Adamw

I suspect the lambda oscillation may be due to CLL update rate too high. I suggest you turn that and the long term trim off while tuning. Your Lambda 2 cal is also not a straight line so something is set up wrong there. I suggest turning on async injection - set min async PW to 0.5ms. You accel fuel is not working since the cold correct table is all zeros. Populate that with all 1.0 for now and start with numbers about 3 right across your accel sense table. If lambda is still unstable when inj PW is stable with those changes then the next step would be to fit a fuel pressure sensor, it could potentially be an intermittent misfire on 1 or 2 cylinders also.

Either will work. Realdash takes a bit more to set up, will need CAN wiring, the correct USB CAN adapter, is more difficult to set up, but is quite a bit more flexible, it can do laptiming, music, navigation, etc, as well as for example send button presses to the ecu to activate high boost or whatever.

I dont see any signs of interference when cranking in that log. Probably more likely a volt drop issue or possibly arming threshold. What do you have for an ignitor? Is the battery close to the starter motor or in the boot/trunk? Can you do a triggerscope when cranking.

I have helped a couple of guys with CAN to HTG. The last one I helped with HTG had just recently changed their software around a lot so it didnt match their documentation and the CAN set up was very buggy, im not sure if even HTG could get it to work. If you attach your ecu and GCU configs I will set up some that I can, but be aware you may need some help for HTG.

The RPM looks pretty normal for a cam mounted trigger. This is valve train resonance bouncing the CAS backwards and forwards. Your wideband is not working at all in either log.

G4X can have 10 ID's on each bus. 100 receive or transmit channels max in user streams, preconfigured streams such as "MXS dash" arent counted in the 100 channel limit. In your case most likely the gauge and dash would both use the same stream so only 1 ID needed for those. CAN keypads arent supported directly yet, although it is a feature we will add in the future.

Do you have a G4+ or G4X? This mod was only needed for G4+ which we haven’t made for over a year

E-throttle relay can be done via CAN. In G4+ you assign the function to a virtual aux and send the VA status. In G4X you assign a CAN aux.

The white wire in the trig1/2 shielded cables is ground. You can use either one. Snip the other off or just fold it back on itself and shrink over it.

The "DisplayLink" stream is identical to the "Transmit Generic dash" stream. Notes from the help file:

Assuming thats an i series vipec then yes it is compatable with CAN Lambda. You will need to be running the latest Link G4+ firmware. Connection would be easiest with a CANJST: https://dealers.linkecu.com/ADA_2 and possibly a CANEXT if needed: https://dealers.linkecu.com/ADA_7

Unlikely the problem I think. Please do a PC log of the engine cranking, we will be able to see if there is any interference while cranking with that. Can you give some info about your ECU and car also.

Your cam is not "locked retarded". Your cam teeth are approx 80 degs further retarded than it should physically be able to move. The problem is the cam edges (pink lines), are sitting just about right on top of the sync edge (yellow lines) and move across the sync edge whenever the cam moves a little. If you compare to the known good pattern, the cam edges will never cross the sync edges:

Can you load this map in and do another short log - say 1 minute of running. I have added some custom CAN to help with diagnosis. Lambda test.pclx

The coolant temp gauge will usually have its own dedicated sensor - the ecu sensor will be dedicated to the ECU. The gauge will usually have a single wire sensor and will use a 12V pull-up where as the ecu will have a 2 wire sensor and only uses 5V pull-up.

It is not something I have ever wanted to do, but the maps I see with it are usually done something like below. Ign timing bounces from +30 to -40 either side of target. In G4+ you would use the Idle ign table, in G4X you would turn off idle ign and set up a 4D ign table instead, use a GP output to enable it only under idle conditions. You can put a DI switch on the Y axis to turn the effect on or off (this example turns on/off based on ECT between 40-60°).

Can you attach your tune.

The only one that will likely be connected to the ecu is the tacho and all aftermarket ones that I have seen would be compatible.

You can buy it direct from Seeed for half that price. There are users on this forum successfully using both the Blue adapter you have and the seeed studio one.

Most of the basic setup looks ok to me. The VE table numbers are a bit bigger than they should be but the shape looks pretty normal for a rotary. I see a few problems that should be fixed, but I dont think any would fully account for the high VE: Injector deadtime table should have Differential fuel pressure on the Y axis, not "AN4 fuel pressure". Fuel corrections>IAT Fuel correction should be turned off in modelled mode. Alternator is not charging, batt voltage is only 12.5V. Injector timing>Injection timing split should be turned off, this is only used for siamese port engines. A couple of other thoughts: Does the MAP read the same as BAP with ign on, engine off? I generally find with rotaries it is easier to tune the staged injection transition if you bring the secondaries in much quicker. You have them coming in quite gradually which could be difficult to compensate for the wall wetting effects in the dry secondary ports. Example staging table below from our FD, you can see I have it going from 0% to full blend over only 20kpa. I would also do a quick validation test, log a run at low load with the staging table set all to zero, then do another similar test with the staging table all set to 100%. You should have the same AFR for both. This just confirms all injectors are flowing the same - both rails have same pressure etc. But as a general comment your fuel control looks quite acceptable, just the VE is a bit higher than expected.

Is the lambda not working? Can you attach a log.

Start by following the guide in the help file:

Sorry, Im not sure I can help with that. It sounds like you have an issue with the adapter. I only have experience with the seed studio one.

Try changing aux 4 active state to high. I also found a variation of the ST185CS pinout where the fuel pump speed relay is on pin 56 (not connected on our ECU), so if changing active state doesnt solve it you could try moving the wire from pin 56 to aux 4 (pin 39).

"Ghost cam" is usually done with ignition timing rather than valve timing.