Adamw

Picture doesnt show, but an initialization failure message would suggest you have a problem with your code or wiring between your micro and CAN controller. This basically means the micro didnt get the "all set up ok and now ready to go" message from the CAN controller. The ECU will show errors because it is not getting an acknowledge from your CAN device.

There is no other setting in the ECU that can affect it. You might have to get an oscilloscope on to the speed wire to see what is wrong with the signal. Im not sure about subaru's, but in Nissans and toyotas from around that same era it is common for the reed switch that generates the signal in the back of the speedo to get sticky/erratic.

Did you perform a store before shutting off? Scope looks good.

You would receive these as CAN Dig inputs. You can then use them to turn on/off functions or in virtual auxes etc. Not really possible. There are some native inputs that can be assigned as CAN inputs such as oil an fuel pressure, but the list is fairly short. There are 8 generic "CAN analog inputs" but these can be re-named and cant have units assigned - they will always be displayed as "volts". You can still use the generic CAN analog inputs for axes on tables or switching things on/off etc but they are a bit less useful when without labels and units. As above, you would receive the input as a CAN DI, then you can do whatever you like with that - swap boost/fuel/ign tables, change RPM or boost limits, turn on WI, flash a light etc.

What size injectors? I notice in the log it looks like it gives a small cough of life initially then very little - does that describe a typical start? What does it have for a coil/ignition system? To answer your question about the min crank speed, it is very low, somewhere around 25RPM. A couple of things I noticed; The pre-crank prime table has very little in it. For 1000cc injectors, typical numbers would be around 30ms below zero DegC tapering to about 4ms @ 70deg. If your injectors are smaller than that then scale these numbers up accordingly. The warm-up enrichment table has smaller numbers in it than I typically see in traditional fuel mode - but it looks like someone may have already tuned it? Once started is the AFR near target during warmup?

In that case it sounds like it is tune related then. If you can post a log of a start up with the adaptronic I can compare injector PW etc during cranking and maybe suggest some further changes to your tune. Otherwise it really needs to go back to the tuner, get him to first fix the staged injection and confirm the trigger offset as I mentioned above, then mess around with the pre-crank prime (currently turned off in your tune) and the cranking enrichment to see if he can improve start up with those. A couple of other things I just noticed; There is only about 5 deg trailing split in the cranking region, open that up to about 15 may give it a better chance of catching. Injector timing is set to start 360BTDC which is very early for a rotary, something like 90-150BTDC is more normal. With the bridge port overlap I suspect you have a lot of fuel spraying straight out the exhaust. Also, although there is no sign of trigger error in your original log, your trigger arming threshold is a bit unusual. I would suggest you change the 1000RPM cell in both the trig 1 & 2 arming threshold tables to 1.0V as below.

I had a quick look through your tune. Nothing really jumping out to me that would cause a hard start but one thing I noticed in your tune that I dont really like; The injector staging setup is pretty weird with the lockout at only 500RPM and sec injectors fully in by 40Kpa, this means the secondary injectors are on the whole time (maybe just above idle depending on PW). Normally you would bring them in say above 3000RPM and maybe 80 or 100KPa.

As for your wiring, pin 30 will need to be connected to a relay that switches on with ignition. With it connected direct to battery as you propose when the ignition is off you will get a back feed through the ECU. As for settings, generally you would set up a GP PWM, set some conditions so it only turns on above a certain temperature, then in the PWM table you would have ECT on one axis and possibly RPM or boost on the other to vary the speed of the pump with temp and load. Frequency needs to be determined by trial and error, it depends on the motor. Usualy 100-500Hz will do for a water pump, the motor will sound most happy when you have it right.

So you dont want to use an android head unit that already has a tuner built in? The problem with many tablets is they only have a single USB port so they cant charge and have something connected to it at the same time. Even with a hub you usually cant do both, Im not sure if this has changed with USB-C tablets. As for radio on a tablet you will need a USB FM tuner/receiver.

In the old G4+ "1 tooth/TDC" mode the offset was locked to zero. So I would do as you suggest - start with offset at zero and set timing by rotating the distributor. This would usually result in your rotor phasing being correct assuming it is an unmodified OEM distributor.

Yes, if it was connected to a reluctor sensor originally it will be expecting a sine wave that crosses zero volts. You can try the de-coupling circuit below, no guarantees for you but it will generate a simulated AC signal that crosses zero. It has worked for a nissan skyline speedo which is uses a reluctor sensor.

In the help file: Wiring Information > Output Wiring > E-Throttle > Specific E-Throttle Applications > Ford BA Falcon Electronic Throttle Wiring

Yes, the rpm spike is due to a trigger error. This means the trigger pattern received by the processor didn’t match what it expected to see. Most likely a tooth didn’t meet the arming threshold so it was ignored.

Nothing at all. Just found "Day 2 triggerscope 1" is a valid capture but neither of the other two show anything - just 0.1V of noise. Both cam and crank are only just making 0.3V in that capture though so that is very low. Its almost like there is no ground connected to the sensors or something. Is it a piggy back install? Try your arming thresholds something like this: No, the triggerscope can only capture about 720deg.

None of the trigger scopes you have attached have any data but based on those screenshots above I would say the arming thresholds need to be much lower. Set the whole table to about half it’s current values on both trig 1& 2 The VVT trigger modes don’t have the same restrictions in regards to cam tooth position. Yours looks normal for 1J/2J.

If its been sitting for a while the first thing I would do is a compression test with a proper rotary comp tester. Both the side seals and apex seals are pretty prone to sticking if they have been sitting for more than a couple of months. Nearly always hard starting of a rotary is a compression issue. There is no indication of a trigger issue in your log but it is only a couple of seconds of cranking so I wouldnt say that is conclusive. They also wont start easy if the plugs are fouled and it doesnt take much to foul them when the tune isnt spot on. What size injectors does it have? Is the AFR in the ball park when it is running? I notice also the trigger offset hasnt changed from the base map setting, this possibly means the tuner hasnt checked the base timing so make sure he has done that too.

Yes, and you should add a 120ohm 1/4W resistor across the CANH/L wires somewhere near the AEM. The other condition you could use instead of RPM is something like Engine running time > 30s, that may help get a little more of the condensation out of the way before powering up.

Start by doing a triggerscope and attaching here. Do one at idle and one at say 4000RPM might be helpful. Heres a quick vid showing how to do it: https://1drv.ms/v/s!AiYbYlZQuRHPmidU5V2CmTcv6t2y?e=i2nfuZ Also, your VVT control is very bad (this is just something I noticed - it is not part of the trigger issue), go to VVT settings, change PID setup to custom then go to the new LH Inlet folder that will show up and half the proportional gain.

Most of your map looks ball park to me. Have you checked base timing is close with a timing light? Provided timing is close then my first suggestion would be "not enough fuel" contrary to your comments. Your master fuel is set to 6ms currently, I would have expected somewhere more around 10ms for 800cc injectors, so I would try bumping that up to see if it improves or not.

Your settings look correct. The 24 tooth "crank" signal needs to be connected to ECU pin C12 and the single tooth cam sensor needs to be connected to pin C14.

You are correct, spark duration and arc duration are the same thing. In your case this setting may only just start to come into play a high rpm. The spark duration setting usually only comes in to play when you have a distributor with single coil and more than 4 cylinders. It is used to shorten the dwell at high RPM when there is not enough time to both charge the coil and spark for the full duration. As an example think of a V8 with a distributor running at 6000RPM. At 6000RPM one engine cycle (two crank revs) takes 20ms. All 8 cylinders have to fire in that time. Lets say your coil needs 2ms of dwell to charge and the spark lasts for 1.0ms. You have to do that 8 times in those 2 revolutions, 8 x 3ms = 24ms - but wait - you only have 20ms to do that in? The ECU cant do anything about the spark so if it knows how long the spark needs and there is not enough time then it can shorten the dwell to make everything fit into the time that is available. A rotary in direct spark mode at 8000RPM will have 7.5ms of cycle time to charge and spark. So if you have a spark duration of 2.9ms that leaves you with about 4.6ms for dwell. You generally dont need much more than 3ms on those coils unless very high boost or Methanol. With it set at 2.0ms now is not a big problem because as you can see it is quite likely you are not running into cycle time issues anyway.

Geez you dont see 2TG's much anymore... You will have to grind 3 teeth off that to just leave a single tooth in the dizzy.

Scope looks good and polarity is correct. The map looks like a good starting point and I dont see anything unusual.

Just set the trigger mode to "Ford BA XR6 VTC". The Thunder and Fury base maps are from Links barra test car so will be a reasonable starting point for many settings. Be careful with that, hall sensors can be pretty fussy with tooth design and many reluctor wheel designs arent ideal for hall sensors. I have never seen a barra one so I dont know how relevant this comment is but make sure you keep an eye on it initially once it is running.

This is the one I had in mind: https://www.hpacademy.com/previous-webinars/049-setting-up-knock-control-link-g4-vipec/