Page 24 of 28

Re: Prelude conversion project - some questions

Posted: Tue, 26 May 2020, 14:28
by jonescg
Based on the roughly 18 Ah that went back into the cells, and the fact the voltage isn't too far off the rest of them, I reckon the cell is out of order.

Re: Prelude conversion project - some questions

Posted: Tue, 26 May 2020, 21:59
by jonescg
I reckon the cell has been in contact with leaking coolant 😟. Reservoir is low and there's an accumulation of crud around some of the boltholes under the battery pack. Might need to abandon the cooling system to save the pack.

Re: Prelude conversion project - some questions

Posted: Tue, 26 May 2020, 22:04
by brendon_m
Hopefully it's just the hose connections on the outside of the modules but it's not sounding promising. :(

Re: Prelude conversion project - some questions

Posted: Wed, 27 May 2020, 01:01
by antiscab
yikes.
can the car run a module down?

Re: Prelude conversion project - some questions

Posted: Wed, 27 May 2020, 06:55
by jonescg
It will run with two modules down but I hope I only need to abandon one cell group if possible.

Re: Prelude conversion project - some questions

Posted: Wed, 27 May 2020, 07:44
by francisco.shi
If the cell goes up together with the rest I would expect the cell may be ok.
Any chance it was discharged at a different rate than the others?
Maybe the bypass is continuously on for that cell?
What BMS are you using?
Does it have one board per cell or one board does many cells?
I would expect the cells to be sealed on the bottom so how can a coolant leak damage the cells?

Re: Prelude conversion project - some questions

Posted: Wed, 27 May 2020, 09:19
by jonescg
Responses in bold
francisco.shi wrote: Wed, 27 May 2020, 07:44 If the cell goes up together with the rest I would expect the cell may be ok.
I didn't fully charge the battery - juse enough to get the car to my workshop. It is still sitting 100 mV lower than the rest, which could be as much as 20 or 30 Ah out.
Any chance it was discharged at a different rate than the others?
Almost certainly. The other cells weren't being discharged at all while this one found it's way down.
Maybe the bypass is continuously on for that cell?
I will have to check the functionality of the BMS module when I pull it out
What BMS are you using?
ZEVA with EVMS3
Does it have one board per cell or one board does many cells?
Each module manages 12 cells. Cell 6 of Module 3 is the offending cell. It happens to be located at the lowest part of the battery pack, and would be the first to be compromised by a coolant leak.
I would expect the cells to be sealed on the bottom so how can a coolant leak damage the cells?
The cells are sealed, but in a polymer/alfoil pouch which is highly susceptible to corrosion should it ever get wet with an aqueous liquid. If a single pouch cell in that group of 10 paralleled cells were to be drained by means of water spanning the anode and cathode, all 10 cells will be drained. If this is the case and that cell has been compromised, I will just drain the cell to 0 V and bypass it with a buslink. Then stop using liquid coolant in the battery - it's probably too risky.

Re: Prelude conversion project - some questions

Posted: Wed, 27 May 2020, 11:37
by francisco.shi
From the pictures of your pack it looks like the coolant is at the bottom so I assume if there are any leaks it would have to first fill up the box before it got to the terminals which are at the top. From memory I think your cells are epoxied to the bottom coolant plate so the foil should not get wet unless the liquid level goes up to over half the height of the cells. If the liquid got to the terminals you would have got an earth leak and would have completely destroyed the bus bars and anything around it.
It sounds somewhat unlikely it is coolant damage.
Send some pictures when you pull it apart .

Re: Prelude conversion project - some questions

Posted: Wed, 27 May 2020, 11:41
by jonescg
It can be drawn up by capillary action, and it can travel a long way through narrow seams. It would be just my luck that it's able to pond right where I least expected it to. So for example it might be possible for it to work it's way up a screw hole and into the bottom of the module.

Re: Prelude conversion project - some questions

Posted: Wed, 27 May 2020, 13:30
by T1 Terry
I would suspect the "bypass" failing as the most likely cause. A short or cell failure would have pulled the cell voltage down much lower than 3v and any attempt to recharge would not help. It looks as though it is a small discharge rate and that is more likely to be "bypass resistor" slowly draining the cell.
This problem lends support for not using boards across the cells themselves but rather running cables back to the centrally mounted balancer module. Much easier to put a clamp meter over the cable from that cell to check if a current is being drawn when there shouldn't be any drain via that cable. Also makes it much easier to recharge an individual cell group if needed.

T1 Terry

Re: Prelude conversion project - some questions

Posted: Wed, 27 May 2020, 13:57
by jonescg
Unfortunately I can't test that while the battery is in pieces. Not without a lot of difficulty anyway.

Re: Prelude conversion project - some questions

Posted: Thu, 28 May 2020, 08:48
by blwnhr
jonescg wrote: Tue, 26 May 2020, 21:59 I reckon the cell has been in contact with leaking coolant 😟. Reservoir is low and there's an accumulation of crud around some of the boltholes under the battery pack. Might need to abandon the cooling system to save the pack.
Have you seen the Kemo water switch? This may have given an early warning of leaking coolant.
https://www.kemo-electronic.de/en/House ... 2-V-DC.php

Re: Prelude conversion project - some questions

Posted: Thu, 28 May 2020, 09:02
by francisco.shi
A better option would be to check for earth leak from the battery. The main problem is deciding how much time to spend to cover every possible mode of failure.
Generally you consider it important only when it happens.

Re: Prelude conversion project - some questions

Posted: Thu, 28 May 2020, 09:07
by jonescg
Yeah I did think about installing something like this, but I really need to determine if the coolant is leaking or not first. If it's leaking, there's really little I can do about it since the cells are effectively potted in place.

The battery has been showing a low current leakage path (~11 mA) to chassis ever since it was first built...

Re: Prelude conversion project - some questions

Posted: Thu, 28 May 2020, 09:10
by francisco.shi
The cells are potted to one of the coolant plates. The leak wouldn't be on the surface the cells are potted to. So if there is a leak wouldn't it be below the cells?

Re: Prelude conversion project - some questions

Posted: Thu, 28 May 2020, 09:21
by jonescg
It's still possible for coolant to work it's way up through a screw hole and into the base of a module. I won't know until it all comes out though.

Re: Prelude conversion project - some questions

Posted: Wed, 03 Jun 2020, 10:37
by jonescg
I've had the Prelude in the shed for about a week now, and it seems the offending cell is slowly dropping. I also noticed that the 12 V auxiliary battery was consistently flat, but this could be mainly due to how little Francess was driving. I charged it up with a power supply and left it overnight, and the car started up OK this morning. So hopefully that's dealt with. But the cell module is still showing a low cell (and dropping) while the temperature sensor on that particular module is still reading 8'C lower than the rest. As these are NTC thermistors, it suggests the resistance is increasing. The sensor happens to be located beside the offending group of cells, so I wonder if the two are related. It would be great if it was a naughty BMS module, but we'll see.

I'll pull the battery out this weekend and see what I can find...

Re: Prelude conversion project - Things went bad. Very bad.

Posted: Sat, 06 Jun 2020, 17:01
by jonescg
This day would come.

I went through the process of pulling the battery pack out of the car. After about two hours (video yet to come), I was able to peel the lid off and take a look inside...



This battery pack is in trouble. Water and pouch cells need some distance, and the slow leak which had developed has caused the moisture to spread and condense throughout the pack.
20200606_124510.jpg
20200606_124510.jpg (209.74 KiB) Viewed 1481 times
So I'm personally stumping up for a whole new battery. I'll probably need to sell the race bike. The modules which aren't in a bad way are still useful, so I'll happily sell them for any keen tinkerers, but they won't be any use for a vehicle unless they're easy to access. Ideal for a shed storage project.

I think I'm going to call it - sorry folks, no more battery building for me. It's just too much risk :( .

Re: Prelude conversion project - some questions

Posted: Sat, 06 Jun 2020, 17:58
by brendon_m
Ouch, I'm glad I decided against water cooling. Too much risk (although my build is low power enough to not need it anyway

Re: Prelude conversion project - some questions

Posted: Sat, 06 Jun 2020, 19:26
by jonescg
I will say that no Nissan Leaf or iMiEV will ever suffer this fate...

Re: Prelude conversion project - some questions

Posted: Sat, 06 Jun 2020, 19:45
by brendon_m
Any clue where the actual leak was?

Re: Prelude conversion project - some questions

Posted: Sat, 06 Jun 2020, 19:50
by Rusdy
Hi Chris, was the leak come from the hose connection closest to the busted cells?

Was the water seeps from the bottom (the makrolon interface to bottom plate not water proof I believe?), or due to condensation?

I wonder why the cells swelled though. I thought water damage can't damage lithium cells? Unless the corrosion made some cells drained beyond the low voltage limit?

I didn't realise how much force swelling pouch cells can exert. I can see the makrolon cracked in the video...

Sorry to hear that you'll need to fork out new batteries, I hope your learning can be great lessons for others. :oops:

Re: Prelude conversion project - some questions

Posted: Sat, 06 Jun 2020, 19:56
by jonescg
No idea, but it's clear that the leak in module 3 started right underneath cell group 6. Once the base plates dry out I might see some evidence of corrosion, but it's most likely it was slowly leaking through the screw holes and up to the walls of the enclosures. But the killer was the condensation which was allowed to accumulate and circulate throughout the whole pack. The lids helped it last as long as it did, but there were still plenty of opportunities for it to get in. In hindsight the better option would have been to get the cooling plates welded up and securing the cells with a completely separate structure, but I was also trying to keep the total height to a minimum. The modules in the boot were fine - the coolant lines were blocked with crud, so they never got a chance to leak...

Water will absolutely kill a cell in short order. The pouches are made of a bi-layer of polythene and aluminium, but the polythene is peppered with microscopic holes. These allow the aluminium behind it to corrode, eventually penetrating the cells and spoiling them.

Re: Prelude conversion project - some questions

Posted: Sun, 07 Jun 2020, 05:15
by 4Springs
jonescg wrote: Sat, 06 Jun 2020, 19:56 Water will absolutely kill a cell in short order. The pouches are made of a bi-layer of polythene and aluminium, but the polythene is peppered with microscopic holes. These allow the aluminium behind it to corrode, eventually penetrating the cells and spoiling them.
Any idea how much water is bad for a cell? I'm thinking of where mine are right now, where they are probably getting at thin film of condensation on these cold mornings.

Re: Prelude conversion project - some questions

Posted: Sun, 07 Jun 2020, 08:33
by jonescg
About that much. Yeah.