Adamw

That sounds like you must be using the CANDASH cable instead of the CANF as shown above. The CANDASH cable connects the two serial wires so that will upset PC Comms. A quick fix would be to open up the ECU case again and either snip or depin the yellow and grey wires from the white plug on the CANPCB cable.

There is no "Speed Input" channel in the Link stream. Use whichever one of the 4 wheelspeeds you have set up in the ecu.

As I said earlier you can use the test calculator to see how any CAN input or output works. The problem with the AEM device is when it goes into error condition it still outputs a lambda of 1.00 which is considered a valid lambda so the ECU will not disable CLL etc. I guess at least it will correct richer rather than leaner. The Link Lambda and most other 3rd party devices I know of all output Lambda = 0.0 when in error so the ECU knows its invalid and will either remove it from the lambda average calculation or disable CLL if there is only a single sensor.

We have added a special trap for young players. Turn off the disable input and you will be winning. The "Always ON" means it will be always disabled. There has already been some discussion about how to make this more obvious.

I would say that is more likely not enough pre-crank prime. Everytime you key-on you will get another dose of pre-crank prime. You have a solenoid valve so the response is instant. As soon as the ecu sees the crank shaft start to move it will turn on the idle valve and send it to whatever value is in your base position table plus the extra "start up step" added. The start up step is faded out over 3 seconds. So it sounds like you need to adjust one of these.

Ok, let me ask someone smarter than me and I will report back.

I dont think so. JMP's suggestion is a good one though, well worth a try.

Hmm, thats weird, I havent come across this before. I see a reference to "McAfee" in your device manager, Is there an anti-virus software or something preventing the driver from installing? Can you temporarily disable? I will attach an old and new version of the driver if you want to give both a try. Link USB Drivers.zip New G4+ drivers.zip

I guess I would tune it a bit further and see how VE looks at higher RPM/Loads. Possibly just a short PW issue?

Use the X-Series template that is included in the software. You will not be able to use the status or error data as AEM enumerations for status and errors are different to ours. You could input them as CAN AN Volt channels as a diagnostic if you really want to but it is unlikely useful. Link use a resolution 0.001 for lambda so there is already a native divider of 1000 built in to the software. AEM use 0.0001 resolution, hence divide the input by 10. You can use the test calculator to see how it works. ID should be 384 as per the help file article for the X-series. AEM use Hex, Link ID's are in decimal. 180 hex is 384 decimal.

You are going to have to get an oscilloscope on it to find out why the ecu is not happy about the signal it is receiving.

Yes, you either have it set to always armed, or you have left your arming switch on. The Log clearly shows launch control status as always on and the launch limit activates about 5200RPM.

Can you show us the hardware IDs like my pic below.

I think you are confusing gauge pressure with Absolute pressure. The acronym MAP stands for Manifold Absolute Pressure. A boost gauge as the name suggests reads in "Gauge pressure". If you look at the parameter "MGP" (manifold gauge pressure) in the ecu it should match your gauge resonably well. Most boost gauges are very poor accuracy and have lots of damping so it wont be perfect but it will be similar. In your log you are idling with a MGP of about -42kpa which is about -12.5inHg so that seems close to what your boost gauge reports. 38psi MAP - 14.5psi atmospheric pressure is about 23.5psi MGP, so again that is close to what your gauge displays (I would trust the ECU over the gauge). Your limiter coming in at about 5200RPM is the launch control limiter.

If your switch is connected to ground then the pull-up should be on, if the switch is connected to +12V then the pull-up should be off. Active edge will have no effect for a switch. On level just inverts which switch position represents on or off. If the switch works backwards then change the on level to the oposite.

It is switching off due to the evap temp going below your evap temp lockout of 4°C. I believe the purpose of this lockout is to prevent the evaporator from icing up. You can try lowering that temp, or setting it to -10 will disable it.

Im not sure if you understood that comment correctly. There are 2 fans in the car, one controlled by Ign 3(Fan 1), one controlled by ign 4 (fan 3). In our WRX4 base map, Fan 1 active state is set to low, but fan 3 state is set to high. One of these (or even both) may not be correct so you may need to experiment with both. So you have these combinations to try: Fan1 low, Fan 3 high (current setting). Fan 1 low, Fan 3 low. (I suspect this may be the correct one) Fan 1 high, Fan 3 low. Fan 1 high, Fan 3 high.

It needs to be mostly tuned first. Generally idle and cold start are the last things you tune.

They are not PWM, you will damage something if you PMW it so dont. It should only be the active state that you need to change. Did you try changing the active state on both fan 3 & fan 1?

You have quite an old ECU, the V1.5 bottom board is approx 5 years old. E-throttle was only introduced in Bottom board revision V1.7. The actual ECU (the Top board with all the electronics) has the E-throttle controller built-in and enabled, but the outputs are not connected to anything on the bottom board. You can return to Link for a "bottom board swap" if you want it updated to the latest V.7. Cost to do this is about NZD$290 last I heard.

I would have ignition target lower than that, typically around 10. At 25 you are already sitting near MBT so there is less ability for the ECU to quickly increase torque when it is needed. You may find the MAP lockout will need to be a bit higher so that it works reliably during coldstart - many engines when cold will sit above 50kpa after start for a while. An actuator deadband around 40 is usually better for E-throttle to stop it dithering.

An ECU will have little affect on how much power you can get from an engine. It is mostly depend on how well optimised the factory tune was compared to how well it is optimised when it has a user tunable ECU. If it is a naturally aspirated 2J with a distributor then there is not much you can optimise so you are unlikely to improve it much at all.

I dont see much wrong with the set-up. As above, turn off IAT trim and the 4D table as a start but I dont think these will solve the problem. If it were mine the next step I would take is a basic flow test on an injector to confirm it is actually 1000cc. You can get a pretty good idea just by connecting an injector direct to a battery and holding it open for say 30seconds and measure how much fuel you get. You can usually just hose clamp them in a piece of hose, dont need the rail etc. And I would try to confirm the fuel pressure with a gauge or similar - just to confirm the sensor is reading correctly. I have had mislabeled sensors before.

If your cruise switch is the type that varies resistance depending on which button you press, then yes you will need a pull-up to 5V. Usually 1Kohm works ok.

To test wiring you could quickly short the wire on ecu pin 50 to ground (using a small jumper wire or paper clip etc), just a quick (few milliseconds) touch to ground, the coil should charge while that pin is grounded, and it will spark as soon as you release the ground. If you get a spark when testing like that then that would prove the coil and wiring is ok and the problem is likely the ignitor.