No start fresh build R33 GTST RB20DE Neo+T AEM Series 2 ECU 30-6620

[Neostead2000]

Ok another update dear god I hope this is the last one!!!

 

I managed to make the car start and work with the custom 24-1 trigger disc! I checked the CAS connector signal wires once more and noticed that pin 4 (1* crank) signal had 5 volts as it should. But pin 3 (120* CAS) signal had 0.223v, which should be 5. I then suspected the ECU itself, as I noticed it had been opened before due to the warranty stickers being slightly off.

So I took it apart, carefully separated both circuit boards and started looking at it dumb. I traced the 41+51 and 42+52 pins and they went to Jumper locations which seemingly had no jumpers positioned on them. It appears the previous owner must have swapped them or it may have come from AEM this way. I placed the jumpers on locations JP1 (1+2) and JP2 (1+2), leaving JP3 and JP4 disconnected entirely.

Placed everything back, selected the right type of signal (falling edge if going from 5v to 0) which is CONTRARY to what AEM have listed on their instruction sheet. I put the CAS back in ballpark area and what do you know, she fired right up and running even better! I quickly set the timing back to factory 10* advance and now it's running fully sequential injection and ignition.

Basically the takeaway is, AEM is absolutely trash with very misleading instructions and outdated information. To think they still offer these series 2 ECUs for nearly $2000 or more is insane. I paid $1100AUD for mine so I guess it's worth it.

Now I can finally focus on tuning this damn thing!

[GTSBoy]

Good work. Nothing better than having to spray seemingly random jumper and other settings at a piece of equipment because the manufacturer's doco is crap.

[joshuaho96]

23 hours ago, GTSBoy said:

From a tuning perspective, there is no difference in pressing pgup or pgdn on a cell reading ms or % or some arbitrary index number (ie 0-255). Who cares? Number get bigger, fuel gets morer, 3D view of map gets taller, colours change appropriately.

I'm a big fan of more advanced modeling so it's easier to constrain the problem and there's less opportunity to shoot yourself in the foot. As OP mentioned you know when you're really screwing something up when working with a VE model if the numbers don't make any sense. 100% VE at idle for example. X-Tau constrains the problem of transient throttle such that you're only picking a wall-wetting constant and time constant and everything else just flows out from there. If your wall wetting constant or time constant is some ridiculous value you know you're doing it wrong.

[joshuaho96]

whoops

Edited October 21, 2023 by joshuaho96
pls delete