Adamw

which one does your loom match? https://dealers.linkecu.com/products/plugin-ecus/plugin-ecus-subaru

What do you mean by "knock counter dips"? The knock counter will just count up 1 every time the knock level exceeds the knock threshold. It is dead flat with no dips in the picture you have attached. You are better to attach the actual log file and tune for any help in future, a picture rarely gives adequate info.

The idle control looks somewhere in the ballpark to me now. The "misfire" is unlikely related to idle control, I suspect it is possibly a fuelling issue based on the 20-30% bank-to-bank variation you have in reported lambda. Possibly it is too rich, bank 2 lambda is actually the more correct measurement and bank 1 is reading leaner/looks normal as it has the misfire. Try leaning it out to see if it gets better or worse. There are lots of fuel settings I dont like in your map; IAT trim is still enabled, charge temp approximation is all zero, the fuel table Y axis is not really suited for an alpha-N tune, if you look at the fuel table surface view it is just 2 or 3 flat planes, almost like only 4 cells where actually tuned then someone just "joined the dots"? Then there is also a 4D table that adds 10% pretty much everywhere except idle? Warmup enrichment is active all the way to 90°C, fuel pressure sensor calibration is wrong, injector deadtimes dont look right, differential fuel pressure does not match what is set in the fuel settings and the fuel model has not been set to compensate for fuel pressure. For your APS, the sensors are connected to analog inputs on the trans controller, the trans controller then sends the pedal position to the ecu via CAN. It looks to me like the APS sensors just haven't been calibrated in the trans controller as the offset between sub/main is relatively constant and the rest position seems reasonably repeatable around 2%. This needs to be improved as you are just on the borderline of triggering a limp mode.

If the trigger error only occures at high load it is more likely an ignition noise problem or possibly a ground offset/voltage problem. Can you do a trigger scope capture at a similar RPM and load to when the error normally occurs so we can check if there are any clues showing in that.

Done.

yes. I always like to do this to get it more accurate though: Initially set to calculated ratio. Set staged injection table to 0 (all fuel through primary). On the dyno hold the engine at a medium RPM, medium load so you are up around 50% DC on the primary injectors (say 3000RPM & 30%TP for example), adjust fuel table so that the measured lambda is close to target lambda. While still holding at same RPM/Load, change staging table to 100% (all fuel now through secondary), you will get a little lean blip as it swaps over but if the ratio is correct your measured lambda should settle on the same value it was at before. If lambda did change significantly then adjust Sec/Pri ratio and switch back and forwards between 0 & 100% staging until lambda is the same regardless of which injectors are being used.

Mostly looks like the basics are in the ballpark now, you will now need to tune the dash pot function so it enters into idle control nicely. This varies a lot from engine to engine so you just need to experiment live to get that working nice.

ADIO 5-8 are capable of measuring frequency/speed, but the thresholds are even higher than G4X, so wont be useful for your intent in this case. From memory about 3.6V high, 1.5V low.

I just spotted the e-throttle target table has non-zeros in the top row also so that is not ideal. I have made some further changes as well as those already suggested above in the map that is attached to this post. Load this in, then you will have to log a cold start/warmup and adjust the base position table again but I think this should be a better starting point. 350z 24 may Idle setup.pclx

Why is the trans controller requesting 2 & 3% AP when the engine is meant to be idling? It should only be manipulating the APS signal when there is a torque modification required, so either there is something wrong with the trans controller setup or the APS wiring to it. The APS sub/main signals that the ecu receives should match and should be zero at idle. As for idle settings; increase your actuator integral to about 0.05, set idle ign max clamp to 30° and min clamp to -5°, do a log from cold start up to fully warm and adjust your idle base position to match the actual throttle opening that is required to achieve the target RPM. For example at the beginning of your log when ECT is 40° your base position is 8.7%, but it looks like it only needs about 5.5-6% to achieve your target rpm, at the end of the log when etc is 55°C, you need 5.5% throttle to idle but your base position table only has 4.5% commanded. Other tune suggestions: Increase VVT RPM and ECT lockout so that it isn't active at idle or when the oil is cold. Since your fuel table is referencing TPS, you ideally want the top row to reference normal warm idle TP - so for example when at normal operating temp, if it normally idles at 3.5% TP, then make the top row of your fuel table to be 3.5%TP. Above that you want finer breakpoints for the first 10-15%, so say 3.5%, 6%, 9%, 12%, 20%. Above 20% you can stretch them out - say 20,40,60,100.

Assuming your car originally had the coilpack like the pic below then the conversion harness should just work with no changes to the configuration. If this is what you had then start by removing a coil/plug and using the ign test to confirm if there is a spark or not. If your original coil looked more like the one below then there would need to be an ignitor bypass fitted.

Resistor leads? Resistor spark plugs? Suppressor fitted to the coil power supply? Coils grounded to the cylinder head? Where is the ECU grounded?

There is no reason why oil pressure shouldn't have worked with the original set up but oil temp wouldnt have. Attached is your MAP with some custom CAN added to send the GP temp in place of oil temp. The problem with oil press must be at the dash or CAN adapter end. 2zzmap-new CAN.pclr

Did your car have an external ignitor before you changed to COP or not? A V6 shouldn't, but since it already has an engine swap it could be anything.

File>compare.

Start with the idle settings changes in green below. The base position table looks like it probably needs work, so log a cold start and allow it to completely warm up then use the logged TPS to aid filling out the base position table. For the accel fuel try these changes:

The force required to move the pintle is relative to the pressure differential across its seat - i.e. how much pressure is pushing it closed vs pressure pushing it open. You have rail pressure on the top, manifold pressure on the bottom. This is typically called differential pressure or relative pressure, Diff FP = FP - MGP. Since MGP = 0 when the engine isnt running, "base pressure" or "static pressure" is the same as differential pressure when the engine isnt running. Coincidentally, the pressure differential across the injector is also important to calculate PW from mass flow, so monitoring it improves the accuracy of both deadtimes and the fuel calculation.

There are quite a few pretty odd settings in your map so I have made some adjustments to get it back to a more typical starting point. Can you try loading this map and doing another log. It may not be immediately better since a lot of my settings are just guesses, but provided it is close enough to idle we will likely get it better after the next log. Is your IAT sensor reading correct? 70degC at idle seems a bit extreme for a NA ITB engine. You can use use the file compare function to see what I changed. 5222023 idle tweaks.pclx

I dont, we do have one on our dyno engine inhouse but that has a G5 ecu so the E-throttle parameters wont be useful to you. What issue do you have?

You are not "jumping to 8000", this is the RPM that the engine will be held at on the start line and for maybe the first 2 wheel rotations, then you quickly reduce that down to the slip value that gives you the best acceleration. Once you have the slip under control and the vehicle accelerating, then you ramp the RPM back up to your normal RPM limit.

Yes I agree with Kris' suggestion as probably the best workaround, choose some nominal ratio to reduce the Inj flow rate by and adjust the engine capacity by the same ratio. As far as I can think most of the fuel calculation and compensations would still work correctly, there would only be a couple of metrics like instantaneous fuel consumption and estimated air mass that wont report correct values.

The virtual here USB server that is suggested in the 2nd post of this thread works with G4X. Pretty slow though, not very pleasant for tuning but ok for a display. Otherwise if you just want a display you can use realdash with a Bluetooth or wifi CAN adapter.

This is just a random made up example, but it is a quite normal trend for road racing, I have no idea how things work in snow. But the general idea in cars is there is an ideal amount of slip that will give best acceleration (not zero slip). It is much easier to start off with too much slip, then quickly pull the RPM down to achieve the ideal slip, than it is to start off with too little slip, and then somehow try to invoke an increase in slip.

You have no accel enrichment as the cold correction table is zero'd out, the smallest value would typically be 1.0 in there at normal operating temp and 2.0-7.0 at the cold end depending on injectors etc. For the boost cut you likely have a short spike that gets too close to the limit, you will see it better if you ecu log at say 200Hz, but since you are hitting the cut you will either have to lower boost, raise the limit, or reduce the control range.

It looks like you have about 10 times more accel fuel than normal so reduce that a lot. I dont see any signs of a idle issue in that log.