Tesla Large Drive Unit Support Thread

[Rx7FD]

Yes, the short is in both directions. Do you think it could be repairable? I will disassemble the motor this weekend to check out the inverter.

Not sure what could've caused the failure, only drove for 20 minutes and didn't push it very hard. It didn't overheat or anything. Thank you Jack for all your help to this community, sad to see that you are discontinuing the Tesla boards.

[tom91]

You can find what is broken by removing the triangle busbar and measuring all the body diodes in the IGBTs. if one does not show a diode voltage of +/-0.36V or rising its broken. Remember to measure from motor phase to positive and negative, then swap the meter leads around and repeat.

[Rx7FD]

Any tips or a guide for removing the inverter from the drive unit? I have removed the 10 screws that are deep inside the triangular structure, but something else is holding it on (the ac power connections?).

Also, Tom what do you mean by a diode voltage? Is there a voltage drop across the diodes? Does it have to be powered by 12v in order to get this drop? And which solder joints on the circuit board are the diodes connected to? Thank you so much for the help!

[tom91]

Any tips or a guide for removing the inverter from the drive unit? I have removed the 10 screws that are deep inside the triangular structure, but something else is holding it on (the ac power connections?).

Also, Tom what do you mean by a diode voltage? Is there a voltage drop across the diodes? Does it have to be powered by 12v in order to get this drop? And which solder joints on the circuit board are the diodes connected to? Thank you so much for the help!

No your multimeter needs to be set to diode check and then measure from phase to dc bus.

[Boxster EV]

I managed to spin my LDU at the first attempt yesterday, by following/studying the contributions in this thread. Couple of observations:

Pedal:
The pin outs for my 2002 BMW e46 pedal are not the same as referenced in page 5 of this thread. I believe the correct e46 pin out are as follows:

Pin 1 is ground (taken from pin 15 of the 23 pin ampseal)

Pin 4 is signal (taken from pin 13 of the 23 pin ampseal)

Pin 5 is 5v (taken from pin 12 of the 23 pin ampseal)

Encoder:
I had to switch the middle two encoder wires, seemingly like everyone else.

LV power supply:
I tried to test the LV side with a cheap power supply, but it wasn’t man enough and I wasted 2 hours fault finding. I’d strongly suggest using a fully charged 12v car battery in future.

HV power supply
For HV I powered my motor with 65 Volts DC which was more than adequate for testing purposes. I followed the earlier instructions to set UDCMIN and UDCSW to 0.

You can see a short video of my test in the project section viewtopic.php?f=11&t=210

NB - my motor was not cooled and only had half the bolts fitted, and no fluid inside. This was fine for a very brief low rev test.

[P.S.Mangelsdorf]

Damien, are there any LDU boards left? The webshop lists them as "In stock" but doesn't provide the ability to purchase. I had wanted to purchase one right away but had to wait until I got paid today.

[Sutho]

After a short drive yesterday, my main contactor shut off, and then my precharge resistor and relay overheated/ caught on fire. The cause: my tesla motor appears to be shorted out, as in the resistance between the 2 power leads is 0 ohms. Any ideas on what this might be? I'm hoping it's not catastrophic.

Sounds like the same thing that I experienced with the Small Tesla Drive last week. I have the motor in a support rig (Model X chassis) and have cooling running. Blows precharge resistor if attempting to power-up. Found inverter to be shorted.

I think either one of these scenarios to be the cause:

Possible Cause 1:

- I am using a 63A HV DC Breaker as HV protection in lieu of a fuse as did not intend to push the motor too hard while free-spinning
- While testing the plot function, I did push the motor a little harder than anticipated and the breaker tripped
- Motor was spinning at 12,000rpm when this occurred so the presumption is that, without the battery pack connected, the back-EMF from the motor spinning at high rpm damaged the inverter

Possible Cause 2:

- If the above is unlikely, then maybe the breaker tripping was a symptom of the inverter failing due to the parameters not being ideal for a free-spinning motor and me punching the accelerator pedal
- The parameters used were Marco's which did seem to be working well

A plot of the moment my SDU inverter failed is below:

[tom91]

Possible Cause 1:

- I am using a 63A HV DC Breaker as HV protection in lieu of a fuse as did not intend to push the motor too hard while free-spinning
- While testing the plot function, I did push the motor a little harder than anticipated and the breaker tripped
- Motor was spinning at 12,000rpm when this occurred so the presumption is that, without the battery pack connected, the back-EMF from the motor spinning at high rpm damaged the inverter

Yes she is propper fried now, any removal of HV on a spinning motor above anything but near idle speed will cause a huge spike in voltage and fry the inverter.

So a dropped contactor or blown fuse can be catastrophic.

Unfortunate this means you now need a new inverter, which does not get sold standalone so you will need a new drive unit.

[Sutho]

Thanks for the confirmation...and certainly something that others need to consider in their build and testing.

Fortunately I have some other drive units.

We have attempted to pull the SDU inverter out of the housing to investigate, but not as easy as expected. Wrong thread for this though...

[joromy]

Would the tripmode setting "keep contactors closed" or "keep precharge closed" have helped, to avoid this?
Not if the fuse blows, but in other cases.

[tom91]

Would the tripmode setting "keep contactors closed" or "keep precharge closed" have helped, to avoid this?
Not if the fuse blows, but in other cases.

Yes it should, but running a small and unsuitable DC MCB is highly not recommended.

[P.S.Mangelsdorf]

So a dropped contactor or blown fuse can be catastrophic.

Please forgive my ignorance, but don't we use fuses to protect components? If a blown fuse is going to destroy stuff
A) What's the point in using a fuse at all?
B) What do we do/use to actually protect components?

[arber333]

So a dropped contactor or blown fuse can be catastrophic.

Please forgive my ignorance, but don't we use fuses to protect components? If a blown fuse is going to destroy stuff
A) What's the point in using a fuse at all?
B) What do we do/use to actually protect components?

Desat circuit or other fault circuit should protect your IGBTs and mainboard. Obviously to some extent only...
Primary mission of fusible element is to protect cables inside the box from burning up of too much current whilst causing your car to also catch fire and causing the catastrophic event of burning down your house. Fuse it not to protect your component, but your person when you are sleeping and cant control the circuit. It is "circuit be dammned" lets save some people and property thing.

[P.S.Mangelsdorf]

So a dropped contactor or blown fuse can be catastrophic.

Please forgive my ignorance, but don't we use fuses to protect components? If a blown fuse is going to destroy stuff
A) What's the point in using a fuse at all?
B) What do we do/use to actually protect components?

Desat circuit or other fault circuit should protect your IGBTs and mainboard. Obviously to some extent only...
Primary mission of fusible element is to protect cables inside the box from burning up of too much current whilst causing your car to also catch fire and causing the catastrophic event of burning down your house. Fuse it not to protect your component, but your person when you are sleeping and cant control the circuit. It is "circuit be dammned" lets save some people and property thing.

Understood. I let my racing brain get ahead of itself. To the extent of my experience, I've had lots of contactors and disconnects that keep the circuit broken if I'm not in the car driving it.

i.e. fuse has been there for if I get over zealous tuning the car and ask for too much power.

[Sutho]

My experience is that, when pushing the limits, you are more likely to drop a BMS contactor so it is a major concern if the inverter is going to blow.

While I know that this back-EMF issue can occur in other inverter/drive systems, I haven't experienced this with inverters like Rinehart (Cascadia) when the BMS contactor has dropped due to pushing the limits beyond what was programmed. I am hoping that this particular failure was due to the high RPM which created a voltage higher than the rating of the IGBTs and that at lower RPMs the voltage could be within rated limits.

[P.S.Mangelsdorf]

My experience is that, when pushing the limits, you are more likely to drop a BMS contactor so it is a major concern if the inverter is going to blow.

Which comes back to part B of my question, how do we protect both the inverter and batteries when driving aggressively/racing?

[Sutho]

[/quote]
Which comes back to part B of my question, how do we protect both the inverter and batteries when driving aggressively/racing?
[/quote]

Maybe something like this:

- Fuses sized to protect the cable, but large enough to not blow unless something catastrophic happens....like a dead short in the inverter or other HV components
- The BMS doesn't open the contactor when there is a fault when driving, but instead triggers a large warning light telling the driver that there is a battery issue and they should stop driving immediately
- The BMS should still be able to open the contactor if there is a charging fault

Not foolproof, but probably more suited for motorsport where the driver needs to take more active control of protecting the batteries and drive system.

[Kevin Sharpe]

Which comes back to part B of my question, how do we protect both the inverter and batteries when driving aggressively/racing?

By design... here's what Musk said in relation to 'ludicrous mode' -

"While working on our goal of making the power train last a million miles, we came up with the idea for an advanced smart fuse for the battery. Instead of a standard fuse that just melts past a certain amperage, which means you aren’t exactly sure when it will or won’t melt or if it will arc when it does, we developed a fuse with its own electronics and a tiny lithium-ion battery. It constantly monitors current at the millisecond level and is pyro-actuated to cut power with extreme precision and certainty."

https://jalopnik.com/the-tesla-model-s- ... 1718577723

[Sutho]

Yes. Pretty sure the primary reason was that the fuse system needs to support 1600 Amps

[arber333]

Yes. Pretty sure the primary reason was that the fuse system needs to support 1600 Amps

Maybe for racing we think of another angle. When car is not racing it is usually not in use or is made much tamer to be able to drive in day2day traffic correct? Then what we can do is some preparation maintenance before the race.
1. Take out the original fuse element which is used only in regular driving.
2. Put a direct link inside that will NOT break the connection.
3. Use whatever number of power contactors in paralel (2 on + side and 2 on - side) so you are able to break connection through BMS or on drivers demand.

Since on the racetrack you are in controlled enviroment you can afford to not use a fuse on power lines. BUT you need to connect all GND lines of contactors through a driver accesible (red) switch inside the cabin to break contactors on drivers discretion.

I use this anyway and TUV inspector was satisfied....

[Kevin Sharpe]

Since on the racetrack you are in controlled enviroment you can afford to not use a fuse on power lines.

That's your choice but important to remember that all the Tesla based race cars i've seen follow the Tesla scheme with a fuse and *never* interrupt the contactors during use. We really don't need to reinvent the wheel

[pm_dawn]

In my opinion:
If the BMS reads a to low voltage or to high current it should signal the inverter to back off or be in control of throttle signal to back the current off.
The inverter should be able to control the contactor but also be able to detect if the inverter has a short to be able to break the contactor/s
Usually fuses are used to protect the cables and to provide selectivity in the order of how things break.
In an EV the fuse also protects the battery from selfdestruct in case of a short in the cables.
Regards
/Per

[P.S.Mangelsdorf]

all the Tesla based race cars i've seen follow the Tesla scheme with a fuse and *never* interrupt the contactors during use. We really don't need to reinvent the wheel

Do you know of a source for a smart fuse like you mentioned previously?

I'm building for a drag racing event that requires longevity of components (both on track and street), and want to make sure I do whatever I can to avoid blowing stuff up by accident.

[jon volk]

So tripmode should be kept at 1=Keep DC switch closed for best protection of the inverter? I'm realizing it's been set to 0 all this time and Ive hit overcurrent shutdowns on plenty of occasions.

[Kevin Sharpe]

If the BMS reads a to low voltage or to high current it should signal the inverter to back off

Correct... I've seen several DU's destroyed by allowing the BMS to control the contactors directly... it's no coincidence that Damien designed the inverter controller to maintain control of the contactors.