Main contactor not closing

[AdrianB]

Zombieverter VCU
GS450h inverter and transmission

I have my setup working on the bench now but the VCU isn’t closing the main contactor.

The VCU powers up normally and when I apply the start signal the negative and precharge contactors close. UDC and INVudc come up and the VCU goes into RUN mode. However the main contactor doesn’t close.

I’ve carried out the following tests:

• I swapped the contactors around to confirm the contactor is good

• I’ve got continuity between PIN33 on the ZV and the main contactor coil (-)

• I’ve got 12V at the main contactor coil (+)

• I’ve got continuity between DRAIN on IC24 and PIN33

• I’ve got continuity between SOURCE on IC24 and PIN55 (ground)

• I’ve got continuity between GATE on IC24 and PC7 on SMT32

• When the ZV is in RUN mode I measure 3.4V on GATE of IC24

As all this seemed right I decided to manually earth the main contactor coil. The contactor closed and the motor then span very strongly (and my precharge resistor cooled down!)

Can anyone think of anything else I should check before deciding that IC24 is bad?

[nkiernan]

First thought would be settings in the web interface, especially udclim, udcmin, and udcsw.

Possibly not related from your description and you've probably already seen this post but sharing a link I had discussing a HV+ contactor case I have

viewtopic.php?p=50338#p50338

This however only relates to restarting as it always starts up ok the first time. Have a second GS450H transmission I want to rerun the test on.

Some other thoughts, what voltage is your HV for testing? I found on lower voltages with some variance between actual and reported voltages I had to lower udcsw to match.

[royhen99]

What current do each of your contactors require?
The NCV8402 device that drives the outputs have a peak current limit of about 3.5 - 4A. This varies a bit with temperature and device to device.
Can you test with a lower coil current relay to see if that works? If it works this could then be used to drive the main contactor.

[AdrianB]

Thanks royhen99 and nkiernan for your suggestions. I’ve spent the morning tinkering but haven’t been able to fix it yet.

First I checked my parameters:

• My HV test pack is around 86V

• udcmin = 72V

• udclim = 100V

• udcsw = 75V

The ZombieVerter appears happy with these settings. Negative contactor and precharge contactor close. udcsw is reached and both contactors stay closed but the main contactor doesn’t close.

Then I looked at the contactors.

• Mine are GV200-051 (second hand). I can’t find a datasheet for these but I’m now thinking they may not have economisers.

• I swapped the precharge one for a HV relay out of the Lexus. That made no difference.

• I also tested the main contactor circuit with a small automotive relay. But that relay didn’t close either.

Finally I double checked the circuit:

• ZV is delivering 3.5V to Gate of IC24 at the appropriate time

• I have continuity between DRAIN of IC24 and the contactor coil

Before condemning IC24 I’ll drop the battery voltage and do a test just with relays.

On the plus side I did make a short video of my first proper spin.

[VWJoe]

Just something to try, check what the charger mode is set to, as I have been having problems with the main contactor making when it is set to Volt_Ampera. This happens when the starter is pressed, the Neg contactor makes and pre-charge, but the main Pos does not stay in.

When charger mode is set to say Ext not a problem.

[nkiernan]

• My HV test pack is around 86V

• I also tested the main contactor circuit with a small automotive relay. But that relay didn’t close either.

Is Invudc showing close to your 86V in the web interface? Not sure what to suggest after that, the fact the small auto relay also fails to close might well flag the zombie pin is not doing what it needs to.

Out of curiosity, does the same thing happen with the HV Request test? (Zombie in standby and 12V to HV request pin with relevant charger section setting)

[uhi22]

Out of your measurements I understand, that the Controller correctly delivers high to the gate, but the FET does not close its drain-source-path. This means, the FET is broken. No need to search in configuration or software part.

[AdrianB]

Thanks uhi22, nkiernan and VWJoe - the problem is solved.

It was the FET. I couldn’t get another one because everyone is out of stock until next month. I swapped it for IC25 which drives gpout2 and is currently not used for anything. The contactors are now working as expected and the system is powering up as it should. It’s also given me some desoldering practice!

Thanks again for the solution. Now on to my next snag.