dan l

iTrader: (3)

First, the stock knock sensor signal works great. I have a microphone to clip to my block and I've never heard any knock FWIW.

You can go ahead and tune for 13:1 air/fuel. You may make ok power even on 93 octane. The problem is the real world. The mitsubishi engineers aren't stupid. The stock maps are rich for a reason, to keep things less volitile. The innovative link explains how richer mixtures lower the burn rate. Now, the stock maps are MUCH too rich, I will not deny this (the car has to be able to take a pounding). The problem is if you tune a street car on the edge of volitility with respect to air/fuel mixtures you better be watching (or listening) to the knock every second your in boost. I have my OEM boost guage in my DSM set up to read knock. You'd be surprised how far your "perfect" tune is off from the morning drive to work to the evening drive home. A little boost leak, bum tank of gas, etc.... Tune for the edge all you want, but keeping it a little rich is also a good idea.

RazorLab

iTrader: (8)

Reading comprehension my friend.

Jorge T

iTrader: (18)

Here are some I saved.

l2r99gst

iTrader: (2)

Boy, this thread has went to crap fast.

nj1266, if we wanted to know what Klaus said every second of the day, we would probably just move in with him. Get off of Klaus' bandwagon and just learn to listen to other experienced tuners once in a while. Yes, Klaus is an experienced engineer, but there are people on this board that are engineers and just as experienced as him....even moreso when speaking of 4g63 turbo applications.

Tuning a 4g63 to 12.5-13.1 AFR on pumpgas for a daily driver is just not realistic, in my opinion. For race gas, fine, but pump gas, no way. You will have to reduce the boost a lot, and even then, you are on the edge of safety, as many people have tried to explain to you. We can talk another 10 pages about this and all of the science behind it, but why? Do what you want and then share your experiences if you wish. That's what the forum is about. People are sharing their knowledge and tuning experiences with you, but you just shove Klaus down their throat.

Good luck and keep us posted.


Eric

kapolani

iTrader: (6)

I am using the EcuFlash software to flash my own car.

I DON'T however feel comfortable messing with my own timing maps etc. I disagree with your statement that a baseflash doesn't take any skill. I have no idea how to do something simple like that for myself, so paying $50 is a no brainer for me.

mchuang

iTrader: (11)

That is what the forum is for you ask questions, but whatever. Just make sure you never ever touch that stock knock protection with those timing maps your gonna pay for and good luck.

nj1266

iTrader: (6)

Read the entire thread that I started in the Vishnu section. In it there are at least 4 people who have logged AFR and reported 12.5:1. It is clear to me now why this is so. Shiv loads his DD dyno to simulate fourth gear or dynos in 4th gear. When I testes on the dyno my 4th gear AFR was 1 point richer than my 3rd gear AFR.

nj1266

iTrader: (6)

Yeah, but you did mention 91 ;-) I so hate 91 octane gas on the EVO.

nj1266

iTrader: (6)

I am not shoving anything down anyone's throat. I am just having a discussion about different tuning theories. Malibu Jack is the only one who did not jump down my throat when I mentioned something that goes against the conventional wisdom of advanced timing/increase boost and keep your AFR somewhat rich.

I fail to see the danger if I am using 93 octane gas, the timing is retarded to 91 octane gas levels, the EGT is around 800-900 C, the AFR is at 12:1 and there is no knock.

RazorLab

iTrader: (8)

Finally something we agree on!

NDgsx

I've been tuning 4G63s for 6 years and gone through a few different approaches to tuning. For the first couple years it was just following what everybody "in the know" was doing, as I really didn't know enough about internal combustion to really form my own opinion yet. It wasn't until I put in a Proefi (precursor to the AEM) and started playing with things on my own that I really started learning. My current approach to tuning is very similar to Malibu Jack's. Here are the approaches I've tried:

1. In 00-01 my tuning was HORRIBLE. Moderate boost, extremely rich AFRs, massive timing advance. Worst approach to tuning ever, it doesn't make power and it's not particularly safe either. This was before widebands were viable as a normal consumer though.



2. 02 was all about running the car super rich still, with very limited timing, but a lot of boost. It was decently fast at least I would just tune with the narrowband for .85-.90, turn my base timing way down, up boost to around 23psi, then slowly advance the cas until I got knock.



3. 03 is when the wideband came into play and I found out I had been running around 10.2:1 for the last few years I then tried to lean it out and found that I couldn't run quite the same boost anymore without knocking, but I was running the same times anyway.



4. Late in 03 I started playing with Proefi, and then switched to an AEM in the spring of 04 (it was easier to talked to other people about with the same software). This is when I finally got real timing control and things got interesting. My first tuning attempts I was running timing completely linearly from low RPMs, I didn't really realize that I should be leaving timing lower around torque peak. I also had a horrible knock threshold map that wasn't helping. This approach was slooooowwww and blew two headgaskets.



5. That fall and winter I read anything I could find on tuning, whether EFI or carb, and then went about methodically testing while making more conservative changes. Through that spring and last summer this is what I came up with as my approach (which has some things set specifically for tuning an AEM, but the concepts apply to any engine):



First step is to tune for a flat AFR. I use boost comp in the AEM, so I'll do pulls at 0psi and get a flat 14.7, then adjust boost comp percentage to get the desired AFR at all the other loads. I'll set 15psi at 11:1 to be on the safe side.



Then I'll set up the knock control. I log pulls at 15psi with extremely low timing, and then set the knock table at .2v above the highest spikes I see. I set up for very aggressive timing retard that goes back to full advance very slowly.



After this is done I'll tune for timing. For a car that has a small to medium size turbo without much lag I'll advance timing in the low end, for larger turbo cars I'll leave it lower to increase EGTs and help spoolup (I'll also leave it around 13:1 during spoolup). I slowly advance timing across the board until I see knock retard in an RPM range, then I'll pull it in the RPM range, and advance it at all other RPMs. What you end up with in the end is a curve, here's a completely made up example:

2.5k 5 degrees

3k 6 degrees

3.5 8 degrees

4k 9 degrees

4.5k 7 degrees

5k 4 degrees

5.5k 8 degrees

6k 11 degrees

6.5k 14 degrees

7k 17 degrees

7.5k 18 degrees



That isn't off any particular engine, just numbers to give you an idea of what the curve looks like. Your lowest timing advance will end up being at peak torque. Timing below peak torque stays slightly more conservative, then it dips hard at peak torque, and climbs aggressively past peak torque as cylinder pressure drops off.

I'll keep timing about 1 degree lower than the knock threshold on pumpgas.



After this I'll raise the boost pressure to the next load point on the boost comp table, usually 19psi, and verify that the AFR is the same as it was at 15psi. After that I'll make sure that the boost comp % rises linearly from 15psi up.



Then I copy and paste the 15psi timing map, pull out three to five degrees of timing, and slowly advance it back until knock becomes apparent.



Keep on repeating these steps raising the boost pressure, once you reach a point where timing is running extremely low to avoid knock and EGTs start getting really hot you need to back off on the boost.



OH YEAH OFF MY TANGENT AND BACK ON THE SUBJECT



I've tried a lot of different AFRs, high and low timing, high and low boost, and in practical purposes from my experience AFRs, in a range, do not account for much change in power. As a rough rule of thumb here is what I have seen:

Richer than 11:1 and power drops off



11:1-11.5:1 and power stays high, but it is more forgiving to changes, ie the car is less susceptible to knock being daily driven



11.5:1-12.5:1 power is damn near the same, I haven't seen much point in going to 12.5:1 because the difference in power was miniscule. This is on a car making ~600whp on racegas on the dyno



12.5:1-13:1 Required even less timing advance and did not make more power than 12:1 with more timing



On pumpgas I'll run cars around 11.3-11.5, on racegas 11.8-12.



The key to power in my experience is getting the right ignition advance, cam profile for your particular setup, and the right centerline for your cam and desired powerband. AFRs don't have a huge effect in my experience and trying to run a car in the "best AFR range for power" is just not worth the time.

Oh and another tip, sometimes past peak torque you CAN end up running overly advanced timing to the point of choking power.

dan l

iTrader: (3)

Nick, good post. What you've learned in the past few years is exactly what I've learned in the past few years. The only real exception is that I started with a pocket logger and instead of getting a standalone I did the ecu chip editing thing.

l2r99gst

iTrader: (2)

NDgsx,

Great post....pretty much exactly how I do it and correlates with my experiences, also. The only difference is that I have used DSMLink and a wideband, so I sort of had a 'head start' on tuning with such a great product and board to help along the way. Thanks for sharing your experiences with this board. It's always good to have real world data and not just theory when it comes to tuning.

Eric

hawaiian_evo

iTrader: (13)

oo