ZombieVerter VCU Support

Locked
Kelju
Posts: 128
Joined: Sat Aug 22, 2020 6:54 pm
Location: Finland
Has thanked: 8 times
Been thanked: 5 times

Re: ZombieVerter VCU Support

Post by Kelju »

Yesterday I took Damien's float-refactor with "git clone -b", compiled it and everything was working normally.
This version does not have all the fancy new parameterization so I would assume that the bug is to be found somewhere in that direction...

Edit: Here is the compiled .bin from the latest float-refactor branch is somebody wants to test it. I cannot test this one without adding my own custom code due to my setup...

[The extension bin has been deactivated and can no longer be displayed.]

nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: ZombieVerter VCU Support

Post by nkiernan »

Ok, so some more testing this evening and some interesting results...not the holy grail yet though :)

So uploaded Kelju's .bin above (thanks for sharing Kelju - you are using a Leaf setup with the vcu yes?). To make it obvious if the inverter power relay activated I disconnected it from the inverter and just connected a continuity meter across the terminals so I could hear if/when it was closed. Everything set up, into the car and turned the key to on, then to start. Negative and pre-charge contactors closed...and on came the continuity buzzer, the inverter relay was closed. Positive contactor came in a short time later and off goes the buzzer...inverter relay was off again. From the code it looks like it should come on at ignition on rather than at ignition start.

Shut it off and tried the test again but this time no buzzer even at ignition start (need to check if this is because udc hadn't lowered enough). I shut off power to the vcu altogether, let it sit a bit, powered all up again and retried the test. Buzzer again while ignition start and positive contactor not closed.

Shut all power down again and tried a third time but with no bench power supply so that udcsw could not be reached to close the positive contactor. Once again, at ignition start the buzzer sounds (inverter would be getting power if connected) and stays on as long as key is held at start. This is the case even after the negative and precharge contactors turn off after the 5 second wait for udcsw to be met.

For a fourth test, I uploaded the original bin file I started with, the one Damian posted earlier in the thread as I hadn't noticed this behavior on that. Sure enough, exact same results as above.

So proves that the stm32 can send the correct gate signal to IC6, and power could be sent to the inverter. So must be just software logic? I will step back though and recheck my wiring to be sure. At one of the power-ups (HV-, HV+ and pre-charge contactors closed), I measured the gate voltage on the other FET's out of curiosity (3.3v at gate for IC24, IC25, IC1 and IC3. 0.08v at gate for IC6 and 0.05v at gate for IC26 - sure I recorded this correct :? ).
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: ZombieVerter VCU Support

Post by nkiernan »

Apologies for the ramble below :)

Just doing some investigation in the zombie code with my limited ability to learn a bit more about the issue I'm having where inverter relay isn't closing when needed (assuming wiring and pin outs I used match code versions!)

The inverter relay is controlled by stm32 pin PA8/67 and is designated inv_out in the code. So in the current repo (have to figure out version control on git later), searching inv_out in the repo shows that it only appears in two files, digio_prj.h (shows up once just defining the pin) and stm32_vcu.cpp (six instances). So this is where the issue must originate!

Of the six instances, one looks to be clearing the pin at boot up in the main function. Two are in a charge related if statement concerning the Leaf so shouldn't be relevant to my GS450H setup. the fourth is in an if statement that needs inverter type to be OpenInverter, which mine isn't.

That leaves two instances, one relating to 'set' (turning on) and one relating to 'clear' (turning off):

Code: Select all

if (opmode==MOD_OFF && (Param::GetBool(Param::din_start) || E65Vehicle.getTerminal15() || chargeMode))	//on detection of ign on or charge mode enable we commence prechage and go to mode precharge
   {
      if(chargeMode==false)
      {
         //activate inv during precharge if not oi.
         if(targetInverter != InvModes::OpenI) DigIo::inv_out.Set();	//inverter power on but not if we are in charge mode!
      }
      DigIo::gp_out2.Set();//Negative contactors on
      DigIo::gp_out1.Set();//Coolant pump on
      DigIo::prec_out.Set();//commence precharge
      opmode = MOD_PRECHARGE;
      Param::SetInt(Param::opmode, opmode);
      oldTime=rtc_get_counter_val();
   }

Code: Select all

if (opmode == MOD_OFF)
   {
      DigIo::inv_out.Clear();		//inverter power off
      DigIo::dcsw_out.Clear();
      DigIo::gp_out2.Clear();		//Negative contactors off
      DigIo::gp_out1.Clear();		//Coolant pump off
//        DigIo::err_out.Clear();
      DigIo::prec_out.Clear();
      Param::SetInt(Param::dir, 0); 	// shift to park/neutral on shutdown
      Param::SetInt(Param::opmode, newMode);
      if(targetVehicle == vehicles::BMW_E65) E65Vehicle.DashOff();
The 'clear' function seems to be ok except inv_out is already clear at this point, clears automatically once inverter mode goes to Run, or when HV+ contactor closes, or when start signal is no longer present. To me it would look like the 'set' function should be ok...at the start the inverter is in Off mode and gets an ignition start signal, not in charge mode, and not OpenI inverter mode = turn on inverter relay!? If the FET shut itself down due to overcurrent etc, I presume there should still be voltage at the gate which there isn't, so the stm32 is actively turning off the FET, and this only happens with a 'clear' command?

So does the code sneak into one of the conditions to do with charging or the OpenI inverter or the Leaf settings to get turned off, or back to boot or maybe just thinks inverter is in Off mode and not Run as shown on web interface!?

The only instances I see that inv_out gets turned off or cleared are:

Code: Select all

if (opmode == MOD_CHARGE || opmode == MOD_RUN)  DigIo::inv_out.Set();//inverter and PDM power on if using pdm and in chg mode or in run mode
      if (opmode == MOD_OFF)  DigIo::inv_out.Clear();//inverter and pdm off in off mode. Duh!

Code: Select all

if (opmode == MOD_OFF)
   {
      DigIo::inv_out.Clear();		//inverter power off
      DigIo::dcsw_out.Clear();
      DigIo::gp_out2.Clear();		//Negative contactors off

Code: Select all

gpio_primary_remap(AFIO_MAPR_SWJ_CFG_JTAG_OFF_SW_ON,AFIO_MAPR_USART3_REMAP_PARTIAL_REMAP);//remap usart 3 to PC10 and PC11 for VCU HW
   gpio_primary_remap(AFIO_MAPR_SWJ_CFG_JTAG_OFF_SW_ON, AFIO_MAPR_CAN2_REMAP | AFIO_MAPR_TIM1_REMAP_FULL_REMAP);//32f107
   usart2_setup();//TOYOTA HYBRID INVERTER INTERFACE
   nvic_setup();
   parm_load();
   spi2_setup();
   spi3_setup();
   Param::Change(Param::PARAM_LAST);
   DigIo::inv_out.Clear();//inverter power off during bootup
   DigIo::mcp_sby.Clear();//enable can3
Any thoughts or suggestions welcome 8-)
Zechy
Posts: 44
Joined: Sat Apr 30, 2022 1:10 am
Location: Australia
Has thanked: 34 times
Been thanked: 25 times

Re: ZombieVerter VCU Support

Post by Zechy »

I recently bought a kit from Damien - still learning
The Tmpm is stuck on 6400 and doesn't change if I disconnect the temperature wires from the motors.
Any ideas?
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: ZombieVerter VCU Support

Post by Jack Bauer »

I'm slowly getting the new workshop up and running so will be able to troubleshoot these various bugs out of the system as well as continue development on new features. I appreciate folks bearing with me in what was a real nightmare of a time moving house and home particularly over the last few weeks. My feet are just beginning to touch the ground the past few days.
Attachments
2022-04-26 15.13.31.jpg
I'm going to need a hacksaw
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: ZombieVerter VCU Support

Post by Jack Bauer »

Just to remind everyone again that there will be some missing parts from VCU kits. Some components are just impossible to find at the moment but I am doing everything in my power to obtain genuine parts. I have not changed my prices despite me having to do a LOT more work myself to avoid nonsense like JLCPCB trying to charge 120usd for an stm32!
I'm going to need a hacksaw
Reid0102
Posts: 38
Joined: Mon Feb 14, 2022 4:01 am

Re: ZombieVerter VCU Support

Post by Reid0102 »

Jack Bauer wrote: Sat Apr 30, 2022 4:49 pm Just to remind everyone again that there will be some missing parts from VCU kits. Some components are just impossible to find at the moment but I am doing everything in my power to obtain genuine parts. I have not changed my prices despite me having to do a LOT more work myself to avoid nonsense like JLCPCB trying to charge 120usd for an stm32!
I can use my business account to purchase in bulk if that may help? I can probably get these completed with the ECU case and harness installed here in the states. This may take some load off of you and then to make sure that you still get money to continue doing this I can have the business pay you a royalty per unit sold. I am all for helping in anyway possible.

Just a thought?
Zechy
Posts: 44
Joined: Sat Apr 30, 2022 1:10 am
Location: Australia
Has thanked: 34 times
Been thanked: 25 times

Re: ZombieVerter VCU Support

Post by Zechy »

I can't get the sync serial working in my kit. No pulses on CLK, REQ and HTM.
I used a logic analyser to check the input on pin1 for IC18 (CLK) and it's receiving input:
IMG_1841.png
Is there a trick in getting the can transceivers to work?
Could the can transceivers in the kit be faulty?
User avatar
EV_Builder
Posts: 1199
Joined: Tue Apr 28, 2020 3:50 pm
Location: The Netherlands
Has thanked: 16 times
Been thanked: 33 times
Contact:

Re: ZombieVerter VCU Support

Post by EV_Builder »

Zechy wrote: Sun May 01, 2022 3:24 am I can't get the sync serial working in my kit. No pulses on CLK, REQ and HTM.
I used a logic analyser to check the input on pin1 for IC18 (CLK) and it's receiving input:
IMG_1841.png

Is there a trick in getting the can transceivers to work?
Could the can transceivers in the kit be faulty?
Check wireing first. Who is supplying the clock?
Is the inverter switched on?
Converting an Porsche Panamera
see http://www.wdrautomatisering.nl for bespoke BMS modules.
Zechy
Posts: 44
Joined: Sat Apr 30, 2022 1:10 am
Location: Australia
Has thanked: 34 times
Been thanked: 25 times

Re: ZombieVerter VCU Support

Post by Zechy »

EV_Builder wrote: Sun May 01, 2022 5:49 am
Zechy wrote: Sun May 01, 2022 3:24 am I can't get the sync serial working in my kit. No pulses on CLK, REQ and HTM.
I used a logic analyser to check the input on pin1 for IC18 (CLK) and it's receiving input:
IMG_1841.png

Is there a trick in getting the can transceivers to work?
Could the can transceivers in the kit be faulty?
Check wireing first. Who is supplying the clock?
Is the inverter switched on?
Thanks. I checked and the wiring seems fine.
I then measured at the white arrows and didn't get any CLK+ and CLK- signal:
Screen Shot 2022-05-01 at 7.21.05 pm.png
Then I used external can transceiver modules between the input pin 1 and the 56pin connector:
IMG_1844.jpg
The syncserial then gets to the inverter and I get data back on the MTH cables.
User avatar
EV_Builder
Posts: 1199
Joined: Tue Apr 28, 2020 3:50 pm
Location: The Netherlands
Has thanked: 16 times
Been thanked: 33 times
Contact:

Re: ZombieVerter VCU Support

Post by EV_Builder »

Pohh good job. Then it's time to check orientation?
Converting an Porsche Panamera
see http://www.wdrautomatisering.nl for bespoke BMS modules.
Zechy
Posts: 44
Joined: Sat Apr 30, 2022 1:10 am
Location: Australia
Has thanked: 34 times
Been thanked: 25 times

Re: ZombieVerter VCU Support

Post by Zechy »

EV_Builder wrote: Sun May 01, 2022 11:40 am Pohh good job. Then it's time to check orientation?
I appreciate your help :)

The orientation of the ICs?

Here is a photo:
IMG_1840.png
Domt177
Posts: 219
Joined: Sun Aug 30, 2020 6:05 pm
Location: West Midlands, UK
Been thanked: 11 times

Re: ZombieVerter VCU Support

Post by Domt177 »

Think the transceivers are bad, as im having the same issues.

Fed CAN to 2 of them and received nothing the other side

Fed the same CAN to 2 known working ones (from another project) and received the other side as expected.
Zechy
Posts: 44
Joined: Sat Apr 30, 2022 1:10 am
Location: Australia
Has thanked: 34 times
Been thanked: 25 times

Re: ZombieVerter VCU Support

Post by Zechy »

Domt177 wrote: Sun May 01, 2022 2:52 pm Think the transceivers are bad, as im having the same issues.

Fed CAN to 2 of them and received nothing the other side

Fed the same CAN to 2 known working ones (from another project) and received the other side as expected.
Thanks for confirming. Did yours also come with the kit?
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: ZombieVerter VCU Support

Post by Jack Bauer »

Yes I supplied the can transceivers as due to this nonsense chip shortage JLC and any reputable sellers such as Mouser not only don't have any but wont even say when they will and no doubt I've gotten bitten by fakes. Again. I caught one batch on testing. Anyone affected by this let me know and I'll sort replacements once I find another source.
I'm going to need a hacksaw
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: ZombieVerter VCU Support

Post by Jack Bauer »

Reid0102 wrote: Sat Apr 30, 2022 6:24 pm
Jack Bauer wrote: Sat Apr 30, 2022 4:49 pm Just to remind everyone again that there will be some missing parts from VCU kits. Some components are just impossible to find at the moment but I am doing everything in my power to obtain genuine parts. I have not changed my prices despite me having to do a LOT more work myself to avoid nonsense like JLCPCB trying to charge 120usd for an stm32!
I can use my business account to purchase in bulk if that may help? I can probably get these completed with the ECU case and harness installed here in the states. This may take some load off of you and then to make sure that you still get money to continue doing this I can have the business pay you a royalty per unit sold. I am all for helping in anyway possible.

Just a thought?
Will send you a PM to discuss this offline. Thanks.
I'm going to need a hacksaw
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: ZombieVerter VCU Support

Post by Jack Bauer »

In stock at LCSC for once : https://www.lcsc.com/product-detail/CAN ... 12084.html
Will advise once they arrive and I can test a few.
I'm going to need a hacksaw
Zechy
Posts: 44
Joined: Sat Apr 30, 2022 1:10 am
Location: Australia
Has thanked: 34 times
Been thanked: 25 times

Re: ZombieVerter VCU Support

Post by Zechy »

Jack Bauer wrote: Mon May 02, 2022 5:57 am Yes I supplied the can transceivers as due to this nonsense chip shortage JLC and any reputable sellers such as Mouser not only don't have any but wont even say when they will and no doubt I've gotten bitten by fakes. Again. I caught one batch on testing. Anyone affected by this let me know and I'll sort replacements once I find another source.
Thank you :)
Domt177
Posts: 219
Joined: Sun Aug 30, 2020 6:05 pm
Location: West Midlands, UK
Been thanked: 11 times

Re: ZombieVerter VCU Support

Post by Domt177 »

Zechy wrote: Mon May 02, 2022 4:41 am
Domt177 wrote: Sun May 01, 2022 2:52 pm Think the transceivers are bad, as im having the same issues.

Fed CAN to 2 of them and received nothing the other side

Fed the same CAN to 2 known working ones (from another project) and received the other side as expected.
Thanks for confirming. Did yours also come with the kit?
Yes
Jonr
Posts: 47
Joined: Fri Feb 11, 2022 1:28 pm
Location: Surrey, England
Has thanked: 1 time
Been thanked: 6 times

Re: ZombieVerter VCU Support

Post by Jonr »

Domt177 wrote: Mon May 02, 2022 6:51 am
Zechy wrote: Mon May 02, 2022 4:41 am
Domt177 wrote: Sun May 01, 2022 2:52 pm Think the transceivers are bad, as im having the same issues.

Fed CAN to 2 of them and received nothing the other side

Fed the same CAN to 2 known working ones (from another project) and received the other side as expected.
Thanks for confirming. Did yours also come with the kit?
Yes
I think I have the same issue with my kit. I see clock signals on pin1 of both both IC18 and IC13 when power applied - when ignition plus "Start" is activated, I see the output IC18 pin7 go high but no signal passes through. I'm not sure of the correct working behaviour to benchmark it.
The overall issue is the inverter never hums and the motor doesn't turn but the software setup looks OK.
Project : GS450h into BMW 3 Series Touring?
Jonr
Posts: 47
Joined: Fri Feb 11, 2022 1:28 pm
Location: Surrey, England
Has thanked: 1 time
Been thanked: 6 times

Re: ZombieVerter VCU Support

Post by Jonr »

Jack Bauer wrote: Mon May 02, 2022 6:16 am In stock at LCSC for once : https://www.lcsc.com/product-detail/CAN ... 12084.html
Will advise once they arrive and I can test a few.
I've just tried to buy 10 of these from this vendor using PayPal - Issue for me was that the purchase value was about USD30, which is fine, but the PayPal screen showed the transaction value as USD 0.01 so I aborted. I've not seen that behaviour before!
Project : GS450h into BMW 3 Series Touring?
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: ZombieVerter VCU Support

Post by nkiernan »

Jonr wrote: Tue May 03, 2022 7:50 pm I think I have the same issue with my kit. I see clock signals on pin1 of both both IC18 and IC13 when power applied - when ignition plus "Start" is activated, I see the output IC18 pin7 go high but no signal passes through. I'm not sure of the correct working behaviour to benchmark it.
The overall issue is the inverter never hums and the motor doesn't turn but the software setup looks OK.
You're working with a GS450H yes? You will see a few posts back in this thread I can't get the inverter turned on or motor to turn but it seems to relate to IC6 (inverter power on relay control IC). Contactor control works as expected at ign on and start, web interface shows Opmode goes to Run, but Invstat stays Off. Maybe some checks you can do on that also. Maybe issues are related
Reid0102
Posts: 38
Joined: Mon Feb 14, 2022 4:01 am

Re: ZombieVerter VCU Support

Post by Reid0102 »

I want to see if anyone had any-thoughts on my gen3 set up, maybe i am missing something, but no can communication from the inverter and cant get it to power on.

here is full detail of what I have done so far.
viewtopic.php?p=39897#p39897
Zechy
Posts: 44
Joined: Sat Apr 30, 2022 1:10 am
Location: Australia
Has thanked: 34 times
Been thanked: 25 times

Re: ZombieVerter VCU Support

Post by Zechy »

My tmpm stays on 6400 no matter what I try.

I plugged a 10k Ohm pot in on MG2temp+ and - I can then measure a voltage of 0.2-3.8 on pin A6
When I plug the 10k Ohm pot into MG1temp+ and - I can only see 0.02-0.06 volt on pin A5

The value of tmpm never changes when I change the pot value.

I measured the voltages here:
MG1 and MG2 Temps.jpg
Jonr
Posts: 47
Joined: Fri Feb 11, 2022 1:28 pm
Location: Surrey, England
Has thanked: 1 time
Been thanked: 6 times

Re: ZombieVerter VCU Support

Post by Jonr »

nkiernan wrote: Tue May 03, 2022 8:50 pm
Jonr wrote: Tue May 03, 2022 7:50 pm I think I have the same issue with my kit. I see clock signals on pin1 of both both IC18 and IC13 when power applied - when ignition plus "Start" is activated, I see the output IC18 pin7 go high but no signal passes through. I'm not sure of the correct working behaviour to benchmark it.
The overall issue is the inverter never hums and the motor doesn't turn but the software setup looks OK.
You're working with a GS450H yes? You will see a few posts back in this thread I can't get the inverter turned on or motor to turn but it seems to relate to IC6 (inverter power on relay control IC). Contactor control works as expected at ign on and start, web interface shows Opmode goes to Run, but Invstat stays Off. Maybe some checks you can do on that also. Maybe issues are related
Yes I'm using GS450h motor and invertor. I've got a bench set up so am currently using relays instead of contactors. I don't seem to have a problem with IC6 - when the start button is pressed, my relay controlling power to the invertor closes and power 12v is fed to B and B+ at the invertor. This power is switched off when the ignition is turned off. Forgive me for saying, this is a low side switch where, when activated by the start button, IC6 grounds one side of the relay coil where 12v is on the other side of the coil via the ignition switch.
Project : GS450h into BMW 3 Series Touring?
Locked