openinverter forum

Can I actually use the VCU to operate the pumps and fans based on temperature?

If you are talking about the zombie? Than Yes.

CoolingFan - OUTPUT: activates when FanTemp setpoint is reached.
CoolantPump - OUTPUT: activates during precharge, usually used for coolant pumps

see: https://openinverter.org/wiki/ZombieVerter_IO

yes I read through it, but as I understand it, just switch it on or off, but I have the Tesla pumps in there that need PWM, as well as fans with BMW control units that also work with PWM, it would be cool to run the whole thing via PWM, and not full power

Sure there are alot of things that would be cool. Just need someone to code those in.

Please note there are two possible PWM frequency sources, one for the "oil pump" and one for all other PWM outputs. This means if you want to use PWM you need to make sure they are all happy with the same frequencies.

Thats the "simple" part. Doing thermal management with PWM controlled pumps and fans means deciding how many stages you want/need then setting up parameters for all the possible variables. Then coding it then testing it, all while ensuring it does not stop other functions from working.

yes, of course there is a lot of coding work involved, it's all still in its infancy, the pumps from Tesla run at 2hz PWM, and the fan control unit from BMW, Peugeot and certainly other manufacturers runs at 100hz PWM, I have an oscilloscope here, can someone perhaps help me with how to configure it, i.e. in the settings in the web interface, the explanation in the wiki is a bit difficult to understand and I don't quite understand it.

perhaps some of the information in the link is also interesting for the wiki


https://www.evcreate.com/using-tesla-th ... tem-parts/

There is no functions in the Zombie to control pump or fan speed based on temperature.

I understand that, but I don't know how to set a fixed value, at least temporarily

PWMTim3 on a PWM pin. All others cannot have any PWM functions on them.

Then set the approiate pwm params as found on the overview https://openinverter.org/wiki/Zombiever ... pot_Values look for section - PWM Control

Prescaler has to do with clock divisions, not 100% sure I just kept it and changed period to get what I want.

OC is used to set duty cycle.

Im sure Damien covered it at some point in one of his videos.

Really easy to get the correct frequency. The clock is 72MHz so need total division of 36000000, e.g. 9000 x 4000.
set Tim3_Presc to 8999, and Tim3_Period to 3999. Duty cycle is controlled with Tim3_x_OC, where x is 1 to 3. 1000 gives 25% duty cycle, 2000 gives 50%, 3000 75% etc.

According to ev create the Tesla pump PWM is 5V. This is not possible with zombieverter as well as switching between 12v ground and floating for the valves .

I don't know why, but my pumps don't work with 5 volts, only with 12v pwm do they work

I think it would be better to develop a separate control unit that takes over the thermal management

They probably have something going on

https://www.evcreate.com/shop/cooling-s ... ontroller/

I use a PLC to control Pump, valve and heater as it makes it easier for me then coding for me

@medo, would be very interested to know more about your application, as I'm just coding the PWM support in HeadlessZombie which is a fork of Zombie with more flexible I/O.

There are 12 reconfigurable outputs which can do pull up or pull down PWM, (in addition to 2 dedicated processor driven PWM outputs)

But those 12 O/Ps can only do 80,100,200 and 2000Hz.

If I understand correctly, Tesla pump requires 2Hz (!) open drain drive on the PWM feed.

Can you confirm that?

If so can you try it at say 80Hz?

If not, I could add a 2Hz PWM in software on any of the O/Ps.

jetpax wrote: ↑Fri Jan 31, 2025 3:04 pm But those 12 O/Ps can only do 80,100,200 and 2000Hz.

Why no 1000hz? This is required to simulate the Control Pilot Signal among other things.

Also please keep your development to your own thread as this is quite off topic as its not anything to do with the Zombie Firmware.

at the moment I am using this
https://www.ebay.de/itm/364842234939

and, yes it is 2 Hz PWM, according to the technical data 5V, but with that the pump does not start at all, only at 12V 2Hz PWM the pumps start running, I have no idea why
by the way, I fail to even generate a pwm on the zombie

Chill @tom91, I know you don't mean to, but sometimes you do come over as a little rude on this forum.

Like you, although I didn't create it, I'm contributing to an open source project with code and hardware.

Am also trying to help improve Open Inverter at the same time.

In answer to your question, there are 2 dedicated processor driven PWM for that sort of thing in HeadlessZombie, although this would not be required if using Foccci, am I right?

medo wrote: ↑Fri Jan 31, 2025 3:43 pm at the moment I am using this
https://www.ebay.de/itm/364842234939

and, yes it is 2 Hz PWM, according to the technical data 5V, but with that the pump does not start at all, only at 12V 2Hz PWM the pumps start running, I have no idea why
by the way, I fail to even generate a pwm on the zombie

Thats great, can you try it at say 80Hz out of interest?

Could you share the tech data you have?

I have already tested some frequencies, only the 2Hz work, here in the wiki is all the information

https://openinverter.org/wiki/Tesla_Mod ... olant_Pump

jetpax wrote: ↑Fri Jan 31, 2025 3:50 pm although this would not be required if using Foccci, am I right?

No, there are chargers that require Control Pilot signal to function thus the Zombie needs to replicate it.[quote=jetpax post_id=79435

jetpax wrote: ↑Fri Jan 31, 2025 3:50 pm I'm contributing to an open source project with code and hardware.

Please continue to do this, this way more people are involved. Sometimes things come across very one way, with myself contributing and pushing the zombie coding forward and answering related questions on the forum.

Posting about your project in threads about ZombieVerter related questions will stirr up more confusion if there is not a unified solution between Zombie and your project.

tom91 wrote: ↑Fri Jan 31, 2025 4:06 pm No, there are chargers that require Control Pilot signal to function thus the Zombie needs to replicate it.[quote=jetpax post_id=79435

Hmm now you're confusing me! I thought that Foccci had a CP signal which can be spoofed and the charge current controlled by CAN to satisfy the limits of the AC outlet (EVSE). But that is certainly off-topic.

tom91 wrote: ↑Fri Jan 31, 2025 4:06 pm Posting about your project in threads about ZombieVerter related questions will stirr up more confusion if there is not a unified solution between Zombie and your project.

Absolutely agree, which is why HeadlessZombie is up to date with Zombie master branch, with changes #ifdefd, including the entire port of Zombie to STM32F405.

Its a WIP, but the eventual goal is that anyone will be able to build and debug the Headless code base for either hardware platform, just by changing the platformio.ini file.

My apologies if anything I have written in this thread has confused other forum members, that was not my intention at all.

jetpax wrote: ↑Fri Jan 31, 2025 5:08 pm Hmm now you're confusing me! I thought that Foccci had a CP signal which can be spoofed and the charge current controlled by CAN to satisfy the limits of the AC outlet (EVSE).

I just tried to explain this in the wiki: https://openinverter.org/wiki/Foccci#Q2 ... the_CP_pin?
Maybe for more detailled discussion we could jump to the Foccci/OBC integration chat: viewtopic.php?t=5127

after yesterday's update from Damian, the PWM now works, I was wondering why it didn't work, now I can also set the 2hz so that the pumps run and I don't need an external device anymore, a temperature-dependent control would of course be nice, but maybe something will come along.

Unfortunately, the ls power inverter switch on pin 32 no longer works

medo wrote: ↑Sun Feb 02, 2025 12:55 pm Unfortunately, the ls power inverter switch on pin 32 no longer works

Please provide some more information on this, like your parameters.

with 2.20A the VCU switches on the inverter after the start signal, after the update to 2.22A it no longer works, parameters were not changed, after a downgrade back to 2.20A, everything is fine again