[DRIVING] BMW Z3 [FINISHED]

[Woodfie]

My Zombieverter arrived a while ago and now that the connectors and case have arrived I've got it all built up and tested and ready to go

IMG20231001154601.jpg

Hi, on a search for which way up or down for the wifi board... and thank you for having this helpful picture,

I guess somewhere there is instruction, vid, I found one on the smaller red wifi board, but not the blue one you have.... Where abouts did you find the assembly info please... have received the Zombie board, and keen to get to next step... Thanks

[Woodfie]

[Woodfie]

viewtopic.php?p=69998#p69998

Have found the relevant assembly information,

Looking for ZombieVerter as one word helped,

[andybpowell]

Sorry I was a bit busy yesterday and didn't see your post

[Woodfie]

How is your project going ? Hope all is well.

[andybpowell]

How is your project going ? Hope all is well.

Not much further along to be honest the juddering is still an issue I really need to wait until more vehicles are using the l210 to see if the problem is specific to my installation or a common issue have been doing some work on the chademo charging but been busy doing work at home prepping for our house move next year

[Woodfie]

Nearing doing a run with wheels of the ground with new Zombie board, into Prius gen3 inverter and L210, to see if have overcome the current limit that had struck of around 40A for mg2 and 42 A for mg2.
Will juggle the brake and accelerator in the test wagon to initially check.

At present am unable to read UDC voltage coming from the inverter to the Huenber display , when the wheels are able to turn, so it can still run I.e. the inverter has the voltage, but no report to the Zombie and display screen.

This morning will pick up another Prius inverter which has been de boosted, to see if that sheds any light on the problem.

Re juddering.....
With the adapted Due board , under the same conditions, was striking a stammer, at around 4000rpm, where a current increase maybe 20% occurred.
I know this is not the gs300h inverter yet, am still hunting a communication plug to the gs300h inverter, from local parts yards, with no luck as yet.. so am stuck on the Prius inverter at this stage..

[andybpowell]

Around 4000-5000 rpm is where the juddering is at it's worst for me so that's interesting good luck I look forward to seeing your progress

[Woodfie]

Some more unloaded in vehicle test runs, have revealed that at 3400rpm , @ 290v where the bus current jumps up, 3.5A to 7A, and a more audible whine occurs.
More things to check out, on
Zombie >Prius inverter>L210 set up,

Have obtained plug for the 300h inverter wip

[Woodfie]

Trying to investigate further, when the current jump occurs on the L210, driven by the Prius inverter and Zombieverter board.
Observing motor AC volts with a moving coil AVO meter, (the dvm I use, can not read and reports erroneously).
At 3400rpm is when a sudden jump in Voltage and current occurs..
video recording at 290v dc batt.v

A check at 224V battery , lowered the "event" surge to around 3000rpm, and was much softer.

A check at 370v battery, with some temporary cells added, hardly raised the rpm occurance point, maybe to 3500rpm, where it would surge higher.

I had expected a linear relationship between battery V and the rpm of this surge event.. This does not appear so.

Am interested to do the same check with the 300h inverter, when installed next... hopefully by next week.

It seems this us around the speed you have a judder occurs, not sure if this is connected.

[Woodfie]

[andybpowell]

That's really interesting going to go see if I can find some empty roads to see if I can replicate that under load

[Zechy]

Hi Andy. I have some data to support your findings.
I'm using a gs300h inverter with a gs450h transmission. Zombieverter using GS300h settings. Voltage around 330V
When holding the speed constant and mg2 speed is around 3500rpm, after a few seconds I get juddering.

Steady speed judder

Zoom

I'm currently using a non-linear throttle ramp as suggested by EV8. The behaviour is exactly the same

Non-linear throttle

[Zechy]

I found a workaround for the juddering that I was experiencing. I did a 30km test run at between 80 and 110km/h.

I logged the serial data going to the inverter (HTM) and noticed that the inverter is re-initialised when the juddering occurs. I assume the checksum ==0 because of noise on the line which then resends the 6 lines to initialise the inverter.
I commented out this line and now I have no more juddering

GS300h code

I compared the code with the GS450h code. The GS450h doesn't have that line of code.

GS450h code

[Woodfie]

Discussion with a local Motor Guru, and observing the back Emf of mg1 and mg2,
It seems that this is likely to be where field weakening occurs..
What I am seeing looks like this event mayve at a rpm point, say 3500, and the Lexus use of the boost converter will aid to reduce the losses due to field weakening.
Still learning, so just ideas here..

[Woodfie]

Have done some more investigation, and its seeming that this 3500 rpm judder event point, on the Zombie/Prius3/L210,
battery at 280v,
the back emf is around 76vrms for mg1 (measured)
And 96vrms (calc from gear ratios)

Neither of these are close to the supply voltage, where FW would be expected..... which seems strange..

[andybpowell]

Thanks to Zechy I have tried out the work around for the judder experianced in 300h setups and it seems to work and has completley transformed the way the car drives, it's maybe not the ideal solution, maybe re-routing cables and only re-initialising after a number of consecutive errors(as suggested by celeron55) may be the answer but i will leave that up to the people with a bettter understanding than me

[Ev8]

Great work in finding the culprit guys! Interestingly I can get a harsh juddering if one motors inverter restarts due to over current and restarts whilst moving I put this down to a sync error being introduced somehow by restarting in motion, (totally different setup though so unrelated)

[Woodfie]

Have done some more investigation, and its seeming that this 3500 rpm judder event point, on the Zombie/Prius3/L210,
battery at 280v,
the back emf is around 76vrms for mg1 (measured)

Regarding MG2
The Back emf at the surge event point
@290v battery Mg2 at 3400rpm is 155v rms
@ 224v battery Mg2 at 3300rpm is 124vrms (Softer)
@ 370v battery Mg2 at 3500rpm is 158v rms

It seems that lowering the voltage dragged the surge point down, and not as harsh a surge.
Raising the battery voltage by 80 v did not appear to raise the surge point , I.e 290v to 370v battery did not appear to change things much.
Could it be that 3500rpm is the max point this happens
More to learn here, as in what happens in a Lexus at say 650v..

[nkiernan]

had a quick look from what i can see it looks fine no idea about balancing though

Screenshot 2024-03-14 104418.png

Hi Andy.
In one of your previous posts you had a screen shot of the Simp serial output and it correctly showed the 96 cell voltages. Mine is trying to show 16 cells per module for the 12 modules, so getting a series cell fault. I've tried some different settings for cells, strings, csc variant. Wondering if you have a chance if you could snap your Simp settings, I might see what I'm missing! I have the X5 edit for the 12 modules already.

[andybpowell]

Hi Andy.
In one of your previous posts you had a screen shot of the Simp serial output and it correctly showed the 96 cell voltages. Mine is trying to show 16 cells per module for the 12 modules, so getting a series cell fault. I've tried some different settings for cells, strings, csc variant. Wondering if you have a chance if you could snap your Simp settings, I might see what I'm missing! I have the X5 edit for the 12 modules already.


SimpBMS003.PNG


SimpBMS004.PNG

it's been a while since I looked at it but vaguely remember a similar issue have you tried the i3 csc

[andybpowell]

Not had much chance to work on the car recently but I have managed some time in the evenings so I've been learning freecad designing a connector for the 300h inverter HV dc input and this is what I came up with

[Woodfie]

Nice cover, well done.

[andybpowell]

I'm sure someone can do better but here is my 300h inverter HV dc connector if anyone has a use for it

300h hv conn.stl

(3.04 MiB) Downloaded 154 times

300hhvcover.stl

(132.99 KiB) Downloaded 144 times

300hhvclamp.stl

(37 KiB) Downloaded 137 times

[andybpowell]

I also have a blanking plate for the standard HV input but it's currently printing so havent tried it yet will share if its correct