- Mon Apr 06, 2020 9:48 pm
#41893
Hello all -
Whilst having a bunch of extra time on my hands during lockdown, I thought I'd embark on an EFI project that I've had in mind for a while to help keep me occupied. I'm in central Scotland and have a 25 year old bike with an old and rare ECU (TDD MINJ00). While there's nothing wrong with the OEM ECU as it is - it's not easily tunable (and I've just installed a big bore conversion on the engine..), and besides - it's certainly an interesting thing to learn about.
The bike itself is from the mid-nineties, and as mentioned, already running EFI with a 60-2 flywheel, and Marelli air/oil temperature sensors (no MAP) so it seems like it should be easy..I'm reasonably competent with Arduino and coding, but not so much with electronics.
So far, so good. I've got the DIY-EFI CORE4, with a breakout board (thanks James!).
In order to build my experience and start testing, I set up an Arduino to simulate the 60-2 pulse pattern, and hooked it up to Tunerstudio, and all is well. I've also sourced the correct bike loom connector and started to build a prototype interface using some stripboard - that's going well - at the weekend I confirmed that I could read TPS, IAT and CLT from the bike using Tunerstudio.
Taking a cue from Pazzi88's schematic for his BMW, I'm currently planning on using a pair of ISL9V3040P3's as coil drivers (the OEM ECU uses something similar, but I can't easily tell exactly which devices it uses due to the potting.
My next steps will be to use the TS test mode and be sure that I can get sparks and fuel on both, then to confirm that the VR conditioner is doing its' job. Once happy that it'll get the bike running, then I'll progress to getting a proper circuit board made up for a more robust construction (I've started to explore EasyEDA to learn that too).
I have two questions though, and it will no doubt be my last -
TIA,
Mike
Whilst having a bunch of extra time on my hands during lockdown, I thought I'd embark on an EFI project that I've had in mind for a while to help keep me occupied. I'm in central Scotland and have a 25 year old bike with an old and rare ECU (TDD MINJ00). While there's nothing wrong with the OEM ECU as it is - it's not easily tunable (and I've just installed a big bore conversion on the engine..), and besides - it's certainly an interesting thing to learn about.
The bike itself is from the mid-nineties, and as mentioned, already running EFI with a 60-2 flywheel, and Marelli air/oil temperature sensors (no MAP) so it seems like it should be easy..I'm reasonably competent with Arduino and coding, but not so much with electronics.
So far, so good. I've got the DIY-EFI CORE4, with a breakout board (thanks James!).
In order to build my experience and start testing, I set up an Arduino to simulate the 60-2 pulse pattern, and hooked it up to Tunerstudio, and all is well. I've also sourced the correct bike loom connector and started to build a prototype interface using some stripboard - that's going well - at the weekend I confirmed that I could read TPS, IAT and CLT from the bike using Tunerstudio.
Taking a cue from Pazzi88's schematic for his BMW, I'm currently planning on using a pair of ISL9V3040P3's as coil drivers (the OEM ECU uses something similar, but I can't easily tell exactly which devices it uses due to the potting.
My next steps will be to use the TS test mode and be sure that I can get sparks and fuel on both, then to confirm that the VR conditioner is doing its' job. Once happy that it'll get the bike running, then I'll progress to getting a proper circuit board made up for a more robust construction (I've started to explore EasyEDA to learn that too).
I have two questions though, and it will no doubt be my last -
- on the bike loom, the crank sensor connector has two connections, as you'd expect - one is the signal, and I can trace that to the relevant pin on the ECU connector, the other appears to be simply a ground - indistinguishable in connectivity or resistance from any other ground connection. I suspect that I should be ok with grounding VR- locally on the board, and connecting VR+ to the signal line, but just wanted to check in here, say hello and ask advice...
- On the DIY-EFI break out board, there is a header for bluetooth, and I have a HC-06 on it's way. However - I'm only measuring 1ohm resistance from the CORE4 board to the header on the TX line. I think this means that there isn't a voltage divider, and that the CORE4 will use 5v (where the HC-06 requires 3.3V for RX/TX). As a short-term test, will this be ok, or is it likely to burn out the HC-06?
TIA,
Mike