Cruise Control Can Message

[jamescooper]

I have a Tesla in a GT40. I am trying to get the Speed Limiting Cruise control working.
It is almost there but being in Australia it is driving a bit like a kangaroo, giving power on and off in rapid sequence.
Clearly the Can output is not correct. I wonder if anybody can let me have an example of their output or comment on my attached output.
Be aware that due to the can inspectors sample rate the table appears to be out of sequence, so I have a second shorter table that I have added that has things in sequential order.

I wonder if Bite 3 should ever have a 0

Sorry the headers don't line up. It is more clear if you copy the text and paste into excel.

// Timestamp Bus CANId Rx/Tx d Length 0 1 2 3 4 5 6 7 Decimal K Decimal L
0.450747 2 3F Rx d 8 00 00 00 41 4C 4C 64 00 65 76
0.650747 2 3F Rx d 8 00 00 00 C1 CC CC 64 00 193 204
0.850752 2 3F Rx d 8 00 00 00 01 0C 0C 64 00 1 12
1.050754 2 3F Rx d 8 00 00 00 81 8C 8C 64 00 129 140
1.250755 2 3F Rx d 8 00 00 00 00 0C 0C 64 00 0 12
1.450747 2 3F Rx d 8 00 00 00 41 4C 4C 64 00 65 76
1.650748 2 3F Rx d 8 00 00 00 C1 CC CC 64 00 193 204
1.850748 2 3F Rx d 8 00 00 00 01 0C 0C 64 00 1 12
2.050747 2 3F Rx d 8 00 00 00 81 8C 8C 64 00 129 140
2.250749 2 3F Rx d 8 00 00 00 00 0C 0C 64 00 0 12
2.450748 2 3F Rx d 8 00 00 00 41 4C 4C 64 00 65 76
2.650747 2 3F Rx d 8 00 00 00 C1 CC CC 64 00 193 204
2.850748 2 3F Rx d 8 00 00 00 01 0C 0C 64 00 1 12

1.250755 2 3F Rx d 8 00 00 00 00 0C 0C 64 00 0 12
0.850752 2 3F Rx d 8 00 00 00 01 0C 0C 64 00 1 12
0.450747 2 3F Rx d 8 00 00 00 41 4C 4C 64 00 65 76
1.050754 2 3F Rx d 8 00 00 00 81 8C 8C 64 00 129 140
1.650748 2 3F Rx d 8 00 00 00 C1 CC CC 64 00 193 204

[tom91]

I have a Tesla

So you are saying you parked a Tesla in a GT40?

Or do you mean you have a tesla Large drive unit fitted with an Open Inverter logic board?

This is in the wrong section of the forum, should be here https://openinverter.org/forum/viewforum.php?f=3 or https://openinverter.org/forum/viewforum.php?f=10

[johu]

moved to inverter section.

Please plot cruisespeed to make sure that isn't the culprit.
What you describe sounds like controller oscillation, in that case raising speedflt would help, 1 at a time. Perhaps also lowering speedkp if further possible

[jamescooper]

For some reason the Kangaroo drive has stopped, No logical reason, as I did not even get as far as your suggestion. But no cruise control.
If you are able, could you create a can export as per my example and send it to me. I am guessing no need to detail what speed you are logging. I just want to see that the absolute basics are right.
Thanks in advance

[johu]

Of course there is a logical reason. And I'm still waiting for a plot of cruisespeed. We can tackle this systematically. Or not at all.

If you want to find out how developers get burned out look at this thread. I ask for X, I get nothing. So I have to ask again. And I can't help. Great way to trigger maximum frustration on both sides.

[jamescooper]

The cruise limiting appears to be trying but the speed is erratic. It as if the the power is switched on and off about every 1/4 seconds. both in cruise off and on.
Attached are two Can dumps on Cruise on one off.
I have adjusted the speedflt from the origional setting of 5 to 10 but with no effect.
Each of the spread sheets has details of drive.
My assumption is that there is something wrong with the coding.

[media][/media]

[tom91]

post the params that are currently on your driveunit.

[jamescooper]

Attached Parameter file

[tom91]

https://openinverter.org/wiki/CAN_communication

The mapping is as follows [startbit, endbit]:

pot[0:11]
pot2[12:23]
canio[24:29]
cruise[24]
start[25]
brake[26]
forward[27]
reverse[28]
bms[29]
canrun1[30:31]
cruisespeed[32:45]
canrun2[46:47]
regenpreset[48:55]
cancrc[56:63]

As i taught you have not implemented it as per spec.

Please provide logs that are decodable by savvy-can next time. I have attached the DBC for the open inverter spec.

[jamescooper]

here are the two files from Can Inspector
The Motec can inspector has a file extension of .asx that I can not attach, so I have changed it to .txt.
I will have a look at your file tonight to see if I can spot an error.

[jamescooper]

I have looked at your CAN mapping and compared yours to mine and It looks as if I have no errors.
if I take Byte 4 or 5, convert to Decimal, Subtract byte 3 and multiply by 255 I get the same result for all four lines. Apart from line 1 that starts with the Zero. But Yours like mine if you subtract 1 rather than 0 we get the same constant. 2805 for me and 8415 for you.
This indicates to me that the instructions to the CAN are correct.
My assumption is that there is some setting that is wrong with the parameters.
You did suggest that I increased speedflt It was originally at 5 and I increased it 1 by 1 up to 10 but with no joy. I can try pushing it up higher if you think that makes sense, but my thinking was that you saying "just 1 at a time." you would not have recommended going much higher than I have.
There is another thought. Am I setting a too low speed for the system 40kph? School zone speed. But as the car is jumping about I do not want to take it onto other than back streets.

[tom91]

I have looked at your CAN mapping and compared yours to mine and It looks as if I have no errors.

What are you on about. I made a .dbc file for decoding the CAN traffic and it shows bit 24 toggling. note we start to count at bit 0.

Here its layed out, byte 3 bit 0 is bit 24.

Your "cruise off" log shows byte 3 jumping from 0x00 to some value.

Take away the CAN stream and the vehicle should drive fine again. Set all your settings back to where you started then work through the CAN sending setup until you can decode it as per dbc provided.