code hasn't come back since the harness was removed.
I thought we ruled-out the harness at some point earlier. But in any case I'm not prepared to revisit all that at this point.

All-in-all, things look pretty much on track, so continue with 'the plan'...
 

·
Registered
Joined
·
90 Posts
Discussion Starter #39
Right on. I had pretty much ruled out the harness myself, because there was absolutely nothing I could find wrong with it. No pinch points, nowhere it could be grounded out. Continuity where there should be and none where it shouldnt be. You’re right though, lets not go down that route just yet. Will post back in a couple days with more info.
 

·
Registered
Joined
·
90 Posts
Discussion Starter #40
Ok, here we go. Tap 4 has discharged for 48 hours. Everything is looking pretty even, so I removed the short from tap 4, and I'll move on to the next step tomorrow. Here's where we're at. We have some voltages actually going up, which I didn't expect but maybe that's rebound and not unusual? Those are in red.

Tap11-18-202048hrs disch. (11-20)Difference on 11-20+48hrs disch.Diff on 11-22
115.4214.57-0.8514.73+0.16
214.7714.79-0.0214.68-0.11
315.3413.08-2.2614.18+1.10
415.3815.380.0013.23-2.15
515.1214.79-0.3314.68-0.11
614.7914.70-0.0914.63-0.07
715.2813.08-2.2013.88+0.80
814.9814.60-0.3814.52-0.08
914.5814.51-0.0714.41-0.10
1015.2712.87-2.4013.68-0.81
 
21 - 40 of 49 Posts
Top