openinverter forum

Jack Bauer wrote: ↑Wed Oct 20, 2021 12:01 pm It's been a bit since I messed about with the lexus stuff but sounds like you are good to go. Data exchange is happening with the inverter. 65v is more then enough to spin up the motors. What is invstat parameter? When you press throttle is potnom and torque registering a change?

Invstat is ON
Potnom and torque are staying at 0

My pot ranges from 420 to 2580 when fully pressed
Potmin 400, potmax 2800
Potmode single, dirnode switch..

There's your problem:) Have you set udcmin to a sane value? say 20v? also potmin needs to be greater than your off throttle reading. so if your off throttle is say 420 then set pomin to 430. Potmax needs to be less then your wide open throttle value. So if your wot value is 2580 then set potmax to 2570.

Jack Bauer wrote: ↑Wed Oct 20, 2021 12:16 pm There's your problem:) Have you set udcmin to a sane value? say 20v? also potmin needs to be greater than your off throttle reading. so if your off throttle is say 420 then set pomin to 430. Potmax needs to be less then your wide open throttle value. So if your wot value is 2580 then set potmax to 2570.

That makes sense. To be fair I had been playing about with them already. Now set 500 min, 2500 max, and pressed UPDATE

But it’s still the same result, potnom stays on zero :/

Udcmin is 45, tried 20. No errors but no go.

Edit: tried cycling / restarting it with the new setting, no change. Thanks for the help, will come back to it tomorrow.

Ok few things. You need to always press enter when changing a parameter. Is your opmode in RUN? If not no torque commands are sent.

Are you using a single or dual channel throttle? Dont forget to set potmode accordingly.

Jack Bauer wrote: ↑Wed Oct 20, 2021 6:02 pm Ok few things. You need to always press enter when changing a parameter.

Changes survive a power cycle so I assume I’m doing that correctly.

Is your opmode in RUN? If not no torque commands are sent
Are you using a single or dual channel throttle? Dont forget to set potmode accordingly.

Definitely getting opmode to RUN (like in the screenshot)

It's a dual pot throttle but I've set potmode to single to keep things simple to start with. It increases the "pot" channel as you press the pedal.

Brake input was off (not connected), also.

Yes but is pot 2 reading? I'm guessing it is and if not calibrated this will cause the problem. Calibrate pot2, select dual throttle mode and see if that makes a difference.

Jack Bauer wrote: ↑Thu Oct 21, 2021 7:23 am Yes but is pot 2 reading? I'm guessing it is and if not calibrated this will cause the problem. Calibrate pot2, select dual throttle mode and see if that makes a difference.

OK - the problem was that Throtmax was set to zero. I hadn’t changed that so it’s either the default value or it got scrambled.

The VCU will run with two pot inputs when set to single pot mode. But it looks like pot2 is referenced in the calculation, even in single mode, so yes, beware of that.

Other issues:
InvLSswitch output does not work. Must be a hardware problem I guess? We’ve ordered a new board anyway so I won’t spend time investigating that for now.

Tmpmmax is still coming in at 6400, which does prevent it from working. Where does that value come from, have I got a wiring fault?

In gs450h mode tmpm comes from the MG2 temp + and - pins. These should be connected to the mg2 temp sensor on the gearbox. Regards the inv ls switch the most likely culprit is a bad solder joint. Thanks to the chip shortage quite a few more parts require hand soldering. Would be easy to diagnose with a multimeter.

Alright, sounds easy to fix. I haven’t heard the oil pump run yet so that’s another thing to figure out.

hello,

I hope you can help me!

The first question :
what does the red flashing LED on the zombieverter board mean ?

I have a red one on the ESP and a red one on the board that are permanently lit and one on the board that is flashing.

2. i don't get any CAN data from the board
What settings do I need?

3. i have 3 relays
one for the + one for - and one for precharge
How should I connect them?


Thanks
Peter

Thank you this has helped me a lot !!!

Please link it on the wiki page !!!


Thanks
Peter

Hi strange issue here. Updated the software on my v1 board yesterday(compiled from master on github) I did update as normal and then it just stopped at 80%. No problem pulled power the did tweezers across c12. The board then accepted the update and finished. Only issue is now it doesnt show any parameters on web portal(it did before update) I have tried Johannes tip of clearing the files on the/list page. II have also tried another wifi board that works with the older vcu to no avail.Already tried compiling a new bin on other computer just to be sure. Would i be right in assuming the issue lies somewhere in the communication between stm and wifi module? Any body got any thoughts?
Can anyone share a .bin file with me to see is it i just made a mess when compiling?

Can the Zombieverter act as a CCS controller alone, or does it need to control an inverter?

since im planning to run the M3 motor/inverter

@paa your most likely issue is a bad solder joint on the MCP25625. the latest firmware initialises this device at startup and if not present or has a problem will give the problem you describe. Attached see binaries for an early release without this check in place. Future updates will see a timeout on this so the vcu does not halt.

@JockE28 yes the vcu can be used as a ccs controller in conjunction with a BMW i3 LIM and ISA shunt. I'd also like to think that when the OI model 3 drive unit firmware is released that it will work for that also.

Coming back to it today and the web interface is empty, like paaa described above. I have re-loaded it several times, apparently it accepts this OK (with or without the tweezer reset) but still no life after a reboot. I tried your .bin file too Damien. There should be a LED flashing if it’s alive?

The soldering on the MPU / IC10 looks fine, I can reflow it but not too optimistic right now. Any other ideas?

the led should be flashing when the software is running. The fact that its uploading means the bootloader is alive. The .bin above should run regardless of the MCP25625. Honestly, I can't generate a fault like this here.

Hi, I've got a couple of questions I hope someone can help me with;
Firstly I have real problems connecting with the Web interface, it takes 10 or15 minutes of retrying to get a connection, once connected I do appear to have 2 way communications - I can save a configuration at any rate; Even after I've connected once its no better. Any idea how to improve this?


Secondly should Led1 be flashing - I've noticed it in videos but can't seem to find anything about it.

And thirdly I've got a beta board with a GD chip will the stm32 bins work with it?

Cheers

Hi all, about to order an ISA shunt from EV create. Do I need the can terminated model? I think it's labeled can2.

Thanks!

Uppertown wrote: ↑Tue Oct 26, 2021 7:33 pm Hi, I've got a couple of questions I hope someone can help me with;
Firstly I have real problems connecting with the Web interface, it takes 10 or15 minutes of retrying to get a connection, once connected I do appear to have 2 way communications - I can save a configuration at any rate

I noticed it wasn't a great connection too, removing the back of the box and sitting the laptop right next to it helps a bit. It may be that it's using a specific wireless channel which is getting interference from something else?

Overall I'm finding it very frustrating and opaque to use the interface, not sure if that's due to my laptop or browser or something.

Ok. This is not where I wanted to be at all. It would seem that my stated goal of making an easy to use and configure open source and community supported vcu is an abject failure. If this is indeed the case then not only am I wasting my time but that of others and having the opposite of the desired effect. The hardware and software of this project are running in my 3 cars for the past 6 months. One of which has accumulated over 10k miles now, driven across Ireland and CCS fast charged along the way. I have a bench unit in the workshop running a Gen3 Leaf setup and another on my desk with a VAG and Toyota instrument cluster where I'm working on software to provide menu options to accommodate these vehicle types.

I drive around with this stuff every day. Hell I'm going to move house over the next few months in those cars driven by the very same hardware and software. So my money is where my mouth is but I have to accept and respect that if the result of that is something that is to quote et0 "opaque and frustrating" then so be it and I can't in good conscience continue. The last video I posted goes through that same interface in excruciating detail so if that's not enough to clear the waters and answer questions then nothing I can do ever will. Yesterday I was shooting part 3 and spent 90 minutes trying everything I could think of to "brick" the bench unit. Nope.

Part 3 is intended to go through the operation of the vcu with the Leaf setup on the bench and again explain as much as possible. I'll finish that off and post it up but at the same time I need to reassess my role in this game.

I have posted multiple times about wifi issues. I currently live in an area mostly populated by holiday homes. During the summer I have the devil's own time with wifi because every house is full of people scrolling facebook and watching netflix. At one point in August I did indeed "brick" my vcu in the E39 and was forced to remove it, get in the E46, drive 2 miles up the road whereupon the laptop connected first time and programmed flawlessly. Nowadays in the off season I cannot generate that problem at all. My advice to folks with this issue is to get a wifi analyser app for a phone or tablet and look at what networks are in your area and their relative strength. My wifis are set to channel 14 so if you have something on that channel in your area it will cause HUGE and WEIRD problems like are being described here. Don't ask how I know:)

Despite huge supply issues particularly with the STM32 I have not and will not change my prices or use suspect parts. But yeah, I have some serious thinking to do.

Anything new is always challenging. I wouldn't have even started my project if it weren't for the work you're doing and I look forward to seeing it through.

That being said...
I trust my soldering skills about as much as I trust the gov't so I paid someone to do it for me. As soon as I hooked up power and ground, the component at ic19 let out it's magic smoke. Before I have it replaced, can anyone see anything else that would cause problems?

All leds light up and blink, at least for the half second before the smoke started.

IC19 is an NCV7356 SW can transciever. As such it gets a direct 12v feed to pin 5. It looks from your pictures to be oriented the correct way so I'm going to guess its the wrong part.

Great, so that means at least one other component is in the wrong place.
So of the similar components:
IC2 - 24M01RP
IC11- XL1509
IC12 - NCV8461
IC14 - V7356
IC 27, 28, 29 - 3122C
IC23 - TDA2822G

Whatever is at IC19 is not legible. IC14 is supposed to be at IC19 and I'm hoping that's the only change?

Is this information available somewhere that I missed?

Full schematics and board layouts on the github as always
https://github.com/damienmaguire/Stm32- ... are/Zombie

Looks like IC14 is the NCV7356 that should have been in IC19