You are on the right path with tuning the MAF scaling. Remember the MAF is set from the factory to go with certain engine VE characteristics. By putting an Evo MAF in a lancer, I am certain you have straight pipes going from the MAF to the turbo. This and filter choice are key factors for the MAF based system. Anything that can cause turbulence in the airflow through the maf will cause problems in accurate readings.

I would start with logging Airflow Hz at idle and verify you are within parameters i.e lower Hz value than 40 at idle. Do the same at cruise....you should be somewhere between 100 and 300 hz depending on the incline of the roads.

We can go advanced once you get this in order.

I don't have a stock Lancer rom.

Shoot one to me and I will see what I can figure out for you.

Same SN at G Mail.

Paul

 

Here is one quick quote I found on this forum. I do remember reading other people saying the same thing.

Look at it this way, if the fuel map = afr, then everyones fuel map would look the same. 14.7afr in cruise, then richening up in boost, and end in 11afr in WOT. If you change your way of looking at the fuel map, then you can just tune for the single cells that are giving you trouble and not have to mess around with the maf scaling which would throw off the rest of the tune for an entire row or column.

 

Are those numbers the ideal numbers for an Evo Maf... Both 8/9 Mafs?

 

Yes as you can see in my quote I stated the same thing the fuel map doesnt reflect ACTUAL afrs, I understand this thank you for pointing it out though incase I didnt. This maf scaling is what throws me off big time, I cant use the factory lancer scaling since my intake setup is literally completely changed from stock, the only thing the same is the ports on the head, I ported the intake manifold and every piece of intake piping is different from stock. I have tried using the stock evo 9 scaling since this is the closest to my setup with all stock evo 9 intake piping bov, recirc pipe, uicp, licp,fmic and the evo 9 maf, but it makes my fuel trims go wayyyyyy lean and to compensate I have to adjust my latencies probably a good 3-4x more than what stock wrx 440 latencies are which seems its best to stick as close to the stock latencies as possible but idk as I am still new to this and trying to decipher all the info I read.


Applaud: thanks a bunch man I will shoot you my stock rom right now. I have always logged airflow and do find my idle runs mainly 25hz with drops to 18.75 and raises to the 30 mark at times too. my cruising hz are around 250 at 60mph I believe maybe a little less.

 

Yeah they are pretty close for all Evo 8/9 mafs.

Some will be lower, some higer, but in general they are about the same.

In the case of those that cannot get LTFT low to adjust it is usually as a result of not idleing in a low enough Hz value. LTFT low won't adjust unless you are below 40ish Hz.


Something else to consider with the MAF is that you can essentially lie to it. On the bottom there is a small bit of silicon that covers a large phillips head plastic screw. Remove the silicon and you will see the screw head. Then look though the MAF where that screw goes in to. You will see that it blocks a good deal of space. Back out the screw until the area inside the MAF is open completely.

In most cases you will notice you idle might change a little. The biggest benefit is that you remove one more restriction from your intake tract.

 

Rom as been sent applaudd. Now with regards to the maf screw on the bottom, I have read about this before , this will allow even more air correct? So I should probably leave this alone since I am having a hard enough time supplying fuel as it is without bringing the latencies wayyyy out of normal range right? I have my 440s scaled to 390 btw

 

Your 440s should be scaled closer to 396 ish. You are a little too small.

 

Well I had them at 412 with my stock maf but I had everything in the maf scaled wayyyyy up for this.

 

rough approximation of 2 byte load-

maf hz/rpm * 852

If I am looking at it correctly this would mean at 4500rpm for instance-

800/4500 = 0.17777 * 852 = 151

5500rpm-

950/5500 = 0.17272 * 852 = 147.16

Seems pretty spot on to me, did you have these injectors cleaned or flow tested before you installed them? I wonder if there is a MAX rich table limit in this ECU like there is in some others possibly?

By the by, an Evo 8 ECU is normally 100-150. Not sure exactly how PnP it is though.

 

I stated that incorrectly I think. I am thinking more like the leanspool map where there is correlation between an AFRmap value and the applied value ratio, like this-

 

Well I have found some things out. My timing map is following my ecuload value to the tee. My fuel map is not. It is refrencing some higher calculated load. I ignored the load all together and adjusted the cells the evoscans afrmap parameter pointed me too and got some promising results. At least promising for being a Le to now properly tune the afr. I also found that my ecu has a bottom of 10.01 so this is the reason I could not get any lower actual afrs after I had passed this point in the fuel map. Now this does help part of my issue but still remains the high afr in that specific rpm region. Now if my theory is correct I am going to drop the mad size down and then try to being my latencies to factory wrx spec them use the maf scaling to adjust trims. I did not get them checked which o regret but I am going to try this and see what I get. I will also bring my s ing up to 396 too. Sorry for crappy spelling I am at work posting from my phone lol.

 

Its easier to drop scaling, trim latency to lean back out, and leave the MAF only as an absolute last resort. If your values cant go richer than a 10.1, set all the values to that and tune the MAP till its right. Its not going to look pretty but results are the key not how the map looks really.

 

Thank you John really appreciate the help. I have been afraid to drop the scaling more because everyone says its too low making me think its gonna really screw up my idle. If not I have no problem dropping the injector scaling more and lowering the latencys to get my trims back in check. So your saying to set it all to ten after adjusting my injector scaling down more then raise it where I need to lean it out to get my desired Afr ?

 

Yes the 2003-2004 evo ecu is almost completely pnp the harness is identical except for the swapping of 2 pins one is the fuel pump and the other is a\c. Other than that I paste my maps into the evo maps and viola. What sucks is I have had 2 evo ecus, the first was an 05 which seems to be a little too problematic for our 4g94 for some reason. It ran the car but my idle was pure garbage and kept getting random codes popping up. Then I bought an 03 evo ecu from another member on here only to find it was fried upon arrival and the seller refuses to do anything about it . So my $140 paper weight sits in the back seat and I have to find the money for another ecu. Until then I might as well use my stock one to its utmost potential .
The reason I would actually like to lower the maf size is this would drop my load down to a more managable area around 120 or so would be better for my tiny maps. After that in theory this will bring the load used by the fuel table down closer to the 150 load its scaled to right now. With my stock maf I was making wayyy more power and hitting only 96% load. I do have my timing really low while I work on my afr's as well though then I will raise the timing back up to a suitable level.

 

Aaron, couldn't he change the XMLs fuel parameters (min & max display values) so he could actually adjust it to a smaller value also...