The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support Kiggly Racing
Please Support STM Tuned

Hows my SD cruise log look/knock under light cruise AND timing questions

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

92nolatist

15+ Year Contributor
413
1
Feb 13, 2008
Roxbury, New_Jersey
So like the title says, how does my tune look? I think its pretty good except that im not getting too good of gas mileage and im going rich between shifts. Plus, Im having a bunch of knock while im cruising with light throttle (under 35%). And it only seems to do it in 4th and 5th gear around 2500 - 3000 rpm; every so often in the other gears. My KS is OEM brand new as of a year ago, and Ive had this knock issue since then but I ignored it with Link; but now have enabled it again to monitor whats going on. I took away timing in the areas where knock showed up but its still knocking the same amount (around 15* max.)

I know the AFRest looks a little lean in some areas, but I did an SD fuel trim adjust after I got home and ended the log. So that shouldnt be too big of an issue anymore.

So can this be considered phantom knock or is there an issue that Im not seeing???
 

Attachments

  • drivehomefromworkKNOCK41612.elg
    85.9 KB · Views: 100
Last edited:
did you try and use the SD adjust (CombindFT) tool that will help. Plus you wanna smooth out your SD table there are some big jumps e.i. 2000 rpm goes 66.5, 67.0, 49.0, 62.5, 67.5 that is going to throw your air flow off. I had the same issue when i started. Use the SD adjust tool and make sure your SD table is got a nice smooth flow too it.

How old is your Knock sensor? Plus do you have solid motor mounts? I was getting a few degrees of knock cause of the mounts. And it would pull the timing out of my tune. Ended up just disabling the sensor in link.
 
I said in the first post that I did the SDadjust after this log and that my knock sensor is a year old. I want to ignore the sensor again via link but I want to be REALLY sure that its phantom knock. I cant hear any pinging either, and I know what that shit sounds like because my truck knocked like a mother before I adjusted base timing.

I have prothane mounts in all 4 locations.

Would I just raise the lowest cells up to match the others or bring them up a little bit to sort of blend them together?
 
take a look at my sd table i posted should look something like that.
 

Attachments

  • sdtable.ecm
    4.9 KB · Views: 104
Ok, so now if I smooth my table out real nice, the cells that were there would be obviously totally different so wouldnt that change the a/f ratio? Isnt that the whole purpose of the SDadjust, to keep stoich? And wouldnt that throw the tune out of whack? Im a little confused as to why smooth the table out that much. Does that net better MPGs?
 
it will change the A/F to what you have the max octane table set to. The SD adjust is trying to get everything to match the max octane table. Smoothing out the SD table help with the amount of fuel the car thinks it needs. With SD you are using the Manifold pressure and air temp to calculate air flow. If the numbers are way off from cell to cell then the ECU thinks there is more or less air flow depending on what way the numbers are off. And then in turn add or subtracts fuel to match the max octane table.
 
But the max octane table is used by open loop driving, I'm using closed loop which tries to keep 14.7 a/f so I don't see how its using the max octane table to control fuel ?

What I'm trying to figure out is if I smooth out my table from, for example, cells in the 60s down to cells in the 50s in a certain area, wouldn't the fuel trims be in the negatives?

Sorry if I'm sounding stupid, but I'm just trying to wrap my head around this.
 
But the max octane table is used by open loop driving, I'm using closed loop which tries to keep 14.7 a/f so I don't see how its using the max octane table to control fuel ?

What I'm trying to figure out is if I smooth out my table from, for example, cells in the 60s down to cells in the 50s in a certain area, wouldn't the fuel trims be in the negatives?

Sorry if I'm sounding stupid, but I'm just trying to wrap my head around this.

Im in the same boat as you are in trying to dial in the VE table. calan has something called link tools, its in another thread i have which ill post up, it really helps understand the graphic way the VE table works and shows it in 3D maps.... i think your getting phantom knock when your crusing in 4th and 5th... if you have a waste gate touching your power steering or solid motor mounts you could get some of that picked up by the sensor. just change the thresholds of your knock sensor in the ALS/knock tab in live access. i have mine set to 3500RPM and 35% throttle... i don't see any phantom knock.. but from what you said in your first post it looks like thats what you did before... sometimes you just get phantom knock.. and it sucks.. but oh well.

Link tools:

http://dsmlink.com/forums/showthread.php?t=61388

you of course have to have to have excel though
 
Last edited by a moderator:
Wes, according to my WB, the transition is smooth and doesnt dip lean at all; so Im not worried about that.

Yea Ive seen that link before but now im going to have to read through it.

I just took a 3rd gear pull and theres no knock whatsoever which is awesome except a 0.4 for a split second; but the car was a complete dog until it hit full boost (18-19lbs). Im using the evo8mod1 timing table and I dont like the way the car feels in spool up :notgood: Plus it used to pull a ton harder with the stock 1g table even in spool up.

Heres a log of the pull and a file of a stock 1g timing that I tweaked ALOT (not running it, just want to see what peoples opinions are on it)
 

Attachments

  • 3rd pull after dark 417.elg
    11.7 KB · Views: 75
  • 1Gmaxtimingtableremapped.eda
    6.5 KB · Views: 78
Last edited:
But the max octane table is used by open loop driving, I'm using closed loop which tries to keep 14.7 a/f so I don't see how its using the max octane table to control fuel ?

You are absolutely correct; when everything is working as it should and the ECU is in closed loop, it is shooting for an AFR of 14.7:1... regardless of what the DA AFR tables say. (The exception to this would be if you are simulating the front O2 narrowband sensor, and have the switching threshold set to something other than 14.7:1).

What I'm trying to figure out is if I smooth out my table from, for example, cells in the 60s down to cells in the 50s in a certain area, wouldn't the fuel trims be in the negatives?

It depends on which way you adjust the VE cells. If you increase VE values (or MAFComp values), you are telling the ECU that more air is entering the engine at that point. The net effect of this is that the amount of fuel delivered is increased, the car runs richer, and CombinedFT will be more negative.

For a simplified example, let's say that at some given point, CombinedFT is reading 5.5. This is telling you that the ECU is having to add 5.5% more fuel than it thinks it should in order to hit that 14.7:1 AFR. Other ways to think of it:

1). You are 5.5% too lean and need to add fuel at that operational point
2). You're airflow calibration is off by 5.5% at that operational point, and the ECU needs to be told this so it can inject more fuel
3). CombinedFT represents a 5.5% error in delivered fuel, based on an AFR target of 14.7:1 and feedback from the front O2 sensor

It all boils down to the same thing; just think of it in whatever terms makes it easier to remember. Personally, I just look at the CombinedFT in closed loop (or WBFactor in open loop with a wideband), and adjust the corresponding VE cells by that same percentage. For our example, once you increase the appropriate VE cell(s) by 5.5%, CombinedFT drops to 0 (theoretically) at that particular point since you have corrected the 5.5% airflow error that the ECU was having to compensate for.

*****

As for your log (quick glance)...

1. Your DA AFR table is dotted with some 11.1's and 11.2's, rather than just being all 11.0's. Doesn't matter much, but it's just one more thing that adds some fluctuation to the WOT AFR.

2. Your VE table isn't horrible, but it needs to be smoother in the low-load areas (upper left).

3. Why is anti-lag enabled? Are you using it while still dialing everything in?

4. Your LoadScale value (DA) is set to 114.8%, which is a default value that a firmware update changed, due to some changes in the way ECMLink handles load (See this explanation). Since you are basically starting over on your tune, you may want to set this back to 100%, which should move you up in the fuel and timing maps.

Your tune actually looks pretty good, based on BoostEst/MAP and other values. Get a wideband so you can do some real tuning. ;)

BTW - That knock in the first log is PK...unless you were lugging the shit out of the car in high gear at low speed. It appears to be happening at random RPM, throttle, and load, so my guess is that there is something loose rattling around.
 
Last edited by a moderator:
Wow, craig, thank you so much. This is the information ive been wanting to see.

I understand about combinedFT and the effects of positive or negative percentages, but lets say that a cell is at 65.5 and combinedFT needs to drop 5.5 percent for that one cell; would I lower that cell to 60 and be dont with it?

About the log.
1. So Ill just need to drop them all 11.0 then.

2. As far as smoothing goes, do I adjust some cells up and others down in order to get a smooth transition or is there some different approach to it?

3. I used antilag for launching the car on the old MAF setup and never took it off.

4. I set the load scale to 114.8 because a write up on tuning with speed density that I read said to put it to 114.8 because with SD, there is something different with loadscaling. I didnt really understand why but decided to do it anyways. There is also a thread on the ecmlink forums that kind of explains why but I cant remember the title.

EDIT: Heres the write up: http://www.dsmtuners.com/forums/articles-tuning-ecu/395564-how-tune-1g-v3-sd-ecmlink-no-maf.html A link to the thread Im talking about above is in document.

I do have a wideband but its not hooked into my ecu at all; its just by itself. I guess I have to do that now...

Yea i figured it to be PK as well, but I just wanted to be sure.



Did you by chance get to look at the 1g timing remap log I put up? Because I would rather run with a 1g timing map because I like off boost power compared to the evo8mod1 map I have in there right now. Plus I think the evo 8 map is a little too conservative for me personally.

Thanks for your help so far.
 
...lets say that a cell is at 65.5 and combinedFT needs to drop 5.5 percent for that one cell; would I lower that cell to 60 and be dont with it?

Nope. :)

You'd change that cell to 61.89 (rounded to 62). You want to multiply the VE values by the corrected percentage;

Code:
65.5 * (1-.055)

or

65.5 + (65.5 * -.055)

In English, that would be:

NewVEValue = CurrentVEValue * (1 + CombinedFT)

or

NewVEValue = CurrentVEValue + (CurrentVEValue * CombinedFT)


As far as smoothing goes, do I adjust some cells up and others down in order to get a smooth transition or is there some different approach to it?

You'll usually lower one and raise another, until CombinedFT is pretty flat throughout all load/RPM points and the VE table makes a smooth transition up and then back down. It's an iterative process; just run the VE CombinedFT tool over and over and it will start to work itself out.

I set the load scale to 114.8 because a write up on tuning with speed density that I read said to put it to 114.8 because with SD, there is something different with loadscaling.

The link I posted up there ^ is to Tom's explanation for it over on the link forums. ;)

I would rather run with a 1g timing map because I like off boost power compared to the evo8mod1 map I have in there right now. Plus I think the evo 8 map is a little too conservative for me personally.

If you can safely run that much timing with whatever fuel you are using, more power too ya. But I wouldn't get in too big a hurry to raise it much until you get the wideband hooked up and a little more experience with Link. Just a few degrees past the knock limit at 7500 RPM and high loads can ruin your day pretty quickly. :)

Remember that the higher in load and RPM you go, the more critical the tune becomes and the less time you have to react to a tuning issue. Keep it safe and conservative until you know for sure that your airflow (VE table) and fuel are properly calibrated at full load and RPM...then start incrementally raising timing and playing with the AFR a few tenths at a time to find power.
 
Last edited:
Ahhh, ok, that makes a ton more sense to me. I will definitely apply that equation to my tuning adventures. I was at work when I went to look at the link and I dont remember my password to the forums so I wasnt able to view the link.

Thanks for looking at that, I am using 93 pump so it might be a little iffy. Ill lower the timing some and give it a whirl. If it doesnt work to my liking, then Ill start over again LOL. A little learning never hurt anybody.
 
Last edited:
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top