How to not kill a tesla drive unit

[muehlpower]

Does it make sense to connect a capacitor in parallel to each of the main contactor coils? With a resistor for charging and two diodes. According to my estimate, a 10000uF capacitor should keep the contactor closed for 1s after switching off.

Kondensator.png (4.4 KiB) Viewed 10112 times

[mane2]

What caused the contractors to open? Did you replace the entire inverter or just the burnt phase?

I had max charging rate on Orion BMS set to 200A (I think default value) and tested strong regen. As regen amps exceeded max set to Orion, it lifted contactors.

To fix this issue I of course upped the max amps but also added a relay so that Orion can only control main contractors when car is not driving.

I changed inverter from another motor, but I couldn’t make it water tight. So we changed the other motor completely, which leaked from motor side and now on the inverter side too So I bought a third motor, from newer car and it has been working now almost a year. No leaks or problems.

[mtrehy]

I had max charging rate on Orion BMS set to 200A (I think default value) and tested strong regen. As regen amps exceeded max set to Orion, it lifted contactors.

To fix this issue I of course upped the max amps but also added a relay so that Orion can only control main contractors when car is not driving.

I changed inverter from another motor, but I couldn’t make it water tight. So we changed the other motor completely, which leaked from motor side and now on the inverter side too So I bought a third motor, from newer car and it has been working now almost a year. No leaks or problems.

Your Orion BMS controls the contactors? I have Orion BMS but don't use that for contactors - I wasn't actually aware it could

[mane2]

Your Orion BMS controls the contactors? I have Orion BMS but don't use that for contactors - I wasn't actually aware it could

My LDU controls contactors but Orion has priority over LDU to open them if something bad is about to happen. After modifications it can only open them in charging or idle mode.

[mtrehy]

My LDU controls contactors but Orion has priority over LDU to open them if something bad is about to happen. After modifications it can only open them in charging or idle mode.

What Orion are you using? I'm interested in how you have configured it to control the contactors

[mane2]

What Orion are you using? I'm interested in how you have configured it to control the contactors

You can use those customizable outputs that Orion has to control relays that control contactors. I have Orion BMS 2.

[phillipschip]

Does it make sense to connect a capacitor in parallel to each of the main contactor coils? With a resistor for charging and two diodes. According to my estimate, a 10000uF capacitor should keep the contactor closed for 1s after switching off.

Kondensator.png

Has anyone tried this? This seems like a good solution. I am wanting to add some type of inertia switch. I have watched the video but having the contactors stay shut would not protect the vehicle or responders from high voltage in case of accident.

[jrbe]

Has anyone tried this? This seems like a good solution. I am wanting to add some type of inertia switch. I have watched the video but having the contactors stay shut would not protect the vehicle or responders from high voltage in case of accident.

I added a capacitor footprint to the gen3 leaf board kind of like this but in the mosfet gate drive circuit. I don't believe it has been tested.

If you do decide to try this you will likely need a capacitor with a rated voltage 8? times higher than the coil voltage to handle the inductive spike at turn off.
Also, the capacitor has the potential to delay the turn on depending on available coil current / wiring. Just something to watch for.

[phillipschip]

I added a capacitor footprint to the gen3 leaf board kind of like this but in the mosfet gate drive circuit. I don't believe it has been tested.

If you do decide to try this you will likely need a capacitor with a rated voltage 8? times higher than the coil voltage to handle the inductive spike at turn off.
Also, the capacitor has the potential to delay the turn on depending on available coil current / wiring. Just something to watch for.

Ok great. Thanks for the advice! So something like a 100v 10000uF capacitor on each contactor could potentially fix this issue? No harm in having them in there even if not?

[jrbe]

I added the capacitor footprint in the leaf board because some people were reporting their contactors chattering. It wouldn't fix the real issue but could mask the issue.

What are you trying to accomplish with the extended contactor on time? Is there something in the system that could be setup / adjusted so you don't have to make up for something?

You could add a contactor power loop that a responder can pull to shut down the battery.

Some inertia switches are junk, be careful with what you use.

[muehlpower]

I'll do it like this, close to the Contactors. It maintains the high voltage for a second even when the ignition is switched off while driving. The capacitors of the power supply on the OI control board should be empty more quickly. The diodes and the resistor limit the inrush current and prevent the current from flowing back to terminal 15. Unlike digital or timer-based solutions, it switches off safely after one second.

[phillipschip]

I'll do it like this, close to the Contactors. It maintains the high voltage for a second even when the ignition is switched off while driving. The capacitors of the power supply on the OI control board should be empty more quickly. The diodes and the resistor limit the inrush current and prevent the current from flowing back to terminal 15. Unlike digital or timer-based solutions, it switches off safely after one second.
Contactor delay.png

This seems like a great solution (I’m not an electrical engineer
What would the specs b for the diodes and resistor? Thanks!

[phillipschip]

Would something like this do the trick? I have these parts already except the capacitor. I would think the trip mode option to close the pre charge would be a perfect protection if given the option to then open after 1 or 2 seconds.
I was thinking of using the Ford fuel pump cutoff switch pictured. I have seen it recommended in forums.

[johu]

Guys, what are your issues with just keeping the contactors closed? You can always open them by turning the ignition off if needed.

[uhi22]

I understood the capacitor approach addresses the use case "ignition off while driving kills the inverter".

[phillipschip]

I understood the capacitor approach addresses the use case "ignition off while driving kills the inverter".

Yes, and the option to possibly use a crash sensor.

[muehlpower]

I understood the capacitor approach addresses the use case "ignition off while driving kills the inverter".

Exactly. In my imagination, first the direction signal goes to neutral (RND), then the capacitors of the power supply on the OI board are empty, and then the HV battery is disconnected. And everything is so safe that the TÜV accepts it as a real emergency stop, not just the signal for forward or reverse is switched off. Perhaps the TÜV or someone at the inspection for a racing event also wants a demonstration that the drive should survive.
How it is then switched off, ignition, emergency stop (also several) or crash sensor is then irrelevant.
Of course, this delay must then also be applied to the second contactor, which is not controlled by the drive unit.

[jrbe]

Would something like this do the trick?

That old style Ford inertia switch is one of the problematic ones I mentioned. It used to be funny to kick the front bumper of one of these cars and this switch would open. Car wouldn't start and the owner wouldn't know why. There might be a newer revision that's less sensitive though. I believe hitting a small to medium size animal would also open the switch possibly rendering your drive exploded. I believe these were added because of the lack of a fuel pump safety circuit in these models.

I'm not seeing what you guys are with this extra one second of contactor on time. If you're going highway speeds and the inverter is shut off, I only see this as giving you 1 second to turn the ignition back on before the contactors open / inverter fails.
It might buy you a little in the speed above which the inverter fails if contactors are opened - allowing generated regen power to flow into the battery. But this would likely go full regen during the inverter off, contactors still closed time, likely causing loss of control.

All of the emergency stop tests I've seen are stationary. But I think this should still be thought through.

OI has an e-stop loop in at least some inverters. Doesn't opening this set the inverter into like a neutral mode, a no regen / no acceleration mode to allow coasting safely to a stop?

It seems like what you guys are after is more of a vehicle computer that's watching for vehicle speed or crash info to cut the contactors when safe or in an emergency but I might be mistaken.

[barracuda816]

Not sure I get the idea either. If you shut off the 12v supply to the inverter (i.e Oi board or Zombie) then the motor cannot keep running under power so surely it makes more sense for an E-stop to cut the 12v power to the board and keep the contactors on latched?

Or am I missing your point?

[phillipschip]

Not sure I get the idea either. If you shut off the 12v supply to the inverter (i.e Oi board or Zombie) then the motor cannot keep running under power so surely it makes more sense for an E-stop to cut the 12v power to the board and keep the contactors on latched?

Or am I missing your point?

Yes. However, that still leaves HV live throughout the vehicle. I am trying to imagine a situation where, in the event of a crash, HV is shut off protecting both me and first responders.

[phillipschip]

That old style Ford inertia switch is one of the problematic ones I mentioned. It used to be funny to kick the front bumper of one of these cars and this switch would open. Car wouldn't start and the owner wouldn't know why. There might be a newer revision that's less sensitive though. I believe hitting a small to medium size animal would also open the switch possibly rendering your drive exploded. I believe these were added because of the lack of a fuel pump safety circuit in these models.

I'm not seeing what you guys are with this extra one second of contactor on time. If you're going highway speeds and the inverter is shut off, I only see this as giving you 1 second to turn the ignition back on before the contactors open / inverter fails.
It might buy you a little in the speed above which the inverter fails if contactors are opened - allowing generated regen power to flow into the battery. But this would likely go full regen during the inverter off, contactors still closed time, likely causing loss of control.

All of the emergency stop tests I've seen are stationary. But I think this should still be thought through.

OI has an e-stop loop in at least some inverters. Doesn't opening this set the inverter into like a neutral mode, a no regen / no acceleration mode to allow coasting safely to a stop?

It seems like what you guys are after is more of a vehicle computer that's watching for vehicle speed or crash info to cut the contactors when safe or in an emergency but I might be mistaken.

Any suggestions for a good inertia switch? If a really good one was installed that would only trip for serious accidents then the inverter wouldn’t matter :0

[jrbe]

I don't think you'll find published cut out decelerations for most inertia switches. I'd stay away from early ones from the 90's to the 00's.
This one does though,
https://www.pegasusautoracing.com/produ ... oh38GdCptu
Might be a mini Cooper one repurposed as a universal kit. Not too bad of a price with the connector and terminals with it. I haven't ever tested these.

[phillipschip]

I'm not seeing what you guys are with this extra one second of contactor on time. If you're going highway speeds and the inverter is shut off, I only see this as giving you 1 second to turn the ignition back on before the contactors open / inverter fails.
It might buy you a little in the speed above which the inverter fails if contactors are opened - allowing generated regen power to flow into the battery. But this would likely go full regen during the inverter off, contactors still closed time, likely causing loss of control.

If regen was turned off(I wouldn't have a problem not using it), with the capacitor at the contactor+diode+resistor setup, wouldn't the motor just shut down safely in the event of an unexpected 12v power cut in any situation. If not, could you explain to me a little further? I am super new to all of this, haven't started wiring, and want to be as safe as possible with my build. Thanks!

[jrbe]

If regen was turned off...

The inverter controls the regen level. Turn off the inverter and the motor / generator goes full regen. It doesn't matter if you have regen set to 0. No pwm from the inverter, no regen control.

If / when the regen voltage that has no where to go (open contactor) builds above the power stage's rated voltage it blows.

If the contactor is opened on light regen and low speed it can still blow the power stage from the field collapse energy / voltage from inverter pwm, voltage with no where to go.

If the contactor is closed power can feed back into the battery and keep voltage at a safe level but it will go full regen deceleration if the inverter is switched off.

One of the things I've mentioned previously as a possible solution is to turn on the vehicle's resistance heater if the contactors need to be opened for some safety reason. It potentially could help by consuming at least some of the power from regen that can't go into the battery. If it can keep the voltage low enough it could save the power stage.
But this is planning on saving something in an otherwise likely catastrophic event and the juice might not be worth the squeeze.

[muehlpower]

The headline reads “not kill a Tesla drive unit”. There have been cases here of destroyed LDUs and SDUs caused by a missing HV connection. In some cases it was a blown fuse, contactors opened by the BMS, or loose contacts on the circuit board or in the wiring harness. Often incorrect tripmode settings in conjunction with overcurrend. Setting 2 (precharge on) does not seem to be sufficient for deadsleads motor. Tripmode is the easiest to change, but does not help with the other causes. Since Tesla motors are three-phase asynchronous motors, a rolling vehicle should not be a problem after correct shutdown, as no current is produced. With leaf and Model3 it will be different.