Change of direction handed over to LDU?

[medo]

Hi, I have a problem with the Drive direction from the VCU to the LDU. The VCU clearly shows that the Drive direction is changing, but this isn't being passed to the LDU. Do I need to specify this in the canmapper? It should normally work without it. VCU version 2.30, LDU version 5.39

[tom91]

It should just be the "standard" CAN Map. I am currently away so can have a look when im back home at some point in the middle of January at my bench setup.

It was working before, so something must have changed.

[medo]

No problem, I'll do another test with the Jaguar gearshifter. Have nice holidays and a happy new year!

[medo]

Quick update: after testing with the Jaguar shift lever, there was no reaction on the LDU side, so it could be a software bug.

[johu]

What is the value of canio, seldir and the respective din_ spot values?

[medo]

Pot1 and Pot2 are sent normally from the VCU to the LDU, as are the brake and start signals; only the drive direction remains at 0 and does not react at all.

[johu]

That is not the answer to the question I asked

[medo]

google übersetzer ist schon beschissen

LDU

VCU

with the car in forward gear and the brake pedal pressed down.

[johu]

ah!

ok, canio, which is the unprocessed value received over CAN just receives Brake. So for some reason Zombie is not turning on any direction bit.
Does it behave differently when downgrading inverter firmware?

edit: on Zombie din_forward is Off as well

[medo]

I'll test it right away with version 5.35; yes, no signal is coming from the VCU via CAN for drive direction, only brake, start, and accelerator pedal are transmitted.


edit: i test it, The same applies to the downgrade; it can only be due to the VCU.

[medo]

News about the faulty CAN communication to OI ?

[manny]

Does the gear inputs change on de vcu spot values?

Please post parameters json files.

[medo]

Does the gear inputs change on de vcu spot values?

Please post parameters json files.

Take a look at the screenshots above, and maybe read all the posts as well.

[tom91]

Did you ever have this working?

I did do some dev work on this, likely in the vehicle testing branch waiting on people to verify it all works as it should before putting it into the master branch.

I have not had any feedback and logs on this so likely never moved into the Master Branch for release.

[medo]

Yes, that worked a while ago, we even discussed it in another thread, but it hasn't worked since the new update 2.30. I sent the signal to the LDU via savvycan, and that works perfectly.

[tom91]

So you are saying, it worked before. What version did it work on.

You upgraded from what to V2.30A?

[medo]

Back then you reprogrammed something and it worked, unfortunately I can't find that working firmware anymore.

https://openinverter.org/forum/viewtopi ... 664312091a

[tom91]

So okay, vehicle testing branch feature needs to be introduced into the release then. And further testing done also by others.

[medo]

Is this the file that's responsible for all of this? CAN_OI.cpp
Oh, I see, a lot of code is missing.

[medo]

Quick report: I replaced this in Can_OI.cpp

Code: Select all

if(Param::GetBool(Param::din_forward) && opmode==MOD_RUN) tempIO+=8;//only send direction data if in run mode
if(Param::GetBool(Param::din_reverse) && opmode==MOD_RUN) tempIO+=16;//only send direction data if in run mode

with

Code: Select all

int dir = Param::GetInt(Param::dir);

if(dir == 1 && opmode==MOD_RUN) tempIO+=8;
if(dir == -1 && opmode==MOD_RUN) tempIO+=16;

now the direction change via the shifter works almost perfectly, only the direction doesn't match what's displayed on the web, it's the opposite. Also, is it normal that if the engine is still turning and you select a direction and accelerate, it still drives in the previous direction, or do you have to be completely stationary for it to accept a direction change?

[medo]

I've noticed another thing: the fan starts immediately, even when the temperature is at 8 degrees and 40 is the default setting. I have to set the temperature in the web interface to over 65 degrees for it to turn off; this must also be a bug.

[medo]

I've now tested the updates from master-dec-update, everything is running and switching as expected, but the LDU is still running backwards and communication with the web is not possible (Node3), and the temperature is still not correct with fan activation.