openinverter forum

More Batteries!!!

Thanks to Marty on the forum here, I got hold of a BMW X5 24kWh pack Still trying to hunt down shorter length Tesla Model 3 module for the other pack, but in the mean time I wanted a handy smaller pack that was easily controlled and a working BMS option for testing and getting things proven. The BMW packs suit with work by Damien and others on the SBox reverse engineering and then Tom's SimpBMS to keep an eye on things.

This will be used to test the Mitsubishi heater and A/C compressor, and have some Leaf stacks to test with a second Zombie vcu. It may make its way into this project for initial drive tests. I believe the SBox is fully integrated into the Zombie including shunt duties, which would be tidy. Although the SBox component ratings don't look like they'll suit something this heavy without limiting some parameters!


Not far off fitting under the car as it is, but wasn't the intention

The benefit of the BMW packs, only a couple of hours got lids and casing removed, Tom's SimpBMS and a Nextion display screen connected up, and an Arduino Due running Damien's SBox control sketch. Shout out to Greg for his excellent videos on connecting to the BMW packs too!

Actual connector DC voltage matches up with both the SimpBMS report and the SBox controller output

First time using SimpBMS so didn't dive into the settings but obviously need to adjust the cell numbers expected. Thanks to Andy and Tom for the tweak needed to allow Simp work with the 12 module X5 pack

did you see that?


https://www.ebay.de/itm/285744978081?it ... R8iGvdbJYw

muehlpower wrote: ↑Sun Mar 17, 2024 5:15 pm did you see that?

Hi Muehlpower,
I did thanks, and I've sent an inquiry. Price per module is too high for me. Others I found won't split the pack. But will keep trying. Thanks

Hi Nkeirnan,

Amazing build, somehow I missed it but I've just read the whole thing in one go. Extremely good quality of work and great detail for us to follow along.

It's like you read my mind, connecting the gs450h to a transfer case for 4wd and using short m3 modules under the floor.

One question for what I'm thinking of doing. I see you mentioned the height of your battery boxes as 224mm. Could you throw a tape over the length and width of your battery boxes please?

Looking forward to seeing more progress soon, Thanks.

Evkiwi wrote: ↑Sat Mar 30, 2024 10:16 pm One question for what I'm thinking of doing. I see you mentioned the height of your battery boxes as 224mm. Could you throw a tape over the length and width of your battery boxes please?

Thanks Evkiwi,

The main outer body dimensions of each box are 1395mm x 342mm x 224mm. with the way I've overlapped the lid to the front and back, overall length at the lid is 1450mm.

This is allowing space in the width for the tabs etc on the sides of the battery modules. Could be narrowed approx 30mm if those tabs were trimmed down. It also allows space front and back for coolant and busbar connections. If space was tight and a smart connection detail used, there could be scope to shorten the boxes too.

Thanks Nkiernan,

That's great information,

Definitely subscribed to thus build

Really interested in this project as I have an L322 I want to convert. You have done an amazing job. very detailed.

So, do you have an idea on top speed and average distance able to go on a charge?

Hoping to keep the functionality so have you managed to communicate with the L322s ECU?

I am still doing the fun stuff repairing the rear arches and sill.

nemmi69 wrote: ↑Fri Apr 12, 2024 6:09 am Hoping to keep the functionality so have you managed to communicate with the L322s ECU?

That's an interesting question!!! I'll hopefully find out very soon. Have some CAN logs but pretty sure I'll need more before I'm done. Make sure you get your CAN logs!

Is yours a 3.6?

Just read through your entire build awesome work there

nkiernan wrote: ↑Wed Apr 17, 2024 8:05 pm That's an interesting question!!! I'll hopefully find out very soon. Have some CAN logs but pretty sure I'll need more before I'm done. Make sure you get your CAN logs!

Is yours a 3.6?

My beasty is currently a 4.2SC so also trying to match its power level (torque is not an issue obviously). Planning on using a Tesla motor and the input and secondary gear set only to reduce down for the gearbox. Having fun finding information out the input shaft location in relation to the bellhousing of the Rangies box.

HV measurement precautions!

Looking for some help! It's been a while and I have a backlog of updates to add here shortly. Right now, I'm back finalising the build of the HVJB and running some final bench tests to make sure PCS, DC-DC, charging etc are funcitonal. While doing random voltage measurements to check for HV isolation issues, I noticed some voltage readings between HV+ busbar and the HVJB case/chassis. Hoping I can get some guidance or suggestions from the forum just to be cautious.

I didn't expect any voltage reading between HV- busbar or HV+ busbar and the casing/chassis.


The HVJB for this build houses the Tesla Model 3 PCS and a backplate with the busbars, shunt (not necessary for now with BMS SBox in system), DC charging contactors (for when that's implemented), some fuse/breakers for PCS and future external loads (Mitsubishi heater and A/C pump)
I have a BMW X5 battery pack and running Damian's arduino script to control pre-charge and full pack voltage
During testing, I am reading aprrox. 0.4 VDC between the HV- busbar and chassis, but getting between 25 VDC and 30 VDC between the HV+ busbar and chassis (starts at 31/32 and drops slowly)!
Now I'm not expecting any voltage reading here, and proceeded to remove/isolate various components to see if there was an issue. I suspected dodgy busbar isolators, PCS doing something odd, or poor insulation at any crimps. Keep getting the voltage reading, so finally removed the HV+ battery lead. Still getting the voltage reading. What I'm getting from this, is that there is nothing leaking from the HV+ side to the chassis, and possibly this is expected...a difference in ground planes or something similar? With HV removed from the HVJB, I get open circuits between the busbars and chassis
I did some checks with a 12V bulb to see if there was any current and the bulb didn't light to either HV+ or HV- and chassis
Because its HV, and to show it respect, hoping someone with a better understanding could confirm if this looks to be an issue or expected before I start putting it back together?

I note that your Junctionbox and Battery box do not seem to be connected together to equalize the grounds.

Have you done any measurements on the HV battery, leads to casing?

When you did the light bulb test did you have your multimeter in current reading mode to check the level of current flowing?

Do you have a hi-pot tester that can do resistance measurements at 500V or 1kV? Then you can test each subsection of your junction box.

Edit also noticed you did not pull the plugs out of the PCS, you only pulled the positive fuse input wires. This would only leave the PCS to HV negative.

EDIT 2 The BMW HV contactor box has built in isolation monitoring, so there is a chance you are seeing this. I do not know if there is a way to turn it off.

Thanks for the reply Tom, you raise some good points so I have more checks to do

tom91 wrote: ↑Tue Nov 19, 2024 1:52 pm I note that your Junctionbox and Battery box do not seem to be connected together to equalize the grounds

Ok, you're right, I didn't think to connect both cases. I'll do that and recheck

tom91 wrote: ↑Tue Nov 19, 2024 1:52 pm Have you done any measurements on the HV battery, leads to casing?

Will do this too

tom91 wrote: ↑Tue Nov 19, 2024 1:52 pm When you did the light bulb test did you have your multimeter in current reading mode to check the level of current flowing?

I was relying on the bulb lighting so didn't pay attention to the meter

tom91 wrote: ↑Tue Nov 19, 2024 1:52 pm Do you have a hi-pot tester that can do resistance measurements at 500V or 1kV? Then you can test each subsection of your junction box.

Unfortunately not

tom91 wrote: ↑Tue Nov 19, 2024 1:52 pm Edit also noticed you did not pull the plugs out of the PCS, you only pulled the positive fuse input wires. This would only leave the PCS to HV negative

I did pull the HV connector from the PCS for one test so HV+ and HV- disconnected at PCS and was still seeing the same voltage HV+ to chassis, so was happy it didn't appear to be the PCS

tom91 wrote: ↑Tue Nov 19, 2024 1:52 pm EDIT 2 The BMW HV contactor box has built in isolation monitoring, so there is a chance you are seeing this. I do not know if there is a way to turn it off.

Interesting. If I hadn't the cases connected as per your first point, this presume this wouldn't affect the HV+/chassis voltage I'm seeing?

nkiernan wrote: ↑Tue Nov 19, 2024 2:21 pm Interesting. If I hadn't the cases connected as per your first point, this presume this wouldn't affect the HV+/chassis voltage I'm seeing?

In theory you are correct. However with things being a bit unknown best to test every possibility.

Redoing the earlier tests with a meter set to amps in series with the bulb for both HV+ and HV- to chassis, 0A for both and bulb didn't light, but voltage shows when measured (approx. 30 VDC and drops over short time). Before connecting the battery pack case and the HVJB enclosure for common ground, I checked both HV+ and HV- leads to battery pack case. This shows approx. 180 VDC and then dropped off over a short period
Next I connected the battery pack case and the HVJB enclosure and redid the HV+ to HVJB enclosure/chassis. Photos below show the same results as above, approx. 180 VDC (initially higher than when cases weren't connected) and then drops off over time. What I notice here is if I remove the multimeter probe and wait a short while and reapply the prove, the voltage has risen and then starts to drop off again. If I wait a couple of minutes and then reapply the probe, its back up to approx. 180 VDC before dropping off
With the comment about the BMW isolation monitoring above, I assumed this would be part of the SBox unit, so I removed the HV+ and HV- connections between the SBox and the battery pack to isolate the battery modules. However, same results from both the battery pack HV+ and HV- back to battery pack case! I then removed the BMS connector from the pack to the BMS master, incase that was a final HV path from the 12 modules in series and anything outside the pack...but same initial voltage that drops off.
This baffled me, surely there shouldn't be any voltage showing at this stage, unless its something to do with the multimeter. Had a good look around the internals of the battery pack to check for any module or cable damage or other connection, but nothing obvious showing up. Seems to be battery module level related at this stage, or is there any way this is normal!!?

At this stage it means there is a potential break down of isolation in one of your modules.

Due to the construction the "grounded" case will behave as a capacitor to the Cells/HV circuit building charge/voltage over time. A multimeter is a high impedance device thus will be discharging that charge causing a leakage to ground.

I would proceed to remove all module connections and measure each one by one to see which module has the issue.

tom91 wrote: ↑Tue Nov 19, 2024 6:41 pm I would proceed to remove all module connections and measure each one by one to see which module has the issue.

Appreciate the quick response and guidance Tom. I will work throught this and see what shows up. Thank you.

tom91 wrote: ↑Tue Nov 19, 2024 6:41 pm I would proceed to remove all module connections and measure each one by one to see which module has the issue.

Disconnected all the modules that were accessible on top and checked these individually again...similar but lower initial voltages that dropped off to approx. 0.002VDC in most cases. Will remove each module but started with this one to check and BMS was also disconnected. Is a standard multimeter of any use to truly check the modules!? Small voltage readings that drop off but this could be the meter and swapping between positive and negative. Resistance shows open.

Is there a sensible priced DIY option for a proper module tester I wonder? Will check if there's something like a Fluke 1507 insulation resistance tester in work maybe

The multimeter is fine, you saw it drop to 25-24V and stay there right? So no need for an HI-pot tester/Megger

The voltage is coming from somewhere, please not to look at what the structure under the modules is, if its closed by alu do not worry. I have seen situations when a module is out of the box/off the cooling plate the issue goes away, bolt it back in issue comes back.

Might be worth doing a megohm resistor battery - to battery box and calculate the voltage divider calcs to see what resistance you have here. You can do this with the + side too but it's measured in the difference to the + side. HV safety gear of course if it's high voltage.

Have you checked things like your shrink tubing for resistance readings?
You can occasionally run into static dissipative items that can be easily overlooked because they are thought of as 100% insulating.

tom91 wrote: ↑Wed Nov 20, 2024 6:22 pm The multimeter is fine, you saw it drop to 25-24V and stay there right? So no need for an HI-pot tester/Megger

Without the HVJB and battery enclosures connected, from HV+ lead to HVJB enclosure the voltage would start at 25 to 30 VDC then drop off over a short period to a volt or so. HV+ lead to battery enclosure started up around 180 VDC and dropped off the same way.

After connecting the enclosures, the HV+ to HVJB enclosure was now also at 180 VDC and dropped off. So the voltages always dropped to a volt or so. Each module is sitting approx 27V. It'll be Saturday before I get to work through the rest of the modules. But reading your comments, reminded me the pack is sitting as shown below with the two lower case halves not connected. I get the same result on both halves, so will connect these as a check before dismantling further

jrbe wrote: ↑Wed Nov 20, 2024 11:54 pm Have you checked things like your shrink tubing for resistance readings?
You can occasionally run into static dissipative items that can be easily overlooked because they are thought of as 100% insulating.

I did initially check any crimps and shrink wrap, but now I've disconnected the modules from the SBox so isolating just the modules in the lower battery enclosure cases to remove any external crimp or build issues and the symptoms remain. So does seem to be pack related at this stage

nkiernan wrote: ↑Thu Nov 21, 2024 7:49 am So the voltages always dropped to a volt or so

Wait what, you are saying all the measurements you took always decayed to around 1V or less? This is news to me your previous posts did not mention they would reach this, what sort of time does it take. Please be as specific as possible.

It sounds like this is not an issue as it first appeared. Please note things like temperature and moisture along side the multimeter used will greatly impact your readings.

tom91 wrote: ↑Thu Nov 21, 2024 10:05 am Wait what, you are saying all the measurements you took always decayed to around 1V or less? This is news to me your previous posts did not mention they would reach this, what sort of time does it take. Please be as specific as possible.

It sounds like this is not an issue as it first appeared. Please note things like temperature and moisture along side the multimeter used will greatly impact your readings.

Apologies, yes, all the measurements drop over the space of maybe 10 to 30 seconds from the initial peak to low voltages in the region of 0.002V to 5V depending on full pack or module only (I expect that 5V would keep dropping if I held the probe long enough but the rate of change slows up when it gets that low). Take the probe away and reaply after a short period to the same test point and the voltage has risen (wait long enough and it'll go back to peak) and then start dropping again as before.


Just found there's one of these in work I could use over the weekend for more accurate checks if needed

nkiernan wrote: ↑Thu Nov 21, 2024 10:51 am Just found there's one of these in work I could use over the weekend for more accurate checks if needed

Those are perfect to use for sanity checks. I always would recommend checking any HV sub assembly and verify it has the required resistance at 1000V.

If you have the time it is highly recommended to be able to close this topic and have definitive results.