[WIP] Mercedes 310E (310D campervan) - GS450H/GS300H

[Jacobsmess]

you can frig it to run without the shunt I did during bench testing but I wouldn't recommend it

I was of the assumption that you don't need a shunt to bench test anyway, is that not the case? I've dropped UDCmin to 0 and am manually precharging, I've a contactor/precharge setup on its way but it uses Lem DHAB current sensing which I'll need to work out for use with the zombie later on. How do you go about "frig"ing it

[Jacobsmess]

In additional news.... I've still not got anything. Prior to running on my InvUDC is 0 but then drops to -2V once I turn the inverter on....

[andybpowell]

I found that selecting s-box as the shunt it didn't fluctuate into negative values

[tom91]

Does your InvStat change to 1? Just to confirm you have communications with the inverter. invUDC should get populated once the communications is working with the inverter, and the "HV voltage" side is connected to a correct voltage level.

[Bigpie]

Invstat is changing to on, but the invudc isn't getting populated

[tom91]

so in theory there is comms as it should go to 0 when comms is lost or faults out. Best to look at how you hooked up the HV test voltage then.

[crasbe]

This would be a good opportunity to post pictures of the wiring. Wouldn't be the first case where inadequate wiring caused issues with communication.

[Jacobsmess]

This would be a good opportunity to post pictures of the wiring. Wouldn't be the first case where inadequate wiring caused issues with communication.

Will do. I'll have to make some space and I'm certainly not ruling it out as an option!

[Jacobsmess]

I found that selecting s-box as the shunt it didn't fluctuate into negative values

So you were getting negative values with other options selected?

[andybpowell]

Of udc with no shunt not invudc

[nkiernan]

I have seen something similar with the GS450H...chased a lot of communication issue theories initially. I've used used 4x12V car batteries (48V) and a 60V PSU as test HV supplies with no issues. In early testing the 12V LV supply just wasn't properly powering up the system (seemed to be too low or dropping voltage at start up with the contactors being pulled in). On suggestions by others I used the PSU and the 12V battery in series so voltage was kept at 13.5-14V and fit to take the voltage dip. Inverter would turn on but no invudc and no spin...like inverter had come on but dropped out before this.

Later I rewired the 12V positive and negative to zombie with larger wire size (1.5 from 0.75 I think, and connected grounds direct to 12V battery (removing a number of junction points) and this got everything working reliably. Good ground wire runs and ground cases. Maybe something to double check

viewtopic.php?p=49496#p49496

[Jacobsmess]

Later I rewired the 12V positive and negative to zombie with larger wire size (1.5 from 0.75 I think, and connected grounds direct to 12V battery (removing a number of junction points) and this got everything working reliably. Good ground wire runs and ground cases. Maybe something to double check

viewtopic.php?p=49496#p49496

This could very well be it, I've refused some daisy chained spade connectors to link both positive and negative wires.... Looks like I might be best spending some time pulling it all apart again.

[Jacobsmess]

Quick video of the current setup. Finally got invUDC showing on the web interface but no spin still.
I think I need to revisit the signal wiring.
Re: setup, I also tried doing all the LV off one battery with the PSU adding current but that didn't help. I've got a piece of aluminium bar touching both the inverter casing, transmission casing and the shield signal screen touching the inverter case with another 4mm wire going from the inverter case to the battery negative but still no dice.
No obvious errors on the web interface from what I could see.

[Bigpie]

Check the sync serial connections are good.
You sure the main contactors are closed and not trying to pull current over the precharge resistor?
Got inverter whine?
Keep it simple and check and double check wiring. It's gotta be close now.

[Jacobsmess]

Check the sync serial connections are good.
You sure the main contactors are closed and not trying to pull current over the precharge resistor?
Got inverter whine?
Keep it simple and check and double check wiring. It's gotta be close now.

Main contactors have all been closed manually but will check.
Yes got inverter whine
Will double check all wiring when I next get time.

[nkiernan]

Is the actual gearbox gear selector in a neutral position (R-N-D) and not in park position so output is free to rotate? In spot values, I presume OpMode is showing as Run, but is Brake showing as Off and either Forward or Reverse showing as On when you're applying throttle to spin? Is LastErr or Status below it in spot values showing any message?

If you have the inverter whine and its showing in run mode and showing InvUDC correctly, that should be good to start spinning motors. Unless the amp supply of the dummy HV pack is not enough? I doubt that though. There were a few variations of throttle calibration required during the development of Zombie, curious if LastErr might show a throttle range error?

[Jacobsmess]

Is the actual gearbox gear selector in a neutral position (R-N-D) and not in park position so output is free to rotate? In spot values, I presume OpMode is showing as Run, but is Brake showing as Off and either Forward or Reverse showing as On when you're applying throttle to spin? Is LastErr or Status below it in spot values showing any message?

If you have the inverter whine and its showing in run mode and showing InvUDC correctly, that should be good to start spinning motors. Unless the amp supply of the dummy HV pack is not enough? I doubt that though. There were a few variations of throttle calibration required during the development of Zombie, curious if LastErr might show a throttle range error?

Output spins freely and gear selector in Drive.
Opmode shows run when ai press the momentary button
Direction is forward
Handbrake shows off
No lasterr or status message

The dummy HV is quite old so it could be the batteries are too weak but my battery tester tests them as ok and they're charged. I've also tried running the PSU alongside them, Damian's videos show them spinning with 6A at 60v so I should be good for that.

No throttle range error and potnom shows the correct value for when I hold the pot in a particular position. I've updated to the latest zombieverter firmware and set potmin and max values with 100 added to min and 100 subtracted from max.

[nkiernan]

And does the Torque spot value plot to show the matching torque request vs throttle position? Sounds like everything should be good to go, but maybe it is some slight wiring issue as others have suggested

[Jacobsmess]

And does the Torque spot value plot to show the matching torque request vs throttle position? Sounds like everything should be good to go, but maybe it is some slight wiring issue as others have suggested

I've gone through and checked wiring and continuity this evening but nothing but I also checked the torque Vs pot and there is no torque request with pot change.

[royhen99]

The fact that torque is 0 suggest potnom is also 0. Previoulsy you stated that you had potnom values which would also mean there were torque values. I do not have a GS450H but I have checked with GS450H inverter selected and if there is a potnom value >0 there is also a torque value.
To get potnom/torque values the only requirements are to be in run mode and forward or reverse direction selected.

The wiki suggest checking that throtmax is not set to zero and that that tmpm is less than tmpmmax.

[Jacobsmess]

Thanks, here are my parameters and spot values.
I've a value for throtmax (100%) and tempt is lower than tmpmmax although it is an odd value at -27 for some reason.

[Jacobsmess]

Ok made some changes on BigPies recommendations and I'm getting potnom, pot and torque all moving when I plot values but still no spin.

[andybpowell]

Udcmin is set to 50v but is reading 0v as you have no shunt

[Jacobsmess]

Udcmin is set to 50v but is reading 0v as you have no shunt

Yes I've dropped that down now as well on BigPies recommendations.

[Jacobsmess]

I guess the next step would be some trouble shooting testing. I can probably pick up an old oscilloscope relatively cheaply or I have a hook up for using a known good one. Would there be merit in testing the twisted pairs with an oscilloscope at the zombie end with zombie disconnected or something similar?

Any other recommendations for troubleshooting?