openinverter forum

Hello,

I recently purchased an LDU logic board from the shop. After attempting a bench test, I found that the a few measured Spot Values are behaving quite erratically.

"udc" and "temphs" are both swinging around wildly.

udc: Seems to swing around between 20v and 550.15v.
tmphs: Seems to stick at exactly 100c, but will randomly jump around 43c for a ~3 sec, then go back to 100c.
uaux: Sticks at exactly 16.43v despite my changing the 12v up and down a bit. Does not respond to changes.

These problems occur both on the bench, and while hooked up to the drive unit. Precharging the drive unit doesn't seem to affect the readings.

I double checked the connector soldering job I did under a microscope and things look fine.

EDIT Additional notes:
I tested VCC_5V (both), 3V3, and -5V at the DC BUS Sense connector. The VCC_5Vs are reading at 5.41v, -5V is at -5.35, and 3V3 is right at 3.3v. Not sure if this is cause for concern or not.

Any thoughts?

uaux is indeed not implemented. The others shouldn't jump around though. I'm currently not home, will get back to you tomorrow with some more detailed checks.

Thank you kindly!

I discovered a loose solder ball between some pins on the STM32, between the UDC_11 pin and another pin (VDD I think?). I reflowed the solder on that side of the IC and the udc measurements are behaving better now.

Since tmphs is still giving me trouble I attempted to reflow the side of the chip that it's pin resided on, however that didn't resolve the issue. It's still reading incorrect temps.

I did attempt to spin the motor up on ~24 volts but it keeps tripping overcurrent right when I "start" it. Wondering if there's something else going on.

Ok, loose solder ball sounds not good. Can you post a hires macro picture of the processor and the entire board?
Also let me know your order number via PM if you prefer.

Attached are the requested photos. I had to take two photos of the micro controller as I couldn't zoom out far enough with my microscope.

I also attached the photo I took when I first discovered the solder ball. It can be kinda hard to see, so there's two, one that is edited to point it out.

I PM'd you my order number. Let me know if you need better photos of anything!

Ok, so the remaining problem is tmphs. tmpm reading is stable? What about pot and pot2?

Can you measure the voltages on IC12, Pin 1, 2 and 5 with "IGBT DRIVER" connector is plugged in?

tmpm stays at around 7.3c-8.1c, which is strange because ambient temp is 21c. pot1 and pot2 appear to be stable.

Pin 1: 1.68v
Pin 2: 1.72v
Pin 5: very low, maybe open circuit?

Then Pin 5 is probably the culprit, or rather the signal that feeds it (TEMP_B). Maybe cold solder joint on the connector (pin 16 of IGBT DRIVER).

I tested from the connector pin back to the chip and it actually tests fine. So maybe there's something wrong with the temp sensor within the inverter? I'll have to take a look later.

For a quick test, I raised the max temp and tried to spin. The inverter makes a whining noise, however, the motor does not spin. The "angle" value is changing quite a bit despite there being no movement.

Going to try to put some more time into it later tonight.

I put some more time into this last evening and the results are the same. Motor refuses to spin. It also still randomly trips the "overcurrent" safety despite me never seeing >10a in the spot values.

Still trying to decide whether it is board or inverter related. Could the phase with the erratic temperature have some kind of problem?

overcurrent unfortunately can mean all sort of things, from actual over current (unlikely as you say) to desat, uvlo or over temperature. Check on phase B whether 12V supply is stable. Also while running measure IC10 pin 6 and 2 whether they are high (about 5V) all the time. Of course scope would be preferred but multimeter might show something, too.

Too bad it takes like 4 weeks for another board to arrive, otherwise I'd just shoot one over.

KingJohn wrote: ↑Fri Mar 03, 2023 3:56 pm I put some more time into this last evening and the results are the same. Motor refuses to spin. It also still randomly trips the "overcurrent" safety despite me never seeing >10a in the spot values.

Any news on this?

Some interesting developments here. I did get it to spin! It appears my flux is more conductive than I remember and was causing temp sensor to read wrong. I gave the board a bath in IPA and scrubbed off all the flux I could see.

It also wouldn't spin up on 24v with just 1850 boost. Setting it to 6000 let it draw around 22a and start rotating.

Once rotating, though, I heard a strange "knocking" sound. After some research I discovered this is likely the "milling noise" others have suffered from, so I disassembled the unit. The bearing on the motor shaft (on the motor side) appears to be the culprit as it feels the worst, but it's hard to tell.

I have a new bearing on the way. We'll see how things develop from here!

Alright, electronic gremlins purged then

With the control board, yes sir!

However, I let it run on 24v and I did notice a bit of vapor (or maybe even smoke) coming from the power electronics.

It was slowly coming out right near where the inverter electronics meet the aluminum case. It didn't have a scent that I could detect. Nothing was hot either. So I'm assuming it was vapor of some sort.

The motor housing was full of rust, so I assume this had water intrusion through the coolant seal at some point, so maybe the coolant is aresolizing somehow? Not sure.

Once I get it back together I'll run it for longer and see if it continues.

hello all
I am trying to get my LDU (in the removed state on the table) to work. The temperature tmphs is stable already at 100°C when switched on. udc shows 7.9V, but is disconnected from the HV batterie. (four times 12V starter batteries) Engine temperature tmpm reads 12°C, which is realistic. When I raise temphsmax to 110°C, the motor very slowly starts to turn unwillingly, but the 20A fuse (in the HV supply line) melts right away.
Is 20A at 48V enough for a test run without load?
IC12 pin1 and pin2 about 1.7 volts, pin 5 zero volts. Pin 16 on Conn10 is also zero volts. All measured with an oscilloscope.
I checked the solder joint under the microscope and cleaned the flux.
Do you have any idea where the problem is?
Did KingJohn get his motor to work?
Thanks a lot!

Find out why it reads 100°C, i.e. what voltage are you seeing on the respective HS temp inputs?
Never run motors with just a low value fuse, always put a cooker element, kettle etc. in parallel to the fuse as unexpected disconnection from the battery can blow up your inverter.

Hello johu, in the meantime I was able to find a replacement LDU. After installing the Openinverter LDU drop-in logic board from the first LDU, I now have the same result. Tmphs shows 100°C. Could the drop-in board have a malfunction?
Thanks for the tip with the cooker element!

Ok, so it's obviously nothing with the drive unit then.
What are the voltages in the temperature inputs? Also did you check for swapped encoder cable (swapped A/B)?