Confused

Double check that you don't have another input set up as TP (Main) - looking at the pinouts, by default AN3 is TPS, if this is still set as TP (Main), it will be picking this input up, not the one you've defined later. If it's not that, you don't *have* to do the calibration this way - you can manually move the throttle plate, and enter into the setup the min & max voltages you see.

If you buy a Plug-in ECU specifically for your Subaru, it won't have an on board wideband controller. You'll want a Link CAN Lambda to go with it.

The scan tool, unless it cost you hundreds or thousands, is probably only trying to connect to the engine ECU - which you've now removed. You'll need a professional scan tool which can connect to multiple systems, especially the SRS system, which in many cases will not reset by simply removing power for a few minutes. For my '98 Mitsubishi Legnum VR-4, I found that the following scan tools worked for resetting SRS issues caused by things like a broken clock spring, removing steering wheel/seats (then turning ignition on) etc, or another issue (once fixed): Mitsubishi MUT-II and MUT-III Snap-On Solus Bluepoint G-Scan Launch X431 Pro 3 But the following didn't: Launch EasyDiag 2.0 (with XPro software) So it's definitely possible that you're just not using the right tool for the job.

Try the same test, but remove one additional relay at a time from the board, until you find out which one stops it from happening. Then, add the others back one at a time. That should narrow it down to one smaller circuit.

You're going to have to help us help you, so please upload a copy of your config, plus some PC log files of every parameter. There are many people successfully running E-Throttle on Link ECUs, so it's most likely a configuration issue, but you've not yet given any details that will help anyone troubleshoot it.

Yes, that's a great solution for the G4X range, but doesn't help with the G4+. The G4+ has some great Knock control, which is highly configurable via the Knock Target Table, but no easy way to inform the user that it's occurring unless you've got a laptop hooked up or looking at a log file after the fact. I think that having a "Knock > Target" parameter on an Aux Out would satisfy this nicely on the G4+.

Bumping this one up, I'd really like the ability to trigger an Aux/Virtual Aux on Knock > Knock Target on the G4+. Hopefully that's a relatively simple inclusion, as I guess all of these values are in memory for the knock trim calculations anyway?

I'd like the option to automatically save Logs to a specified folder, also. The other day I had the laptop hooked up to the car, then when I plugged it into my USB-C hub to charge (which also has a display attached), PC Link G4+ crashed, and I lost the log. If it had been automatically saved to disk, this would have prevented me losing the log.

Yes, for Closed Loop boost control to work properly, you need to first set up Open Loop boost control as close to perfect as you can. It uses the settings you perfect in Open Loop as a baseline of where to start when you're in Closed Loop - the better this baseline is, the better your Closed Loop can respond, and it will be making minor adjustments, not massive ones. The Help File really is VERY good at explaining almost every aspect of the ECU, and how it applies, if you've not done so already, you should be using it as your bedtime reading!

I think you'd want to activate the fans from the ECU, rather than your separate fan switch - that way you've got full control over when they activate, and can log when they're on and off. You can wire them up individually as two separate fans, and use the Engine Fan 1 and Engine Fan 2 parameters to turn them on based on ECT, or just use a single fan output from the ECU and run them both together. It depends on how many spare outputs you've got. You'll want to add a relay, as the Link's outputs won't drive a fan directly! Feed one side of the coil with an ignition switched feed, and wire the other side of the coil to the output on the Link.

I finally got round to actually trying to run the engine this week with the R8 coils after fitting them back in March - after setting the new dwell times (from another post of yours @Adamw with settings for ~1.4 bar), it fired up and ran beautifully! I'd still like to look at modifying the base board to remove the link between ECU Hold Power from DI3/AI4 so I could go sequential, but it'll do for now!!

@Gsab I've been visiting this forum every day for a couple of years now, and have been reading every new thread that's posted, and I agree - you can learn a LOT about how the ECU works just by doing so, and also HUGE respect to @Adamw for his professionalism and patience (even when he didn't work for Link for a while - I think he's back officially now?) - especially when, as you said, most people do not come back to say thanks! It reminds me a little of this: XKCD: Wisdom of the Ancients One thing I don't like about Haltech is that they discourage you from using the forum, and prefer you to call them. That means as a new (or potential) user, you don't have the ability to look back over issues people have had and see their resolution, so it's likely you'll make similar mistakes to others.

Thank you, I'll give that a try!

Did you manage to have a check of your calibration?

It needs to be positioned a couple of mm away from the surface it is detecting. So, looks like you'll need to make up a bracket.

Yes, I want to start using traction control, so that needs at least 2 speed inputs, so 2 Digital Inputs. It is waterproof, been exposed to a lot of typical English weather, so lots of rain!

Here's a snip from one of my log files which shows some low speed. The yellow line is from my 4 pulse per revolution, the purple from my prop-diff pickup. As you'll see, as soon as the rear detected movement, so did the front. By 15mph they were both practically identical, despite the rear having nearly 4x the resolution. Whether that's enough for you, I can't say. I decided I wanted more pulses per revolution, so have made modifications to do so. You probably could attach this sensor to the thin metal backing plate, it's fully threaded so you can adjust the depth accordingly.

Potentially, yes. I've got a friend who is modifying a non-ABS Mitsubishi FTO hub to take that sensor and pick up on an ABS ring. But if you've got an ABS hub, have you got the original ABS sensor? You just need any rotating part that is either ferrous (a magnet will stick to it) or that you can stick a magnet to. For example, you could stick magnets onto your CV joint casings.

If the ECU is showing what the AEM gauge is transmitting, either your AEM is correct (and configured correctly within the ECU), or the dyno is correct. To know which is correct, you'll need at least one extra sensor to compare to.

It's 3 wire - 12v power, ground, and signal. This goes into a Digital Input on the ECU. Calibrate it (as per the help file), and have a nice accurate speed. The bracket took me about 15 minutes to fabricate.

I've used this Hall effect sensor to pick up on the heads of the wheel studs on the back of the hub on my non-driven wheel, as it was quite an open design, and they were easily accessible. I made up a small aluminium bracket to hold it in the right place. This gave me 4 pulses per revolution. I did, however, notice that, as my driven wheel speed (picked up from the driveshaft-diff flange) had 3.7x the number of pulses per revolution, the "slip percentage" for traction control was growing with each pulse of the driven wheel, before being "reset" when the non-driven wheel speed pulse was received. To combat this, I've added some extra bolt heads to the back of the hubs, to give me 3x more pulses per revolution, which will recalculate the speed (and therefore slip percentage) more regularly.

This issue is still present.

You will need to change your relay's coil feed to be ignition switched, rather than permanently powered. This way, the ECU will still pull to ground to activate, but when ignition is off, the coil cannot be energised, even with the ECU defaulting to grounding the outputs.

Thank you for fixing the Cruise Light functionality, I'm hoping that should help with being able to change to a different E-Throttle Target Table to minimise speed overshoot when accelerating. Plus I can have a light on the dash, too!

I've got an Evo 1-3 ECU in my Anglia running a 6A13TT, so I replaced the connectors when I rewired the car. From RS Components the part numbers you'll want are: 12 way - 720-1017 16 way - 712-2163 22 way - 712-2485 26 way - 712-2497 Pin inserts for these are: Small - 712-1909 Large - 712-1911 The Evo 1-3 only uses 3 of the 4 connectors (as does the Legnum/Galant VR-4, but the later Evo 4-8 uses all 4 sizes. If you wanted to make some kind of patch harness, then you might want 720-1014 (or 720-0979, can't tell the differences!) as a 4 way female, where you could present the 2 sets of expansion connectors on the plug-in board out to the extra plug, which might make the loom nicer! Or 720-0976 as a 3 way female if you want a straight replacement.