Adamw

The motor will be connected direct to the aux, but you still need the E-throttle relay on the aux9/10 supply so the ecu can cut power to throttle if there is a safety issue.

Yep all looks fine. I would just swap aux 9 & 10, it will work fine as is since you can reverse the control logic in G4X, it has always been links convention to have aux 9 connected to motor +ve. And yes “sensor ground” is labeled “ground out” in some of our pinouts (green wire in link looms).

Yes it is driven at cam speed so it is a cam trigger.

Any internal combustion engine will need air to "rev". If it can reach 5500RPM with the throttle closed then you have a large amount of air getting in somewhere (ie a big leak). Your idle valve isnt open much so I dont think that is the "leak". A PW of 7ms would only be 70%DC at 12000RPM. At say 2000RPM, the cycle time is 60ms, so 7ms out of a 60ms cycle would be around your 8%DC.

This is normal, it is due to the "fuel puddle" (fuel film on port walls etc) evaporating when the manifold vacuum increases. Sometimes called wall wetting or Tau. It will happen regardless of alpha-N(TPS) or SD(MAP). Injector size and spray pattern can make a big difference so this may be why you see more in this engine than previous. It is unlikely the cause of your stalling though, it is usually only present for a short time after the MAP changes significantly, by the time you are getting close top idle the manifold pressure should be stabalising. Perhaps your fuel tables numbers in the high RPM/High vacuum area are too big?

25-35% is common, start at 25%.

How do you know there is no spark? Your stall looks more like an idle speed control problem to me?

That is all. You likely wont need much. Set it initially to 2.5V at 14V and say 2.8ms at 13V, 4ms for all voltages below 13.

Nothing much out of place in your log or map, it appears the ecu is commanding everything that is needed for life. I would start by squirting some starter spray into the intake while cranking, this is a good way to isolate whether your problem is fuel or spark. If it doesnt fire with starter fluid, then I would confirm spark, confirm ign timing and also try changing the trigger offset by 360 deg.

Please do a trigger scope while cranking and attach that and a copy of your tune.

Ign switch input is B136-25, this is not to power the ecu - this is basically just a low current signal to tell the ecu to turn on the main relay. When the ecu receives 12V on that ign sw pin it then sends 12V out of pin B136-32 - this wire goes to the coil on the main EFI relay. When the main relay receives that 12V signal, its contacts close and supplies the main high current 12V to pin B136-8 & B136-9. G4X pinout in the help file is more complete than G4+ so will probably be more useful.

There is quite a bit of complexity in the subaru CAN, 3 different types of torque measurements, requested torque, many of the messages have frame counters for error checking, some odd codes for the ABS etc. I dont think you will be able to do it with the user tools in G4+.

Is it on 5.6.8 firmware? You could try connecting the DI-ve to gnd but it shouldnt be needed.

The ECU is outputting a 0-12V squarewave, the tacho originally worked off the high voltage coil primary (~100V), the tacho needs to be modified to work with the lower voltage signal, or you can use a tacho booster with an unmodified tacho. I have never studied the MR2 tacho or modified one so I cant offer any advice on what it needs.

That is fine with the single 3 core.

If the ecu is seeing lower voltage than the rest of the engine then you really want to fix that. Injector dead times and dwell etc will be wrong unless the ecu is has an accurate indication of the voltage powering these items.

The place to start would be to set up ECU logging with a range of common parameters - things like RPM, dwell, trigger error counter, ign & fuel cut, bat voltage, inj PW, all engine sensors, set it to log all the time. The next time it plays up download the most recent log out of it and you should hopefully have some clues whats going on.

The alternator control wasnt set up in your map, I have set it up and attached below. Load this in and give it a try. For the tacho, can you try changing the tacho fuction output to Aux . mx5tune alt set up.pclx

Yes it will need to running, the ecu needs to know crank angle define the knock window.

The alternator control should already be set up and working in our basemap. For the tacho, can you attach your tune

Bat voltage is only showing 12.4V in the log. For the dwell - yes, put 1.8ms in the 14V row and maybe 2.0 in the 13V row.

Yes, there is no signal from your cam sensor at all. It could be a wiring issue or a failed sensor. If you unplug the sensor and poke a paper clip or similar into the signal in the loom plug then touch that against ground on an off, you should see trigger 2 signal say "yes" when you do that is the signal wire is ok.

During bootup it will first display 8888, then the next screen should look something like below. The first 2 digits are the sensor type (LSU4.2 in this example), the 3rd digit has the dash at the bottom (narrow afr range), the third digit is the fuel type (looks like 6 but is inferring G for gasoline).

Hollinger usually have a proper 0-5V potentiometer on the gear barrel. That sensor is not going to be useful for gear shift control. We can still do open loop shift cut but it is going to be difficult for you to tune with no gear position feed back.

Can you attach the tune/map from the ecu and your AEM dash config file. Also give us a bit of info about what ecu you have, which CAN ports are wired together etc.