ZombieVerter VCU Support

[Jack Bauer]

As stated on the webshop :
"Due to the ongoing semiconductor shortage boards may be missing some components. Where possible these parts will be supplied in the kit but this is not always possible. Please see the relevant support threads on the Openinverter forum for further details as the situation changes from day to day."
Only a few posts back I updated this information.
viewtopic.php?p=45375#p45375
IC10 is the MCP25625 and is as of today still unavailable from suppliers and must be harvested from development boards and as such is not included in the kit versions of the VCU.
X2 is the 16MHz crystal for running the MCP25625 : https://lcsc.com/product-detail/Crystal ... 13738.html

Perhaps you could add that information to the wiki least it get lost in the woods with the caveat that it is a moving target.

[Jack Bauer]

@nkeirnan if you are still struggling after today please contact me offline and we'll either work out a video call or a site visit. I'm keen to get to the bottom of this one myself.

[nkiernan]

Well if you ask me you should tell us where you think you connect everything? Let's check pin 1 etc..might be stupid but I bet something small is causing this.

Ok, will see about doing this later on

[nkiernan]

These two files are identical.

Not sure which version of the code is on there, but if it is 1.01 Inverter=6 is undefined. This is an easy way to drive yourself crazy is to set a value that is valid in one version and then put on new code where it isn't and it will just keep doing undefined things. I'd set it to GS450h, save to flash and power cycle to be sure it gets initialized properly.

That's an interesting point re the inverter. The Zombie came with latest version and when I connected to wifi first, the GS300H VCU was set as Damian commented it was bench tested with his GS300H set up. I just saved the parameter file before changing any parameters. I then changed inverter to GS450H for my setup, and changed some other settings for throttle, and sure I saved to flash and restarted before downloaded the new parameters file. Will look at this again to make sure.

[nkiernan]

Nope. Opmode would be in "charge" instead of run. Does the oil pump run with the zom?

Tested both the GS450H VCU and Zombie this morning with the oil pump connected and oil pump runs as expected with both. With the Zombie, I can update the oil pump speed on the web interface and hit enter and the oil pump speed changes accordingly, and quieter than I was expecting too.

So the Zombie looks to be accepting its dealing with a GS450H setup alright

[Jack Bauer]

Ok thats good in one sense as it wont run the oil pump in other inverter modes.

[nkiernan]

Well if you ask me you should tell us where you think you connect everything? Let's check pin 1 etc..might be stupid but I bet something small is causing this.

Not easy to explain where all the checks are happening so photos below should cover it. So 8 off inverter comms wires come in at top right to one side of the connector blocks, and the GS450H VCU (small black plug to the right of the connector blocks) and Zombie header connectors are joined and connected to the opposite side of those connector blocks

This is how I'm reading the pin numbers (from wiki's and github pinouts)

And with the inverter plug disconnected so I'm not getting continuity across the termination resistor on each pair, below photos show where I get continuity for each of the 8 comms wires between the VCU header plugs. The GS450H is working ok, so unless I'm reading the pin numbers wrong, I think the Zombie should be ok! And just to note, all the other connected wires work as expected, 12V and ground, shunt CAN, contactor control, throttle inputs, ign on and start, so pretty happy I have the pin numbering ok

[nkiernan]

These two files are identical.

Rechecked this and I must have saved the params without saving to flash or power cycling the VCU. Changing the inverter option, saving to flash, power cycling the vcu, reconnecting and downloading and reading the parameter files does change the inverter number in the parameter files.

[EV_Builder]

Pin 16= REQ -....

[nkiernan]

Pin 16= REQ -....

Yea, do you see something wrong in the photos? I have 16 = req-, 17 = req+, etc

[EV_Builder]

No can't see it, I think next step is that the VCU program is adapted to spit out something on the web interface to confirm us: receiving/ succesful sending of data.

[EV_Builder]

Z004.JPG

If you see that coloured levered clip / join block.
Why are there more then one wire in?

[nkiernan]

If you see that coloured levered clip / join block.
Why are there more then one wire in?

I have two wires coming out from each, one goes to the zombie connector and the other goes to the GS450H VCU connector. This is so I can plug in either the zombie or the GS450H VCU to run the one driveline...for testing purposes

[EvSteeve]

Good afternoon all!

Having an issue with my ZombieVerter / Leaf Gen2 Setup..

I have the setup in the car - (complete gen2 motor, inverter & PDM)
I am using LG Chem batteries, 300v (4x 5 batteries in series strings)
and obviously the ZombieVerter!

I am currently trying to get the setup to charge.
When the EVSE is plugged in, (& ign on) IT will close the contactors and charge for a short period but will soon after cut out, and I'm struggling to work out why. (probably worth noting, the can is communicating and the motor will spin)

I have attached some pictures of my parameters to help anyone kind enough to take a look!

I have noticed, in the 'spot values' the 'SOC' IS at 99.9% which could well be the reason its stopping it from charging but I cannot see what parameters are used to calculate this.. (UDClim & UDCmin etc are all correct)

Any advice?

Regards and Thanks in advance!

[Jack Bauer]

SOC is calculated using data from "BattCap" at the isa shunt. The interaction of the charge parameters is explained in the instructional videos to the best of my recollection. However I suspect your battery is full. Charge is terminated when 1)Idc drops below IDCspnt AND 2)UDC exceeds Voltspnt. So check if those conditions are being met firstly.

[EvSteeve]

SOC is calculated using data from "BattCap" at the isa shunt. The interaction of the charge parameters is explained in the instructional videos to the best of my recollection. However I suspect your battery is full. Charge is terminated when 1)Idc drops below IDCspnt AND 2)UDC exceeds Voltspnt. So check if those conditions are being met firstly.

Hi,

Thanks for your response,

Battery is not fully charged - Currently pack is at 280v max charge for the pack is 328V.
Car will charge successfully if used in 'HV_ON' manual mode, but when using EXT_CAN it fails. 'UDC' doesn't exceed 'Voltspnt' and IDC doesn't drop below IDCTerm (I presume you meant IDCTerm rather than IDCspnt). Weirdly when the EVSE is plugged in the system will precharge twice before going into 'Charge' mode for 5-10 Seconds before opening the contactors again. I have validated CAN wiring & termination resistance and all seem to be ok!

do you have any links about how the ISA shunt use the 'BattCap' Data to calculate soc? and if the zombieverter sees that it is at 100% Soc will it use that to terminate charging or does it only use the parameters you have mentioned in your reply to terminate charging? (Excuse my ignorance)

Regards!

[Jack Bauer]

SOC is for informational purposes only. Please see the video links in the wiki page. Are you running the latest release firmware?

[EvSteeve]

SOC is for informational purposes only. Please see the video links in the wiki page. Are you running the latest release firmware?

Have checked over the videos, decided to reflash the VCU with another version,
It did go to 100% and 'update complete' on the interface, so I power cycled it and now have no wifi! (ARGHH!!)

The red 'pwr' led is on the wifi chip, not sure if there should be others? It is a 'ESP8266'...
doesn't seem to be any other way of reflashing the VCU... SO frustrating!!

[nkiernan]

The red 'pwr' led is on the wifi chip, not sure if there should be others? It is a 'ESP8266'...
doesn't seem to be any other way of reflashing the VCU... SO frustrating!!

Have you looked at the Zombie youtube videos linked in the Zombieverter wiki. From your comment not sure if you have the red activity LED flashing and the 3.3V power LED on the Zombie board, so from the wiki, below might be of use:

1:06:23 Resolve update fail/hang - activity led stops flashing, no data on web interface (power cycle)

[EvSteeve]

sorry should have included that info, still have power led and activity led flashing on the VCU board itself so not the same as in the wiki videos

[Jack Bauer]

Unless for a specific reason, the only firmware folks should be running is here : https://github.com/damienmaguire/Stm32-vcu/releases

[EvSteeve]

Unless for a specific reason, the only firmware folks should be running is here : https://github.com/damienmaguire/Stm32-vcu/releases

If this is reference to my previous post, we are only using firmware from github. Any advice as to why the wifi was lost?

Tia

[Jack Bauer]

Yes I've just came across a few different firmware versions out in the wild so want to make sure everyone is on the same page. Regards the wifi I suspect its a victim of the Olimex problems I described a few posts back.

[dadiowe]

So, further progress yesterday on the BMW Z4 2004 GS450H.

It appears that all the vehicle can messages for the BMW E39 selection from the Zombieverter to the BMW Can Bus connection that was connected to the engine management ECU of the BMW are the same for the Z4. Happy days.

I now have a fault light free instrument cluster and functioning power steering. Thanks so much for this great step forward community.

However, when I test drove the car the drive was very badly Jerky in forward and reverse unless only slight and slow throttle was used.
I did see a suggestion of adjusting the torgue settings but that option appears to be only in the older V.3 GS450 software.

Today was not such a success as I managed to short out the HV sense wire from the positive side of the battery pack despite the positive terminal being disconnected from the battery. Further investigation appears to show that the capacitors in the Lexus inverter are not discharging and even after a few hours are HOLDING HV VOLTAGE almost to the level of the battery pack i.e. 400 Volts.

Obviously not a great state of affairs!!!

I hope no one else has experienced this issue. I have the Voltage sense wire connected after the master safety fuse, the negative contactor is not energised or welded. If I take a voltage reading, with everything powered off, at the terminals of the inverter I have just short of 390Volts.
Any guidance regarding this would be really appreciated.

[dadiowe]

Further to my Explosive start to the week. Can anyone tell me the part number of the component just below the 4 pin connector in the IVTS shunt? It is a long shot that nothing else is damaged but worth a try before spending £300 for a new unit.