[k90]

D_BSD wire was added to DME for probing purposes and later to be connected to PWM generation board residing in electronics box. No extra BSD pins found so I used the vacant IBS pin. Having no IBS makes this slightly easier as the bus has only alternator, main coolant pump and oil level sensor. Couldn't be bothered to start pulling individual BSD slaves now until I try to guess first.

I performed some waveform captures on half a dozen conditions for now.
With cheapest tools ofc. FX2LP board(saleae 8ch clone) and resistor divider 10k / 2.4k probably does well enough for such slow comms. Software used with the particular board is Pulseview. I'm going to try to write the decoder for pulseview now.

Photos:

1. Gray wire added to pos 3. The vacant IBS slot.
2. First time using the heatshrink tape for labeller.
3. Test points for clips to attach to.
4. My logic analyzer setup for 12v signal. More than worth all the 10dollars or so. I should get one with a plastic case though. And maybe hantek usb scope or something that can stream analog signal to PC. And to Pulseview too for having both analog information and decoded bus information on the same timeline.

Captures in attachment ZIPs in chronological order:

1. Wakeup the car.
2. Switch ignition on
3. Start the car.
4. Car is running now. It has been running for couple of minutes now.
5. Stop the engine and switch IHKA rest on.
6. Open the door and switch IHKA rest off. The bus is in inactive state yes until I open the door.

Maybe the first captures of BSD bus in the internets? Couldn't find much.

[k90]

Bus state, bit decoding, frame bit numbering, bitfield decoding done.
Warnings for insufficient amount of pause bits, parity mismatch and frame length variations done.

BSD seems to be the slowest thing ever seen. 1200 symbols per second. Symbol being sync, 0 or 1. Frame lenght is 19 bits and this particular DME seems to do 5 sync bit pause between frames so 50frames per second.

To be added next is address decoding. I think I need to start pulling BSD slaves. Is it completelly safe to pull alternator and run the car? It's the easiest ofc. Actually I might not need to run the car as with ignition there is address 0x3, 0x4 and 0x6 and I have total of 3 BSD devices. Without ignition after wakeup it seems to talk to 0x3 only. Idk why.

By quick look I haven't seen any slave to master frames. IBS might do something like that to wake up DME.

[k90]

Summer is here. Project done by next winter?!

Added frame filter list and comma separated formater to the decoder to get values to "Excel" for plotting.
These charts indicate what I already knew. I think I need to buy or build a massive adjustable load to really see the values bottom out and see which direction means what. Judging operation from start transient where everything is moving everywhere is way harder than load applied to car in steady running state. EDIT: Also chart with clampmeter amps measurement from alternator lead show that nothing strongly correlates to amps output when talking only 40-70amps. Expectation is to see values change drastically when approaching rated current.
EDIT: Damn I'm lucky to find the coolant pump outputs it's supply voltage value.

Been writing bsd tranceiver C code for a microcontroller. Hope to test it to run old coolant pump soon as I have other things to do and probably need to pause the project soon.
EDIT: coolant pump is not the relevant part, but working tranceiver code is.

[grey78]

Hello everyone.. I did a diagnosis on my F36 with Bimmerlink and I see that my auxiliary heater is always "ON" even in summer and with a hot engine. With the engine OFF or with start and stop it switches OFF at 6% but then when engine start it always stays at 93% without switching off. Any idea?