Leaf Gen2 and EM57 Bench Test Problem [SOLVED]

[E46Driver]

Hi guys, received my Leaf Gen 2 package and installed it with no problems. Powered up with 12V, accessed the WEB interface and all looked good. Adjusted the following parameters for bench testing:

Code: Select all

"udcmin": 30,
"udcsw": 30,
"udcofs": 0,

"udcmax": 520, <------- No change, default setting

I'm planning to use 12 Sinopoly cells that I've used for testing for a few years. Full charge voltage 41.5V.

PROBLEM : persistent OVERVOLTAGE error, even with no HV connections.

Controller will not energize precharge relay (although, I'm guessing the OVERVOLTAGE error kills everything before it even gets to precharge). I do not get a precharge error - or any other errors.

I've temporarily wired the system like this post for basic function testing:

https://openinverter.org/wiki/File:HV_wiring.jpg

I tried with and without HV battery connected.
Did I miss a setting somewhere?
Why would I get an OVERVOLTAGE error with no HV battery connected?

[johu]

What is udc showing? What is udclim? Because that is setting the OVERVOLTAGE threshold

[E46Driver]

Hi johu, thanks for the quick reply!

udc showed 748.6 V -- there's the problem.

I thought the brain board might have been calibrated before shipping since it had the following values:

Code: Select all

"udcgain": 5.65,
"udcofs": 250,

I tried to recalibrate as per instructions here:
https://openinverter.org/wiki/Schematic ... structions
But how do I "...Now make sure the input to the sensor board is disconnected. Get the value of udc and write it to udcofs:..." How to disconnect sensor board when its inside inverter?

I tried with no battery to get offset 3842. And then connected battery to calculate gain 6.03.

So my new values:

Code: Select all

"udcgain": 6.03,
"udcofs": 3842,

get udc yields 41.34 V --- close enough for testing.

BUT -- when I remove battery, reading still shows 41.34!!!!

I suspect a problem with the voltage sense, but not sure where to look next. Suggestions?

EDIT : udclim was not changed - left at 540 as received

[Isaac96]

Okay just to make sure you're doing it right.
Calibration:
gain 1, offset 0.
Change offset until udc reads 0 at 0v input.


Connect your voltage. Adjust gain until the reading equals your actual voltage.
And it should work.

'Disconnect the sensor board' really just means don't have a pack connected. And be sure to wait until the inverter capacitors have discharged.


3842 for udcofs seems very high so that may be part of the issue.


When you connect a battery - are you measuring voltage at the inverter itself? If the contactors aren't firing then you won't get any voltage.

[E46Driver]

Hi Isacc96, thanks for the reply.

Yes, that is how I am performing the calibration.

I also thought my offset was strangely high.

I am measuring at the inverter and using manual precharge engagement. The caps are definitely charged - and I can measure the voltage slowly drop after removing the pack. When I perform "get udc" the value never changes regardless if pack is connected or not. Something strange going on. I am dreadfully fearing an inverter problem. I hope I can find the problem and repair if necessary.

[E46Driver]

So I disassembled the inverter and removed the adapter board. My theory (and hope) was that the brain board was not receiving a signal from the adapter board. And I was HOPING there was nothing wrong with the inverter.

Sure enough - to my embarrassment - I had 3 cold solder joints on the header for JP7. Pin 4 was one of them --- UDC.

So all is well and I have a spinning motor! Thanks for the tips which led me down the solution trail !!!

Ended up with these settings:

Code: Select all

"udcgain": 5.439,
"udcofs": 299,

So recall the original settings as shipped were:

Code: Select all

"udcgain": 5.65,
"udcofs": 250,

These should have been close enough to give an approximately correct reading. Thankfully it was an easy fix and I didn't break anything! Whew!

[Renegate2020]

Hello.
You didn't have contact at JP7?
I have a similar problem - an overvoltage error (667v) appeared while driving. When the high-voltage battery is disconnected, the error does not disappear.

[johu]

Yes sounds like it. The signal is pulled high on the adapter board and Nissans board sends an open collector PWM signal. No signal from Nissan -> maxed out reading

[Renegate2020]

Thanks Johu. Demolition will take place tomorrow.

[Renegate2020]

On pin 4 of the JP7 connector, I have + 5v. The high voltage battery is disabled. So there should be no tension there? Which side to look for the problem?
When disconnecting all wires from the inverter to the adapter board, the voltage on pin 4 changes to 3.4v. The error remains.

[johu]

which version adapter board is it? Can you post a picture?

[Renegate2020]

V3+ Nissan Leaf adapter board

[johu]

ok, that's super proven I also run this version myself.

On pin 4 of the JP7 connector, I have + 5v. The high voltage battery is disabled. So there should be no tension there? Which side to look for the problem?
When disconnecting all wires from the inverter to the adapter board, the voltage on pin 4 changes to 3.4v. The error remains.

So R2 pulls the udc PWM signal to 5V via 10k. Then it is divided with a 47k/(10k+10k) divider so it should idle around 3.5V

It seems the inverter pulls udc PWM to more than 5V then. If you unsolder R2 what happens on JP7.4?

[Renegate2020]

With R2 soldered and wires connected 5.8v and there is an overvoltage error, with wires disconnected we have 0v and the overvoltage error disappears.

[johu]

ok, something is wrong with inverter voltage signal then. It shouldn't act like this

[Renegate2020]

Thank you. Can you tell me what should be the signal from the inverter and on which connector to check?

[johu]

The pin that has the trace going to R2 and R5. CN501

[Renegate2020]

I disassembled the inverter a bit and such a nuisance. Could the reason be in uncontrolled regeneration?

[johu]

I found these inverters to be pretty invincible (and Zapatero too). But that mosfet looks quite unhappy and it seems the inductors let the smoke out...

[Renegate2020]

Absolutely agree. I want to understand why this happened so that it does not happen again in the future.

[Romale]

Absolutely agree. I want to understand why this happened so that it does not happen again in the future.

I had exactly the same. First I broke the wheel drive, the second time the inverter burned down. Every time it was a problem of uncontrolled regeneration

[johu]

Due to testing I have had a lot of strong, uncontrolled regen. It never caused any harm because
a) I never ever open the contactor once it is closed - only when I turn ignition off
b) The fuse can withstand it (Nissan original)

If you're using the inverter for controlling the contactor set tripmode to DcSwOn

[Renegate2020]

I had exactly the same. First I broke the wheel drive, the second time the inverter burned down. Every time it was a problem of uncontrolled regeneration

Something similar happened to me - after a sharp start, the wheel drive broke. I restored it. only now, after lightly pressing the accelerator, the car gives an error "OVERCURRENT" and does not go. Could the inverter be damaged? like the car accelerated, but not regenerated.