Adamw

What specific coil are we talking about?

Are you sure it is wired to CAN1? I notice you have the AEM device set up on both CAN1 and CAN2 in the ecu.

The Atom doesnt do VVT control so you would need a Monsoon as a minimum.

A high error on two temp sensors occurring at the same time would suggest a bad ground. Either a loose ground connection to the sensors somewhere or something like a bad engine ground so starter current is passing through ECU ground when cranking etc. Set up ECU logging to run all the time, recording the raw voltages on the temp pins as well as the bat voltage, temps, fault codes etc, then you will have a log saved in memory that you can download next time it happens.

Have you tried using the same settings/profile as the gt86 with the link? Can you show us some pics of the wiring.

The CLL functionality only allows either narrow band or wideband control, not both at the same time. The NB's would not be useful when there is positive pressure in the ex manifold anyhow. So if you want to do individual cyl control with sensors preturbo you will want pressure compensated widebands such as Link CAN lambdas.

Lol, that's a little more than I was assuming. That's getting up there for sure, you might be pushing your luck even with R35 coils at more than 300Hp/cyl. Although im quite surprised the K series coil actually made it to that power level already. Are you running a 16V system? I assume now we are talking methanol, a very fat mixture and 50psi+. I would be leaning more towards an IGN1A for an application like that, as it would give you a bit more headroom for when things arent perfect - ie mixture goes a bit too rich or battery voltage drops etc. I wouldnt put complete trust in the accuracy of the test results from PRP below but the general trend shown seems to fit with what I observe in everyday use. I would put the K20 denso somewhere between around number 8-12 in this list.

Bit rate should be 500K. I just noticed you had OBD enabled on CAN 1 so turn that off and try again.

Can you zero the CLL trim in the overrun/idle entry area to see if that helps the near stall.

You need to hit enter after typing in the value, you should see it turn blue to show its been updated.

The fuel press scaling was fixed several years ago. On the ecu stream tab in rs3, check you have v2.00.07. If not, click on change ecu, then just select the same CAN BUS BASE LCC ecu and it will reload the latest

Depends what you are comparing them to I guess. But they are probably one of the more powerful OEM COP coils that I have come across. Make sure you get genuine Hitachi ones or some other known reputable equivalent. Avoid Amazon/ebay etc.

Ok, attached is a copy of your map with CAN setup to hopefully duplicate the GJP IO device so you shouldnt need to mess with the existing Racechrono setup too much. There are a few mistakes in that GJP documentation so let me know if something doesnt look right. I didnt set up steering angle or brake press since you dont have these sensors. In Racechrono you will just have to add the lambda channel since the GJP device didnt have lambda. Lambda channel details: PID 1803 Equation: bytesToUint(raw, 3, 2) * 0.001 Metz-e36 GJP CAN.pclx

That would be impossible, are you sure it is not inHg? I would confirm fuel pressure as the first step, most of the basics in the log look ok. Lambda doesnt mean a lot when there is a misfire so we cant put much weight on that. I would also go to your GP output that is controlling the VTC cam and with the engine idling put the aux in test mode to confirm the VTC is actually working too. Probably not relevant to your current problem, but you log shows there is no acceleration enrichment working, so that probably needs to go back to the tuner to sort. With the engine off, does MAP read the same as BAP?

Spark edge is falling. Try this for dwell: https://www.platinumracingproducts.com/pages/platinum-racing-products-r35-dwell-settings

It looks like you have a trigger issue. Please do a trigger scope.

Your ECU has 2 CAN ports, you can use either one. It is fine to connect multiple devices all to the same bus, this is exactly what CAN is designed for.

First of all, do you have the vanos working so it follows the commanded target?

I think this is using raw CAN, not OBD2 PID's so PID is really the wrong term that is confusing the question. Message ID or CAN ID would be more appropriate. Maybe their app was originally only designed for OBD2 commands. What ecu CAN port is the RC device wired to?

The fuel pump can be connected to different pins depending on region. Try changing the fuel pump jumper to the alternative option.

The CANopen and J1939 are just the standard keypads, not the motorsport version. You will likely have to contact them directly.

It doesnt have the driver onboard, you would need the external e-throttle module.

Our test car which the maps came from is a JDM sti which use the better 2.0 engine. The maps use modelled fuel equation so just change the engine capacity and injector size and it will run. Some of the 2.5L also have a different MAP sensor calibration so confirm that reads correct and you shouldn't need much work to make it drivable.

It would not be able to control the exhaust cam without a position sensor. Typically on a NA engine the exhaust cam timing has very little effect on performance so is probably not the end of the world.