SDU can be fried with nothing connected on the HV side? [SOLVED]

[ardillolambo]

Folks,
After wiring all the 12 DC control signals...to do the encoder testing and realizing that I have to refresh every time (I guess the auto refresh doesn't work on mine) to see the status, find out that I have and overcurrent fault....and guess what, it doesn't go away. It is possible that I inadvertently fried my inverter? if so how is possible to do that with nothing connected to the HV terminals

Regards,

[P.S.Mangelsdorf]

Its possible that the inverter was fried in the accident that totaled the car. Otherwise, to the best of my knowledge no, just 12V shouldn't be able to fry the HV side.

[johu]

Check the power stage (2nd item): https://openinverter.org/wiki/Electronics_Basics

[ardillolambo]

Check the power stage (2nd item): https://openinverter.org/wiki/Electronics_Basics

Here is the sequence of events as I recall
1) checked all my switches, wiring etc., to be OK, turned out to be OK
2) gave the drive the "start" signal, looked at the drive status, says "waiting on start signal" or something around that.
3) Turn OFF the 12 volts check my wiring again, wiring was fine and dandy.
4) turned back ON 12v, gave it the "start" signal...hit refresh but this time I was away from the status page...did some stuff can't recall, when scroll down to the status page I saw "running"
5) At this point, I decided to power OFF the 12 dc and start the whole process again fresh and in order and remind me to hit refresh every time I do something like pressing the Start button, after that it is showing "overcurrent"

I guess when I was like a chicken without a head (Step 4) I may had the FWD & REV switches both in the ON position , that's the only thing I can recall that I did wrong, but again there is nothing connected to the DC bus bars.


I did the testing with the meter in the diode mode before moving forward to the 12 volts test. Will do it again as you pointed out.
What else to look if the test turned out ok?
what if the test turn out bad? Can I recover from this, or I have to find another inverter?

Regards,

[johu]

Nothing you did could have destroyed it. So that's fine.
If it turns out ok you can measure some spots on the board, current sensor values and such. Maybe there is a bad solder joint. Recently even an open resistor was found on a brand new board.

[ardillolambo]

Nothing you did could have destroyed it. So that's fine.
If it turns out ok you can measure some spots on the board, current sensor values and such. Maybe there is a bad solder joint. Recently even an open resistor was found on a brand new board.

I performed the tests again and I got the expected results for a good inverter.

1st test DC busbars (Diode Mode)
- Straight polarity: Beep (Short)...then voltage reading climb up to OL
- Inverted Polarity: Voltage reading settles at 0.634

2nd test 3 phase busbars (diode Mode)
- Negative lead on negative DC bus bar, positive lead on each 3 phase busbar: Beep...then voltage climb up to OL
- Negative lead on Positive DC bus bar, positive lead on each 3 phase busbar: voltage reading settles at 0.354
- Positive lead on Positive DC bus bar, negative lead on each 3 phase busbar: Beep...then voltage climb up to OL

I looked at the drop in board for any signs of bad components, can't identify something burned.
Since my project is a dual motor, I have another board, but to be honest don't wan to use it for now, until I understand how I get to this situation and if I can recover from it.

What else can I do/check for, now that I know the inverter isn't fried...after reading viewtopic.php?p=60992&hilit=overcurrent#p60992 I guess there is not too much to do, I'll end up buying another inverter

[johu]

Now I recall - with no DC cable connected the interlock will be interrupted. Bridge CONN3 on the board

Here's the schematic.
https://github.com/jsphuebner/Tesla-Fro ... ematic.pdf

Verify that the current sensors deliver 1.65V to the STM32. Thats where R9 meets R4 and R2 meets R1.

Then verify all inputs of IC5 are high, that is pins 1-6

[ardillolambo]

@Johu
I figured it out...it looks that I was providing a start pulse is too short, it goes in a split second to run and then to overcurrent, I just keep the start button a little longer pressed and it goes from overcurrent to start. I recreated the scenario of a short start pulse button press, then overcurrent current fault then keep it press a little longer then running. Also recreated coming from a fresh start with a longer pulse an it goes to running right away.

now that I'm sure my inverter isn't fried, I noticed few things....

1) every time I power up the ignition, I have to change the parameters back to "0", I don't want to mess around with it at this time, but later on when I set my parameter I guess there is a way to do keep them permanently on the inverter? if so how that is done.

2) No matter which way I connect my throttle pedal (from a 2017 Model S) using sensor 1 or 2...I get Throttle1 error, is there a way to set up the throttle pedal correctly using the interface?

Regards,

[johu]

1) Look harder buttons on top
2) https://openinverter.org/wiki/Schematic ... rottle_pot (of course you no longer type in commands but put values into the mentioned parameter)

[P.S.Mangelsdorf]

1) Look harder buttons on top

(Hint: Its one that says "save parameters")

[ardillolambo]

1) Look harder buttons on top
2) https://openinverter.org/wiki/Schematic ... rottle_pot (of course you no longer type in commands but put values into the mentioned parameter)

Yeah, I saw the buttons at top, but since I had paranoia about potentially having a fried inverter, and since I was not familiar with the web interface, the most reasonable thing.... is to ask, even when is a stupid question and I probably knew the answer, just want to make sure.

Thanks for all your help trying to figure this little issue out, hope this help someone else down the road.
Will follow the steps to the Throttle pedal configuring.

[Socko]

I was having this same issue with my fresh board in my small front drive that I just got in 2024. I had the inverter in my office after putting the board in and once I wired up the molex when I would put it into run mode it would give an overcurrent error. I tested in neutral forward or reverse. I tested long or short actuations. Always did it every time. Invertor phases tested correctly with a diode tester setting on my multimeter.

I put the inverter on the motor and it still did the same thing so I was fairly disheartened but I figured I would put high voltage on it and see.

I just got it wired up today and turned on 12v. Connected the HV. Actuated run, and no error. Put in gear, no error. Pull throttle and it spun up fine with no errors for first test.

Maybe this is just how sdu's work?

[johu]

During the bench test was the HVIL connected? I think it is a little bridge inside the DC cable that tells the inverter that the DC cable is connected. If it's bent or dirty this will also show as OVERCURRENT error, unfortunately. There is a jumper called "HVIL DISABLE" that gets rid of the problem. Don't see the point anyway.

[Socko]

During the bench test was the HVIL connected? I think it is a little bridge inside the DC cable that tells the inverter that the DC cable is connected. If it's bent or dirty this will also show as OVERCURRENT error, unfortunately. There is a jumper called "HVIL DISABLE" that gets rid of the problem. Don't see the point anyway.

I am not exactly sure how the High Voltage Interlock Loop works. I never had the orange cable plugged into the motor till I had the HV on it, so its possible that is connected to why my inverter was having issues. I intend to use the default cable terminals but I just jammed wires in the other side to spin it. Will it work if I cut the cable in half and just crimp terminals on it or should I short that jumper now to avoid issues later? I know there is a ground sleeve in the cables, that I assumed was to help with HV noise. Is there some other function right in the terminal on the motor?

Thanks for the help!