openinverter forum

Jack Bauer wrote: ↑Mon Mar 23, 2020 3:49 pm Do you know how the byte 2 checksum is calculated?

/*
 * Checksum is done as follows:
 * Sum bytes 1 to 7 and 16 bit can ID
 * Add hi byte of result to low byte
 * use low byte result as checksum WTF !!!!!!!!!
 */
canID = 0xAA;
byte1 = roll;
...
checksum =      byte1 + byte2 + byte3 + byte4 + byte5 + byte6 + byte7 + canID;
checksum = (checksum / 0x100) + (checksum & 0xff);
checksum = checksum & 0xff;

int __fastcall can_checksum_calc(int a1, int a2, int a3, int a4)
{
  int *mailbox_struct; // r7@1
  int v5; // r3@1
  int v6; // r6@2
  int i; // r7@4
  __int16 v8; // r4@6

  mailbox_struct = &can_routine_table[6 * a1];
  v5 = *(mailbox_struct + 0x15);
  if ( a4 )
    v6 = mailbox_struct[3]; // This is the can ID like 0xAA, etc.
  else
    LOWORD(v6) = 0;
  for ( i = 0; i < v5; i = (i + 1) & 0xFF )
    LOWORD(v6) = v6 + *(a2 + i);
  v8 = (v6 >> 8) + v6;
  if ( a3 == 4 )
    LOBYTE(v8) = (((((v6 >> 8) + v6) & 0xFF) >> 4) + ((BYTE2(v6) + v6) & 0xF)) & 0xF;
  return v8;
  

Jack Bauer wrote: ↑Tue Mar 24, 2020 6:37 pm Today saw a major milestone. I drove the Bexus on the motorway today to the local NCT (annual test) and it was just perfect. Despite the crappy test battery sagging horribly at no point was I running out of voltage. I would like to say that this should put the 650V nonsense to bed. A diesel diff swap is a lot easier:) We went from 3.6 as fitted to the 735i to a 2.8 from a 730d. All for a cost of 100 euros.

Here is some video to enjoy:

Jack Bauer wrote: ↑Tue Mar 24, 2020 6:37 pm Today saw a major milestone. I drove the Bexus on the motorway today to the local NCT (annual test) and it was just perfect.

Jack Bauer wrote: ↑Wed Mar 25, 2020 3:28 pm We have reversing lights:) Am I right in looking at message 0xA8 in that it has two bits that flip on press of the brake?

  bn_can_send_temp_byte0 = (32 * v15) & 0xE0 | (bn_can_send_temp_byte0 & 0x1F);

sfk wrote: ↑Wed Mar 25, 2020 4:14 pm
Looks great. Was MG1 in use and contributing any torque at all or was it just MG2 doing the driving?

Jack Bauer wrote: ↑Wed Mar 25, 2020 6:26 pm

sfk wrote: ↑Wed Mar 25, 2020 4:14 pm
Looks great. Was MG1 in use and contributing any torque at all or was it just MG2 doing the driving?

MG1 and MG2

slow67 wrote: ↑Wed Mar 25, 2020 7:02 pm Except when you backup!

xp677 wrote: ↑Mon Mar 30, 2020 11:29 pm

slow67 wrote: ↑Wed Mar 25, 2020 7:02 pm Except when you backup!

You could use MG1 as well for reverse if you wanted, I left it out of the code as it seems pointless for a reverse gear. I never tested it on a car though, so no idea how it handles that mechanically. The rotor will still be spun by MG2, no idea where that current goes. I guess it's just fine looking at Damiens videos.

In the same vein, you could run the whole thing in reverse as easily as forward, in case you have a less-common differential setup which is backwards to normal.

drprox wrote: ↑Sun Mar 29, 2020 9:51 pm Damien, did you remove the engine ECU completely or leave it in?
I am curious as to how much of the electrical system needs 'simulating'. I assume the bits you are currently trying to work out the messaging for is the DSC/TCS system? What about other things such as iDrive, airbags, central locking etc?

Kevin Sharpe wrote: ↑Thu Feb 27, 2020 8:04 pm

mattndex@gmail.com wrote: ↑Sun Feb 23, 2020 2:49 pm Is there a chance of you posting a schematic of how the high voltage junction box is wired.

The participants on next weeks course are building a HVJB for the Grey Goose project. It will be very similar to that used on BEXUS. I'll ask them to document the build and post details in that thread

IMG_20200227_152751.jpg

arturk wrote: ↑Sat May 02, 2020 7:49 am What are AMP ratings on those fuses? Fast acting?