OK guys my issue is a little complicated because the evo ecu is in my 1g DSM. Don't let it scare you. The car idled crappy so I checked the IAC it was bad so I replaced with a working unit. And now the car idles at 1500-2000/rpm. car idling at 2100 rpm when it's cold. And dropping to about 1500 rpm when warmed up. but the car runs perfect with the DSM ECU no boost leaks. Any ideas?

 

Not too familiar with the DSMs but if they are similar to the EVOs ECU, you may have to electronically lower your idle rpm level within the ECU. If not, it could be other issues such as a throttle cable without enough slack. May need to loosen it up a bit or grease up the throttle cable.

Another possibility is that you may need to adjust the BISS. It may be set to allow too much air at idle which would raise your idle.

 

Well it's the evo 8 ecu so it's the same as an evo. I tried setting the BISS screw but it jumps to 2200 rpm when I start the actuator in evoscan and doesn't respond to screw turning.

 

Then I believe you found the problem already its in your throttle body

 

the car runs perfect on the DSM ECU so I assume that it's something with tephra mods or maybe egr related.

 

Open ecuflash and pull up your rom for the idle setting. Or download your rom from the car itself and check idle. What is it set to?

 

It should be stock, never changed it. I can't look to make sure because I'm at work. I'm 100% its set at 850.

 

If AFR is OK adjust your base ignition timing to 5° (first lock it via MUT / evoscan, then adjust the CAS to get to 5° BTDC), you don't want to run the stock 7° base ignition with an evo ECU!
If the ignition timing is OK, THEN adjust the idle bypass screw (again via MUT / evoscan)
The idle stepper has a different base-setting on the evo ECU than the old ECUs had, so it has to be readjusted!

If it's still idling too high try to limit the ignition variation on idle. This might also help with "hiccups" at idle.


For really smooth idling and properly working idle control, the MAF has to be scaled properly. Stalling at rundown or near-stalling after shortly opening the throttle is often (also) related to incorrect MAF-scaling.

And of course *any* leaks *have* to be fixed! especially the old throttle bodys tend to be massive leaks on vaccuum and pressure! replace the shaft seals with new radial shaft/lip seals and also check/replace the lower injector seals within the head.

Also the old EGR-valves tend to open at relatively low pressures (0.5bar), leaking boost into the exhaust, completely messing up AFR readings. replace it with a new one or block off the port on the intake manifold (-> higher cumbustion temperatures! look for decent high-octane fuel or prepare yourself to lower ignition and/or boost levels...).



Last resort: try another ROM ID. My Galant E30 ran absolutely fine on 9055 - started up on first try had perfect idle after a few minutes. 9653 was a PITA to even get it running longer than ~3-5secs and throwing lots of error codes, not to mention getting "acceptable" idle behaviour...

 

DSM timing is 5° BTDC. So I thought it would work. I don't think it's that but I'll double check it. I'm using a EVO 8 MAF shouldn't need to be scaled. I'll also try another ROM. I've been wanting to block of the egr I have just been putting it off. I'll do the egr. Then if that does not fix it I'll do a different ROM.

 

did you set up the ignition timing or just pretended it would be right? The old ECUs and evo ECUs use different offsets for TDC-signal <-> ignition timing. So if you didn't adjust it, it's most likely not at 5° BTDC base ignition...
Same with idle steps - both ECUs have different initial positions, so it has to be re-adjusted when switching to the evo ECU.


The MAF definately needs a proper scaling as well as the injectors. The flow characteristics of the intake + turbo are playing a huge role in the scaling and how the ECU calculates load and air mass from it. I've also used evo MAF + injectors but scalings were way off even on the ROM which belonged to both parts. Injectors not that much, but the MAF curves looked not much like the ones in the ROM that "should fit the MAF" after rescaling. Especially idle and high flow areas needed lots of adjustment and due to the low airflow idle is especially picky about the correct scalings...
Do yourself a vafour and rescale it to fit your assembly - if load is correct and therefore predictable and AFRs dead-on, trouble-shooting and tuning will be much easier. 100 load units should be roughly equal to 100 kPa manifold pressure. use ECU-load for low Hz areas (higher resolution) and 1byte load for higher Hz and check that your evoscan formula doesn't add any corrections on top of any.


I found that with an "unscaled" MAF load is most likely not "off by X%" but just variably wrong all the time depending on rpm, throttle, weather, moon-phase and fertility of the local squirrels... Impossible to get any tuning done right with that.