I think I'm starting to confuse myself here, there's too many things going on at once that I'm trying to tune and it's scrambling my brain lol.

The setup is my friend's 2g eclipse with an evo 8 ecu conversion running ecuflash AND a MAFT box with a 3.5" gm blow-through setup. He didn't have the manual for the maft when I first stopped down so I was hunting blind a little trying to remember stuff from years ago, but I found one online to go through again. The problem I'm running into though is 2-fold...(A) what scaling do I set for the maf scaling in ecuflash (I'd really prefer to keep the Evo scaling if possible, but I do have a rough dsm maf scaling), and (B) what would I set for injector size on the maft? He currently has 650cc injectors installed right now, but once I get a steady base for them he's jumping up to a set of 1000's, so I'm really not sure how I'd set the MAFT for injector size or what switch setting it would be. The manual's are very vague and I personally want no injector compensation done through the MAFT so I can control it directly through the ecu, although I would probably assume a stock setting of 550? Talk about over complicating such a simple swap, I really just wish he put an Evo maf on there instead...would've made my life so much easier.

 

Haha...bump for answers still? Regardless I think I'm going to try both 450 and 550 for injector settings on the maft and see which one responds better. Also it seems as if the base adjustment raises or lowers the actual output values so I'm going to try matching them to the Evo's maf values.

 

I would guess if you are using an evo ECU you would want the maft to "look" like a standard evo input so that everything could be controlled with the ECU. Set it as expected inputs for an Evo ECU and you should be fine to work from there. If I remember correctly the MAFT manual specifies which settings should be used as a base for 1g, 2g, or Evo.

 

That's where I ran into a wall, the owner never had the manual for the MAFT (version 2.0) and all of the online manuals only ever mention the 1g and 2g. Currently I have the ecu very roughly scaled for the 2g maf size and the settings on the MAFT set for the 2g as well.

 

Ok, I just ran into the strangest problem I've seen yet...the load % on this particular setup is moving downwards when you tap the throttle or slowly raise the rpm. It starts out around ~30% at idle, and will spike up a hair before it drops down even lower than the idle % according to my quick evoscan logs.

I figured something was screwy with the MAFT at first, so I swapped the stock DSM maf back on...and guess what, load % still goes down instead of up. I'm honestly dumbfounded and have no explanation at all as to why this is doing this. Help?!? LOL

 

You have to back track a few steps.

Make sure you have ALL the things done correctly that is on this pre-dyno checklist. This checklist is absolutely necessary for proper tuning or else you may be trying to tune around a mechanical problem.

http://fftec.com/PreDynoChecklist.html

Next you need to double check your wiring @ the ecu make sure you have all the wiring correct to the MAF. Then you need to eliminate a variable and use a EVO maf. That way you are leaving the MAF settings alone.

TRY and keep the path from the intake tube to MAF smooth as possible. This will give you a clean reading. Sharp bends directly connecting to the maf will give you odd readings.

Next dial in your injector size and latency to get your fuel trims in in line. I would just go ahead and jump straight to the 1000's and skip the smaller injectors. This can take some driving. Try to get your low and mid fuel trims to stay with +-5%

Next you need to install a MAP (AKA Speed Density) sensor and wideband to datalog with in Evoscan. You will need to do some actual tuning keep on the rich of fuel and super conservative side of timing like a stock car would be tuned. You need to be able to do full WOT rpm sweeps to redline @ the boost level you plan on running.

Make sure you are logging MAP sensor, MUT2 load and HZ along with all the other necessities. The standard Evoscan Airflow will only read up to 1600 and if you plan on maxing out the 1000cc injectors you will be up in the 2500-3000hz range.

Once you have a bunch of wide open throttle datalogs install your GM maf and translator. Start with 500-550cc inj settings and have all other fuel adjustment settings zero'd out. Something to keep in mind a smaller injector size setting on the maf t will give you more HZ and larger will give less.

Keep in mind once you switch over to the GM MAF and translator everything dependent on HZ will be changed. The constants to work with from the old datalogs are RPM, MAP and AFR and HZ.

Do some light throttle short pulls and log the same things as previous. If the HZ is way higher @ the same boost level as the previous logs then you need to adjust the MAFT inj size to a larger injector. Once you have your HZ from the MAFt logs close to but just a tad higher than the old pre-maft logs.
Once you are that far you can go a couple of different paths.
1 use maf smoothing to get your Fueling back to what it was. (this would involve scaling your maf)
2 tune as normal with ECU flash.

I would recommend getting the cars injectors dialed in and just doing the Speed Density conversion that way you don't need the GM maft. It works AWSOME!!

 

I appreciate the info, but I've already covered pretty much everything you listed other than jumping straight to the 1000's. The injectors are dialed in pretty well with no more than +1.5% correction in the low and mid and with an o2 feedback correction of less than 5%. If he had an Evo maf to plug in that would make my life a lot easier and I could get on to the rest of the tuning process. The problem is, the Evo maf and the 2g maf are a direct wire up, so that being the case even the 2g maf on the Evo ecu should still produce a rising load %, not a falling one. I appreciate your help, but I am no noob when it comes to tuning and flashing (having been tuning for well over 10 years now) and I've already gone over everything you mentioned...in the beginning where it should be done.

Pretty much I'm looking more at the coding for 99270000, because it's an EDM rom on a USDM ecu, would that cause the values to be inverted? And if that is the case, I need to find a way to disable all CAS checks (and quite a few others) in the periphery bits to use 94170015 instead.





Oh, and that little part about the AFC's on your site...that's classic! I tell people all the time I don't want to touch anything with an AFC attached to it, guess I'm really not far off with this MAFT box right about now though. :P

 

I really hope you figure out the inverted load issue. That is a little strange. I just thought I would list everything out at once that way we're not posting one sentence yes, no, try this etc etc. Does the Inj pulse and Ign Timing correspond to the inverted load?

 

The timing jumps around...a lot. At idle it jumps anywhere from -9 degrees to 14 advanced, when you raise the rpm it stabilizes a hair but not by much. The injector pulses seem to follow suit with the dropping load %, as load drops so does the pulsewidth. Timing advance increases to around 37-40 degrees advance though when the load % and pulsewidth drop off. I'm going to try hacking up a 94170015 with everything disabled and see what the result is.

 

let me see if i can try to help out a little bit on this one. i know a few people running the 90550001 ROM with only 1 minor annoyances being cel codes for idle circuit and generator FR circuit. i currently running the 96530006 ROM with the same cel codes for idle circuit and generator FR circuit but i am also getting a code for a dead rear o2 sensor. other than that i m running a MAFT gen2 and evo8 injectors, so my setup is pretty close to yours. try setting your MAFT up as if it were installed in a 3000gt with the 650 injectors. my understanding is that the 3si maf sensors flow close enough to the same as the evo maf that on the MAFT gen2 they are paired up on the input and output settings. also something i do want add i don't know if you had this problem as well but when i did my swap i had to switch over the spark plug firing order... i had to do the same with the injectors.... i found out it was causing some problems with the way the car ran especially at cruse on the freeway and at full throttle.

 

Thanks for the info, I'll look into that. Yesterday I tried a few things out, mainly taking the 9417015 rom and applying all the periphery bit settings how they are from the 99270000 rom. Now the load % appears pretty normal and everything seems to have settled down a lot. I have the scaling at the maft set up for 560cc injector size and it's kinda close, although it seems to be about 9% scaling difference off from keeping a nice steady 14.7 afr idle. I have a feeling that the owner's front o2 is pretty much shot, because the feedback doesn't really move around much and the o2 voltage seems to be sitting in one spot all the time. There are also a bunch of codes popping up (everything from crank angle and cam angle sensors which are probably from the 1G CAS swap to boost control solenoid and a bunch of others) but considering the condition of the wiring harness setup I can deal with them. So far it idles smooth, revs smooth, and seems to pull pretty smooth as well, with the exception that I have to add quite a bit more fuel in for his To4E 50 Trim and lower the throttle crossover points. At this point I'm pretty happy with the results from yesterday, at least there was progress, and I never really expect to see this car run code-free since it was never really intended to have a 4g63 in it anyway and the wiring harness has more than a few empty dangling plugs. :P

This is honestly a setup that should be running an AEM standalone.