lebowski controller serial connection setup

[whereswally606]

I tried to get an old controller of mine and bobc running again today after what might be 10years or so in a box. Needless to say i didnt get past the setup part.

if people could help me with the serial settings i.e. baud rate etc and also the terminal settings in say teraterm or termite that would be immensely helpful.

It could have been that i was using a usb to ttl adapter instead of the rs232 to serial but I don't have the port for the connection on my laptop so had to go the way of usb.

sad that i cant remember any of these details and worst that i couldn't find them via searching here or elsewhere on the web. anybody with any better memories could you please post here.

 

[amberwolf]

I never got to that stage with mine yet, but hopefully this info from the manual in this thread will help:

Lebowski's motor controller IC, schematic & setup, new v2.A1

I finished writing the setup manual for my controller IC. I posted it here (instead of mailing it only to the people that offered to test it) so that everyone can get an idea of the features that my controller IC has... What do yo guys think ? The following schematic and manual are...

endless-sphere.com

Some general remarks: Before connecting the battery make sure options d and e in menu c are set correctly. Use a fuse in the battery line. There should be no diode in the battery line. Make sure the current sensors are connected conform the schematic and that the default calibration values are written, even when current sensor calibration is not used (menu d, option l). An error here can cause severe damage to the output stage (again, make sure to have a fuse in the battery line).

The menu system was designed using 'gtkterm' under Ubuntu. The baudrate is 115200 baud, 8 data bits, 1 stop bit, no parity, no handshaking.

Under windows the freely available program 'termite' can be used, I tested version 2.8. The main difference between the two is that gtkterm transmits characters as they are typed while termite waits for an return before sending the characters. This means that whenever the chip waits for 'press any key', in termite the chip will only respond correctly when the return key is pressed. In the serial port settings of termite select 'append CR' and disable 'local echo' The chip can be placed in the setup mode by closing the setup switch connecting pin 19 to ground while resetting the chip. Upon entering setup mode a keypress on the PC is necessary before the chip displays the main setup menu.

 

[whereswally606]

Thanks amber the bit about the baud rate might be key

 

[amberwolf]

BTW, if you can't "talk" to anything with it, then to be sure your USB-serial is working, you can connect the TX to the RX on it without plugging it into anything, open a terminal program, and you should see everything you type come right back on the screen as if you hit every key twice. (loopback test).

 

[whereswally606]

So I had another go trying to talk to the controller. I think I now get a consistently garbled set of ascii every time I hit the reset on the controller so its progress.

@amberwolf Thanks for the tip about the loopback on the serial TTL device that is helpful and I didn't know that so i will store that one in my good to know part of the grey matter.

having googled the garbled ascii problem and knowing I have the correct baud settings leads me to believe that i am simply using the wrong type of adapter. normal ftdi TTL is not appropriate here maybe with its max 5v compared to the rs232 which can be + or - 9. I know this has worked with a standard rs232 port in the past so i should bite the bullet and buy an adapter to usb for my laptop. What is annoying is that I used to have one and i have spent the last hour looking in every inch of my empire of dirt to no avail.

tomorrow I may dig out my desktop from my office and see whether i can talk to it on the native rs232 and if so then place and order for the USB to rs232 cable.

Thanks for the help so far hopefully this will end up being useful.

 

[amberwolf]

Does the board have a Max232 chip or similar buffer chip for level translation? If not, it's very likely to only use the native voltage level of the MCU itself for the TX/RX lines, and greater voltage could damage the MCU.

The one schematic I located in a Lebowski thread doens't show any buffers, just direct to the MCU pins for "RS232", but I couldn't quickly find which specific MCU is used, so I couldn't look up what those pins expect signal-wise.

 

[amberwolf]

For garbled ascii, sometimes having the wrong data bit amoutn or wrong parity will do that, too.

Some terminals also have different character sets or "languages" available and that might have to be set right too.

When I was testing Incememed's sSFOC5 I had troulbe with that a fwe times.

 

[whereswally606]

I have just this minute setup the controller on the floor of my office under my desk at the back of the only computer I own with a native rs232 port. Within 2 seconds of doing this I had the initial setup page rendered clear as day. Knowing this, I will spend my £4 on a new USB to RS232 converter and hopefully I will be able to have a portable configuration changer in the form of a laptop.

Thanks for the help AW you were bang on on all the necessary details. Will keep you posted of any progress I make.

 

[whereswally606]

well i got the lebowski setup screen working on my laptop and then i broke everything. nothing would make my motor run so i did something stupid which was

a) PWM frequency: 8kHz
b) deadtime: 499ns
c) dutycycle testsignal: 27%
d) toggle high side polarity, now active LOW I toggled this from HIGH
e) toggle low side polarity, now active LOW I toggled this from HIGH
f) test PWM signals and ran this.

g) autocomplete

h) loop sample frequency: 20.39 kHz

z) return to main menu


this instantly blew the fuse on the dcdc.

i replaced the fuse on the dcdc but now can not fire up the controller as it is either got a dead short in the power stage or the settings that are incorrect are now saved that way.

not sure how to proceed, whether it is possible to inject the 5v to wake the chip up and just try to reset the settings.

had some scary welding at the battery pole when i tried to connect it using one of the bullet connectors instead. (clearly under a large load)

 

[amberwolf]

That I don't know. I would test the motor-drive FETs to see if any are shorted. (if they are IGBTs I don't know the test procedure, but FETs are pretty easy.

If they are, you could disconnect the B+ buss to them, then at least check out the controller brain.

 

[whereswally606]

Can not easily disconnect the power stage and its not individual fets its a ixys sixpack so i think i need to cut the power connections between the powerstage and the braindriver board. Then power just the brain driver so i can reset those low/high flags again and retest. Every chance ive blown that chip.

 

[marcexec]

Can not easily disconnect the power stage and its not individual fets its a ixys sixpack so i think i need to cut the power connections between the powerstage and the braindriver board. Then power just the brain driver so i can reset those low/high flags again and retest. Every chance ive blown that chip.

Did you get it running again? I've re-entered the Lebowski world recently and started a little guide, maybe the IMA powerstage is of interest to you?