Adamw

After that first run since there was knock detected at ~5500-6500, the knock trim table would have looked something like this: So the next time the engine passes through 5500-6500 this same trim would be applied. These values in this table will stay there until the engine has been operating in the same cell again with no knock detected, then it will start removing the retard. These tables are cleared back to zero also everytime the ecu is powered up. Whereas if the knock trim table had MAP on the Y axis, then only something like these 4 cells would be trimmed (below). This means you would only get retard in low load areas where it wouldnt affect performance. But still it would be better to fix the threshold so that it is not falsely detecting knock in the first place...

All of the Alfa TS engines fire both plugs at the same time - even the later 16v ones with the one central plug and the smaller one off to the side. Firing twin plugs with different timing is pretty rare due to the higher susceptibility of the colliding frame fronts to cause knock. It is usually only very large bore (95mm+) twin spark engines that delay one plug a little - typically only 1 or 2 degs.

You can see after you lifted off at the end of the first pull that your knock level went above the threshold from about 6250 to 5700RPM. This has resulted in a knock trim being applied between these RPM's. The knock trim gets written into the knock trim table so that when the engine next passes through this same RPM band it gets the same knock trim applied (whether it is knocking or not). Only after the engine has been operating in that RPM band for some time (Ign advance delay) without knock, then the ecu will start slowly removing the knock trim (ign advance rate). So a couple of things you could do to fix: Increase threshold in the high RPM/low MAP cells that it passes through after you lift off. I have highlighted the cells in the table below that are causing the problem. Add MAP or TP to the Y axis of the knock trim table so that the trim doesnt get applied to the same RPM regardless of load.

The 1K pull-up is preferable over the 4.6K and will give slightly better resolution at normal to maximum temps. The 4.6K would give slightly better resolution at very low temps, but most engine people arent concerned about low temp. Set calibration to "Std Bosch NTC".

The base map will give you the basic fuel and ignition settings. You will have to manually assign all the correct inputs and outputs to match your wiring. ie you need to set which aux output the fuel pump and fan are connected to, which inputs the TPS, MAP & temp sensors are connected to and set there correct calibrations etc.

You need to set Lambda 1 sensor control to "Link CAN"

Update your firmware, yours is very old so I suspect it doesnt have the AEM calibration included that you are trying to use. Logging is the same as G4, hit F8 to start and stop logging or click the words "PC Log = Off" at the top right of the screen. K key swaps between text and surface view on any table. See this video if you want to set it up with both text and graphic view side by side: https://youtu.be/Eo9zBcIkacs Also, since you are new to the software I would suggest you go to >layout>load layout and choose the 1366 x 768 or the 1920 x 1080 layout depending whether you have an HD screen or not. These have many of the common pages already set up.

If DC is at 100% and boost is not increasing then it is either not capable of bleeding enough air or the exhaust manifold pressure is considerably higher than the spring pressure. You can confirm which by temporarily removing the hose from the wastegate and carefully opening throttle to see if higher boost can be achieved with no boost reference. Make sure boost cut is set to something safe when doing this. If you can get high boost this way then it is just the valve not flowing enough bleed, you may be able to fit a restrictor on the supply side.

I think you will be in bootloader mode if you dont have pin 4 grounded. From the manual:

Attach the tune and a log

I dont think there is much risk of charge robbing on a WRX10, but your EGTs are pretty warm so I would guess possibly the extra oxygen going through the dead cylinder can find its way into the adjacent cylinder exhaust manifold and allow a little extra of the combusted fuel in the manifold of the still working cylinder to burn.

If you want to run two gauges then you can have them both connected to the same bus, they just need to be on different ID's. The AEM instructions will tell you how to change the gauge ID from the default. Then in PC Link you will set up separate receive streams for Lambda 1 and Lambda 2 on ID's to match each gauge.

If the engine is stock internally I would be cautious of using sustained cuts. If it has good seals in it then Ign cut is the best option - if you get heavy backfires you can also use 4D fuel at the same time to add a little fuel which softens down any backfires.

Try loading the map attached below, this is the 7 bar base map but I have made a few small fixes. You will just need to enter the correct injector size and it shouldnt be too far off. To load the map, connect the ecu to PC Link then go to >file>open and choose this map. Toyota Altezza SXE10 7Bar G4X plugin Modelled fuel Adam.pclx

If tech support told you to return the ecu for inspection then I would take that advice. There is a small batch of G4+ Fury boards that gave a high failure rate, if they have identified yours is from that batch then it will need to be inspected and possibly replaced.

Why arent you using our Altezza base map? There are many things wrong in your map: E-throttle target table is full of zeros (you are telling the ecu you always want 0% throttle for any pedal position). E-throttle is set to quiet throttle mode (wont work unless engine is running) There is no aux output assigned to the E-throttle clutch (should be aux 5) Fuel equation is in traditional mode - modelled is needed for fuel gauge to work properly in the altezza. Fix the first 3 and E-throttle will work.

Need a log.

You will need to remove the lock and do a new log of some pedal movements with the E-throttle mode set to on.

We want the complete tune file from the ECU. When connected go to >file>save as. It will be a .pclx file E-throttle mode needs to be set to "ON" for it to work.

Yeah that would work fine.

It would be best to attach a log from your car showing the issue along with your tune file.

You might have to elaborate on your question a little more as I dont understand what you are asking. Typically the injector flow rate is supplied in the data provided by the injector supplier. Usually it is determined dynamically using the linear slope of a graph of flow Vs pulsewidth at a specific frequency. Static "wide open" flow data is not as useful. Most flow test procedures are based around the SAE standard J1832

Also 0280140549 is the two wire straight through version. Note "fully closed" for these bosch valves is usually around 20%DC, below that they start to open again as a fail safe. So for these you set the minimum clamp to ~20% and you wont have numbers in your base pos table below 20.

What do you mean by "fully sealed at 100%"? 100% DC is usually fully open. The common 2 or 3 wire bosch ones are what I normally go for, hose barb fittings so easy to connect and come in straight or 90deg versions.

The IC spray function is not set up in our base map as the set up is different depending on if the ecu is fitted to an Evo 4/5/6/7/8. If you write how you want the switch to work I can help you set it up.