openinverter forum

ianlighting wrote: ↑Mon Apr 28, 2025 2:59 pm So when leaf spy says ‘cell 55’, that’s the best granularity it can provide, but I actually have 3 cells that are ‘55’, and no way of knowing which of the 3 is suspect.

those 3 pouches in parallel are bonded for life and the other two have suffered more if only one would have a defect

marcexec wrote: ↑Mon Apr 28, 2025 3:53 pm those 3 pouches in parallel are bonded for life and the other two have suffered more if only one would have a defect

Thanks. As far as I know it’s not possible to break out the individual cells (or set of 3) from these modules. Do you know if that’s true? If it turns out I need to replace, I assume it’s an entire module. But if anyone knows if individual cells can be extracted, that might give another option.

Not sure about separating them but surely the BMS has individual cell taps. Even trickle charging the set of 3 will work to balance as they'll all sit at the same voltage anyway.

Jacobsmess wrote: ↑Mon Apr 28, 2025 4:30 pm Not sure about separating them but surely the BMS has individual cell taps. Even trickle charging the set of 3 will work to balance as they'll all sit at the same voltage anyway.

It has 1 wire per set of 3 cells. so 9 wires on the tall module for its 27 cells.

Yes I can access ‘cell 55’ and charge that 1 (which is actually 3). Although I need to acquire a charger first to do this.

So any recommendations from anyone for a charger that won’t break the bank? Dont anticipate doing this often, so feature rich is not necessary.

A variable bench PSU should be fine just to trickle it. Current control is useful or just a low output

Yeah you’re over thinking it. That parallel group is reading lower. Just top it back up, no need to split it up. It’s effectively “one cell” at this point.

A basic bench top variable dc power supply is all you need.

I have a 3amp 120v dc one from Amazon for trickle charging large modules, it was 80$ cad.

But no need for one that goes that high. plenty of decent ones on Amazon.

Or check marketplace for a old used one

Bratitude wrote: ↑Mon Apr 28, 2025 5:17 pm Yeah you’re over thinking it.

Yeah. Hazard when everything I touch on this adventure is new virgin territory for me so I try hard (and fail sometimes) to avoid setting a foot wrong. I have no problem with owning my cock-ups, it’s the healthy respect for the dangers involved that make me overly cautious at times.

ianlighting wrote: ↑Mon Apr 28, 2025 4:37 pm So any recommendations from anyone for a charger that won’t break the bank? Dont anticipate doing this often, so feature rich is not necessary.

You can use any old charger from a mobile phone with an output of 5 volts and 2 amperes and watch the voltmeter to perfectly tighten the 55 cell.
You won't have to buy anything!

Romale wrote: ↑Mon Apr 28, 2025 5:28 pm You can use any old charger from a mobile phone with an output of 5 volts and 2 amperes and watch the voltmeter to perfectly tighten the 55 cell.
You won't have to buy anything!

Hmmm…. Not sure where I could find anything like that…

Sanity check before I get into serious charging / balancing efforts.

In a classic leaf module you have access to both positive and negative terminals. Makes measuring and balancing if necessary pretty straight forward. Nice easy to use m8 bolts to connect with aswell. Unfortunately this no what I’m working with.

In my lovely 9s3p tall leaf module I only have access to each positive pin for each of the 9 series cells. Plus the main positive and negative terminals for the whole module. I can’t specifically see a separate negative pin for each cell.
Since I cannot see the internal wiring or construction within the module’s welded case I would appreciate thoughts on the safety of using the next positive pin from the cell in series after my weak cell as my negative point in order to only add volts to the one I’m interested in. There is no way to just deal with the weak cell in isolation from the rest of the series.

Someone will be along to confirm but as I'd understand. You'd use the positive of cell 55 and the positive of cell 56. This would create a circuit across cell 55 only and this bump its voltage up. You can use a multimeter to find those on the connector (using jump leads so you don't short out any cells accidently).

Jacobsmess wrote: ↑Tue Apr 29, 2025 2:14 pm Someone will be along to confirm but as I'd understand. You'd use the positive of cell 55 and the positive of cell 56. This would create a circuit across cell 55 only and this bump its voltage up. You can use a multimeter to find those on the connector (using jump leads so you don't short out any cells accidently).

Yes that’s the idea. Just a slight unknown what’s actually inside these modules.

in general, the minus of cell 55 should be the plus of cell 54. why did you decide on 56??

As stated, the positive of the next cell is the negative of the previous. Or vice versa.

They are in SERIES.

Just remember, these modules can deliver infinite current, and I’m not sure the cell taps are fused, so any shorts will melt the cell taps pretty quick

ianlighting wrote: ↑Tue Apr 29, 2025 11:09 am Sanity check before I get into serious charging / balancing efforts.

In a classic leaf module you have access to both positive and negative terminals. Makes measuring and balancing if necessary pretty straight forward. Nice easy to use m8 bolts to connect with aswell. Unfortunately this no what I’m working with.

In my lovely 9s3p tall leaf module I only have access to each positive pin for each of the 9 series cells. Plus the main positive and negative terminals for the whole module. I can’t specifically see a separate negative pin for each cell.

IMG_7951.jpeg

Since I cannot see the internal wiring or construction within the module’s welded case I would appreciate thoughts on the safety of using the next positive pin from the cell in series after my weak cell as my negative point in order to only add volts to the one I’m interested in. There is no way to just deal with the weak cell in isolation from the rest of the series.

The major problem with using the link between the negative of the previous cell and positive of the next cell in series, as a common voltage reference point, is any additional resistance in that link will adversely affect the voltage read on each of those cells. A dirty or loose connection, a length of cable rather than a short copper busbar, will upset the readings ..... it plays absolute havoc with balancers if they aren't across the positive and negative of each cell, the voltage drops when it is charging the induction coil or capacitor, yet rises when it ties to discharge. If the cell voltage monitor is a sperate unit, the readings are continually changing making them unreadable.

The fix is to have the balancer across both terminals of each cell, and anywhere a cable is involved with joining a negative to a positive in a series connection, a wire from both the negative end and the positive end, with a max 5 amp fuse at the terminal end, linked at the voltage sensing point on the unit. The fuse is to limit the current that can pass through the voltage sensing cables, if a high resistance at any part of the link was to occur, the current would attempt to use the voltage sensing cables as an easier current path, melt the insulation on the wire and potentially start a fire. A blown fuse is a sure indicator that something is wrong with that link .....

T1 Terry

Thanks for all the comments.

T1Terry wrote: ↑Wed Apr 30, 2025 5:56 am The major problem with using the link between the negative of the previous cell and positive of the next cell in series, as a common voltage reference point, is any additional resistance in that link will adversely affect the voltage read on each of those cells. A dirty or loose connection, a length of cable rather than a short copper busbar, will upset the readings .....

It’s not an ideal situation, but I’m stuck with what Nissan made. If I want to attempt to charge up the weak cell, my only way to access the cell in question is as described by using the previous positive as its negative.

But this is not a long term permanent thing. It’s a one-off, carefully monitored to try to charge up the cell.

T1Terry wrote: ↑Wed Apr 30, 2025 5:56 am
The fix is to have the balancer across both terminals of each cell, and anywhere a cable is involved with joining a negative to a positive in a series connection…

All the cell’s series connections are internal to the module and inaccessible. these are manufactured by Nissan and not something I’ve messed with. It’s true that the module to module connections are built by me. The majority are using repurposed Nissan leaf busbars, but some 35mm and 50mm cabling.

T1Terry wrote: ↑Wed Apr 30, 2025 5:56 am
The fix is to have the balancer across both terminals of each cell, and anywhere a cable is involved with joining a negative to a positive in a series connection, a wire from both the negative end and the positive end, with a max 5 amp fuse at the terminal end, linked at the voltage sensing point on the unit.

Adding 96 inline fuses for each of the BMS wire cell taps would be quite an undertaking. Most of the BMS wiring is also OEM Nissan leaf, same original cabling and connections. My hope had been that by sticking to OEM, I could reduce the need for things they did not include in their original build, eg 96 fuses.

The exception to this is the 6 modules I have put on extended wires. This would be a total of 30 cell taps. So adding fuses to these might be the right thing to do if I can squeeze it in.


Appreciate the input.

You can likely carefully charge the cell through these wires to get it back up to the other cells. But you will have to keep the charge current very low.

The Nissan BMS should be able to balance this cell when it's active.
This seems to be full circle again. Why is it low compared to the others? Unless there's a good reason why this one cell is low there's likely an issue with the cell. If there is an issue, now is the time to swap it out with a known good cell / module.

Is it worth trying to recover this low cell? Depends on what your good enough looks like for your project.

jrbe wrote: ↑Wed Apr 30, 2025 9:24 am You can likely carefully charge the cell through these wires to get it back up to the other cells. But you will have to keep the charge current very low.

The Nissan BMS should be able to balance this cell when it's active.
This seems to be full circle again. Why is it low compared to the others? Unless there's a good reason why this one cell is low there's likely an issue with the cell. If there is an issue, now is the time to swap it out with a known good cell / module.

Is it worth trying to recover this low cell? Depends on what your good enough looks like for your project.

I’ve put the feelers out for a replacement module, but they are very rare compared to 24/30/40kwh leaf modules.

So I will try to carefully charge and see how it behaves.

Project objective - I definitely started on this with the objective of a reasonable range EV VW T2. Which put pressure on me to complete by about now But reality has made itself known and so the VW has just been downgraded in our home to ‘Ian’s project’, rather than ‘our campervan’. The objective remains the same, but the pressure is reduced as it’ll now get done when it gets done.

Lots of retorhic here, just top it up slowly, better balanced temporarily than not at all if your testing/ in your final stages.

It’ll give you time to work on everything else as you wait for a replacement module.

You are now driving which is a huge milestone, now let’s get you driving safely and reliably:)

Manually charging / balancing.

Went ahead and bought a small battery charger / balancer for £30 - the ‘imax B6’. Only wanted to charge up the single weak cell, but thought maybe it’s other balancing features might be useful in the future.

So veeery slowly brought it up to match the rest over a few hours. Actually did it in 2 goes on 2 separate days as I didn’t want to leave it unattended.

Once it was matched, put things back together again and did a few goes around the block. That means 20 - 30mph, just to check everything still worked.

I could see the suspect cell was starting to drop a little compared to the others on Leaf Spy (range shown as 36mV) Nothing dramatic, but noticeable. So assumed this was an indication that the balancing was not going to work long term, which was kind of what I was expecting anyway.

But I’ve actually been pleasantly surprised since then. After some wiring tidying to get the 12V lights etc working with the VCU instead of the OEM mechanicals, went for a faster drive on a nearby main road. Variance dropped to 26mV…
Best I’ve seen it is 20mV, which is pretty amazing.

Following that, doing a slow AC charge, the spread has extended to 38mV, with the weak one lagging behind the others. But once driving, they equalise as the capacity is reduced.

I’ve just done a 30mile round trip, and this is how the battery is looking, according to Leaf Spy now…

And battery temp while I’m at it…
What does it mean long term? Who knows! I don’t intend to be delving into the battery to manually charge up every few weeks, so will have to see if the BMS can stay on top of this for me, but looks kind of promising so far.


Range

So this is the slight disappointment I need to look into next. Based on my 30 mile test (prob about 70% at cruising 50 - 60mph, the rest at 30mph, stop/start at junctions), seems like I consume 1% per mile. So even I can work out that means about 100 mile range assuming you go down to the wire.

But what I’m not certain of is whether the BMS is skewing the numbers in some way. I have a 40kWh BMS with a 60kWh battery. The 40 was a 2p, while the 60 is a 3p. I honestly don’t have a clear idea whether that affects what it reports, or if it is roughly accurate. Would love it to be 120+

More research req…

(Tyre pressures were checked before test, and I avoided regen mode when cruising)

Next step

Only been playing with home AC charging so far, but soon will need to venture forth to find a chademo charger in the wild to see how that goes. Our local BP has one, but it’s also often the place where the local police get their snacks and I don’t want to invite attention just yet! The engine bay wiring isn’t bad now, but the cabin is a bit tangly.

…so then I couldn’t resist when I realised a quiet local hotel had a Chademo, had to try it!

First attempt failed. Think it was because I plugged in with the car ignition off. At home with AC charging this works. It wakes the car up and charging commences. But with public Chademo, at least in this case, whatever handshaking goes on required the car to be ‘on’.

I managed to make the screen go to a black screen of death with a big Red Cross, with some message about voltage too low. It then went through a 5 minute reboot cycle, before letting me try again, which then worked.

You have to wonder if there can’t be a more elegant failure mode that doesn’t require a reboot of the charger. But then I’m not the software engineer.

I had my perverse fun of spending a few £££ for something I can get for free at home with sun and solar PV, and then drove home to use up some of what I’d just purchased.

In terms of Leaf Spy, mid charge, there’s 1 cell (67) that seems eager to charge up more quickly than the others. This peak disappears after the charge ends…

Warmest battery temperature was 23.5deg.

It was a pretty short charge, only 10 mins. But another useful step.

Once back home, things looked how I expected

…reverting back to my Leaf Spy image of in-progress Chademo charging yesterday, I’ve realised there’s a pattern in the peaks and troughs.

More or less all the taller peaks are from the first cell in a module. Here it is with green lines tagged to the start of each module. Also M(edium), S(hort, T(all) added to identify the different types of Leaf module.


So the extra long spike is module 11. This is the first module on extended cell tap wires and extended HV cable. All the modules to the right of that point also have similar length extended cell taps, but they don’t spike in the same way. All the HV from this point are short cable or original Leaf busbars.

I’m sure that’s not coincidence, just not sure what, if anything I should do with this revelation now I’ve seen it. I guess I’ll dive back into the battery box and double check the cell tap connections are good, at least on module 11.