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.

  • Update Your Password Today!

    We have implemented tools to identify member accounts using insecure passwords and will be locking those accounts until their passwords are updated. Don't get locked out of your account - update your password today, and ensure your account has a valid email address on file. Read more here...
Please Support ExtremePSI
Please Support Rix Racing

ECMlink New SD setup Hoping to have some "wiseman" check my logs

Tony J

Proven Member
32
6
Oct 2, 2020
Belleville, ON_Canada
Just recently converted to speed density, I have the car dialed in pretty good cruising and what not but at WOT it seems to get knock 2/4 pulls

FIC 1250cc injectors
255 walbro
-6 lines everywhere
fuel lab filter
aeromotive fpr set to 40 psi
fp red turbo
greddy bov
2.5 inch piping
mishimoto intercooler
3 inch exhaust
external wastegate tial 38mm

balance shaft delete
fresh timing job less than 100 km ago
fresh plugs br8es i believe, gapped to .20 if im not mistaken.

car idles great.

Any suggestions would be muchly appreciated!
 

Attachments

  • Couple bigger pulls unmodified map.ecm
    4.9 KB · Views: 26

Stapl3

15+ Year Contributor
1,597
167
Oct 27, 2004
MI, Michigan
I would make a corresponding VE table change based on the WBFactor at the same time you are adjusting the open loop AFR demand."
This is where we disagree. Look at his log. Throughout the entire 2nd and 3rd gear pull he's a full point lean. That indicates the fuel system is miscalibrated, not the VE table. Adjusting the VE table to richen like that is using airflow adjustments as fuel adjustments and it's wrong. I mean, even look at his table adjustments in the beginning of this thread. Giant increases in numbers that are obviously wrong.
That aside, lets assume he's running 5psi boost, so the loadfactor puts him to dance all over the middle/right VE table instead of just using the bottom row and looking right. You're going to adjust the table all over the place 10-19% more? That's crazy.

The car is 10-19% lean and you're suggesting the airflow is 10-19% off. It's not, the fuel system is. The global or base fuel is wrong.

You ARE targeting 9.3. Fix your tables before going further.
Again I feel this is dangerous. He's lean as it is. It's a gift from the default ecu gods that he has 9.3 in there. Because if that number was 11.0 the output would be in the 12s and engine would be in the scrap heap.


Edit: It's worse than I thought looking back on the log, at peak VE it's getting 11.1 with a target of 9.3. 2 points off. The car is seriously lean.
 
Last edited:

jdxnc

10+ Year Contributor
387
343
Sep 15, 2009
Stanstead, Vermont
This is where we disagree. Look at his log. Throughout the entire 2nd and 3rd gear pull he's a full point lean. That indicates the fuel system is miscalibrated, not the VE table. Adjusting the VE table to richen like that is using airflow adjustments as fuel adjustments and it's wrong. I mean, even look at his table adjustments in the beginning of this thread. Giant increases in numbers that are obviously wrong.
That aside, lets assume he's running 5psi boost, so the loadfactor puts him to dance all over the middle/right VE table instead of just using the bottom row and looking right. You're going to adjust the table all over the place 10-19% more? That's crazy.

The car is 10-19% lean and you're suggesting the airflow is 10-19% off. It's not, the fuel system is. The global or base fuel is wrong.


Again I feel this is dangerous. He's lean as it is. It's a gift from the default ecu gods that he has 9.3 in there. Because if that number was 11.0 the output would be in the 12s and engine would be in the scrap heap.


Edit: It's worse than I thought looking back on the log, at peak VE it's getting 11.1 with a target of 9.3. 2 points off. The car is seriously lean.
Yup, I agree there's many issues to solve but yeah, I only quickly looked over the log and I think the global is off and needs to be corrected first.
 

danl

20+ Year Contributor
721
117
Apr 8, 2002
Severn, Maryland
BoostEST is only accurate at peak VE and should be pretty much ignored at all other times. Afratioest I make line up during idle, cruise, and throughout the pull. Pretty much all the time.

The whole thing with Boostest = boost AND afratioest = wb AND the airflow slider be at 0 or SD table cell being at 100, all 3 of those things occurring at peak VE means the fuel system is perfectly calibrated. Then there is nothing to touch with the afpr or global. Any adjustments (spare messing with deadtime) before and after peak VE would be completely done with the mafcomp sliders or the SD table.
I'm sorry but I disagree. BoostEST is what the ecu thinks the boost is after its calculations. If BoostEST is off, then the ECU's calibration is off. The ecu is calculating fuel based off the ideal gas law (PV=nRT) and the math remains the same for all loads. http://www.ecmtuning.com/wiki/sd101

Now, in practice there are reasons this does not happen at low injector pulse widths, but the correlation is still there.
 

danl

20+ Year Contributor
721
117
Apr 8, 2002
Severn, Maryland
And to sidebar on that. I run as high base fuel pressure as I can on all my cars. One of my cars is at 80psi base fuel and it's absolutely fantastic. Maybe the best thing I've ever done with a DSM. The injectors behave so differently, the response and personality of the car is incredible. Never had an issue doing this, only benefits.

Are you calibrating the IPW dead time map to compensate for that high of fuel pressure? Picking a popular injector like the ID1000 that would be a .2ms change at idle, normal idle is around 2.2 ms, so by messing with just base fuel pressure you have mis-calibrated the IPW dead time by 10%. Now you need to go back into the table to calibrate this for all 6 voltages in the ECU. This could also be a contributing factor for why your BoostEST is only accurate at high loads.

The other reason cranking base fuel pressure is not desirable is because as was previously mentioned. Fuel flow drops off. You loose over 50% of the flow of a walbro 255 by going from 40-80psi.
 

Stapl3

15+ Year Contributor
1,597
167
Oct 27, 2004
MI, Michigan

Tony J

Proven Member
32
6
Oct 2, 2020
Belleville, ON_Canada
I am now full rich at wot. Will post some logs.

I am getting knock in weird places now, not really at the top of my pull however.

The last log I sent you guys the car was working good.

This time she was really falling on its face and then would catch and party... couple random knocks here and there. Worse than the last log.

Comparing my fuel settings I added more global and added deadtime.

Will post log shortly just got home and my wife is about to disown me due to all the time spent at the shop ROFL
 

Tony J

Proven Member
32
6
Oct 2, 2020
Belleville, ON_Canada
couple logs. got some weird knock in the drive log. the 3rd gear pull looked clean the drive too other than the few spots.. thoughts?

Ive adjusted global and deadtime again, not sure what would cause me .4 knock. i dont feel its lean at this point and ive made less power according to the last log on these injector settings...

another pull from tonight with brief knock. not sure why this is happening so infrequently... sure if it happens everytime somethings surely off but its so intermittent and only seems to be when i hammer on it. if i eaaaaaase into the pull theres 0 knock, or soo thats how it seems. Im not ripping through the gears but would like to come spring LOL

Afr #s seem to be closer too in each of the logs i posted.
 

Attachments

  • clean 3rd gear pull.elg
    26.3 KB · Views: 11
  • drive FROM STORE.elg
    462.1 KB · Views: 6
  • QUICK PULL MUDCAT.elg
    26 KB · Views: 12
Last edited by a moderator:

Stapl3

15+ Year Contributor
1,597
167
Oct 27, 2004
MI, Michigan
None of those have a WOT pull through peak VE, so not really much of any information to work from. I do see on the last log that the target is 9.3 and its at 10.4, so,,, I won't type it again.

A log showing a committed WOT pull in 2nd or 3rd or 4th to over 6k rpm is needed. If you're concerned over knock via timing, just put the sliders down a few notches for now.
 

Tony J

Proven Member
32
6
Oct 2, 2020
Belleville, ON_Canada
None of those have a WOT pull through peak VE, so not really much of any information to work from. I do see on the last log that the target is 9.3 and its at 10.4, so,,, I won't type it again.

A log showing a committed WOT pull in 2nd or 3rd or 4th to over 6k rpm is needed. If you're concerned over knock via timing, just put the sliders down a few notches for now.
I knoooow I knooow but that's better than 11.1... I only have about an hour to spend a night or my spouse is about to disown me haha.. little more tomorrow perhaps..

I'm trying to do enough to make a difference but not send anything out of wack.. and also not have to call the tow truck LOL
 

llxkevinxll

10+ Year Contributor
489
231
Jan 29, 2012
Milwaukee, Wisconsin
another pull from tonight with brief knock. not sure why this is happening so infrequently... sure if it happens everytime somethings surely off but its so intermittent and only seems to be when i hammer on it... if i eaaaaaase into the pull theres 0 knock..... or soo thats how it seems. Im not ripping through the gears but would like to come spring LOL

Afr #s seem to be closer too in each of the logs i posted.
The AEMWBGauge flat lining at 10.4 is odd. Even a perfectly setup car is going to have some waivering in this reading. Partially due to the chaotic nature of the exhaust stream and oscillations in the 5V output from the controller + ECMLink sensing error. The signaling appeared more natural in past pulls but this seems odd to me.

EDIT: I looked through all of your logs, the sensor never reads below 10.3. It runs into a wall and unnaturally flattens. I think this is more obvious now as you added fuel back in with the change to Global %. Did you happen to watch the gauge face during any of the pulls? Your setup in ECMLink seems normal for using an AEM WB controller. It is just a preset. But the way signal appears to be clipping at 10.3/10.4 is strange. May want to check the alignment between the gauge reported AFR and ECMLink sensed AFR.
 

Tony J

Proven Member
32
6
Oct 2, 2020
Belleville, ON_Canada
The AEMWBGauge flat lining at 10.4 is odd. Even a perfectly setup car is going to have some waivering in this reading. Partially due to the chaotic nature of the exhaust stream and oscillations in the 5V output from the controller + ECMLink sensing error. The signaling appeared more natural in past pulls but this seems odd to me.

EDIT: I looked through all of your logs, the sensor never reads below 10.3. It runs into a wall and unnaturally flattens. I think this is more obvious now as you added fuel back in with the change to Global %. Did you happen to watch the gauge face during any of the pulls? Your setup in ECMLink seems normal for using an AEM WB controller. It is just a preset. But the way signal appears to be clipping at 10.3/10.4 is strange. May want to check the alignment between the gauge reported AFR and ECMLink sensed AFR.
My gauge actually read 10.0, if im understanding everything @Stapl3 was saying then I just need to keep adjusting global or fuel at the regulator until my afr est and gauge are dead on.

I wanted it closer like this log its much closer 10% instead of 20, so I feel better changing my target afrs from 9.3 to say 10.8
 

llxkevinxll

10+ Year Contributor
489
231
Jan 29, 2012
Milwaukee, Wisconsin
My gauge actually read 10.0, if im understanding everything @Stapl3 was saying then I just need to keep adjusting global or fuel at the regulator until my afr est and gauge are dead on.

I wanted it closer like this log its much closer 10% instead of 20, so I feel better changing my target afrs from 9.3 to say 10.8
Your next steps are correct. But if you are using the sampled AEMWB signal in ecmlink to make these decisions, I'd question its accuracy or setup as it gets clipped at 10.4. As you noted it read richer on the gauge face but ecmlink didn't display that. It got stuck at 10.4. I'd look at the controller output calibration or check the wiring maybe. Not sure what would cause it to clip like that.
 

Stapl3

15+ Year Contributor
1,597
167
Oct 27, 2004
MI, Michigan
I agree, that is kinda weird with the wideband doing that. It's puzzling because it works elsewhere. Either something is up with it, or the SD table is damn perfect.

You could do something rather extreme with the fueling, like a rather drastic global change or a complete full turn on the afpr to see what happens. If nothing happens you would just undo the action and then safely assume the sensor is funky.
Or you could recalibrate the o2 sensor.
Or you jump ship and buy a LC1.

I can't think of any AEM part that is "good" for our cars for some reason, and I do remember over all the years people avoiding AEM stuff on the 'link forum. For some reason their parts just doesn't jive with our cars, spare the standalone which I admittedly have no experience with.

Now that I relook at your mod list, I actually ran 1250s at 40psi base for a while! Wild. I'll fire up the old laptop and give you my exact settings.
 

Tony J

Proven Member
32
6
Oct 2, 2020
Belleville, ON_Canada
I agree, that is kinda weird with the wideband doing that. It's puzzling because it works elsewhere. Either something is up with it, or the SD table is damn perfect.

You could do something rather extreme with the fueling, like a rather drastic global change or a complete full turn on the afpr to see what happens. If nothing happens you would just undo the action and then safely assume the sensor is funky.
Or you could recalibrate the o2 sensor.
Or you jump ship and buy a LC1.

I can't think of any AEM part that is "good" for our cars for some reason, and I do remember over all the years people avoiding AEM stuff on the 'link forum. For some reason their parts just doesn't jive with our cars, spare the standalone which I admittedly have no experience with.

Now that I relook at your mod list, I actually ran 1250s at 40psi base for a while! Wild. I'll fire up the old laptop and give you my exact settings.
That is funny. 1250 at 40 psi seem to be a common.

I'd love to see some settings. I've had the aem, this same gauge installed on 4 cars and never had an issue. If anything it's off .4 at most but is very close to actual gauge.

At this point I feel I'm close, with a small adjustment needed at global.

The cars much richer down low and into boost then previously and I'd rather work with too rich than too lean.

Will keep adding fuel and doing pulls to see how close I can get afr est but 10% seems to b the best I've had it
 

1990TSIAWDTALON

Moderator
9,869
5,747
Nov 14, 2013
Independence, Kansas
Ground the gauge at the same point as the ECU is. My tuner had me do that and it lined up the gauge with link. He said that was important. Just passing it on.
Proceed!
 

AWD-Tony

Supporting Member
6,800
3,688
Sep 11, 2017
Cincinnati, Ohio
When you change this section in DA openloopthreshold to the values you have, it's for ONLY cruise tuning to keep the car in closedloop longer and should not be doing any WOT pulls.

You must be logged in to view this image or video.



Also, someone changed the fastidleiscpos settings for some reason.

You must be logged in to view this image or video.
 

danl

20+ Year Contributor
721
117
Apr 8, 2002
Severn, Maryland
The aem wideband does not have a strong enough driver for the Mitsubishi ecu. My aem gauge will sometimes work, other times it will not. It is a well documented issue on the dsmlink site. First check that you have a good SENSOR ground. If that is ok and the wideband gauge reads ok, but the ecu numbers do not align, it could be the driver issue.
 

Tony J

Proven Member
32
6
Oct 2, 2020
Belleville, ON_Canada
They're known to have problems when you go full rich. I've got to pull some timing and change my target afr from 9.3 to 10.8 at wot. Should be good to go,

My car at full boost likes it, getting there is sometimes a .4 knock here or there.
 

habitatguy187

10+ Year Contributor
3,510
183
Aug 20, 2008
Indianapolis, Indiana
So far I’ve had the innovate mtxl, the lc2, and the aemx wideband. All 3 grounded to ecu ground. The aem is the closest to reading accurate but is still off a tad. It’s my belief that If you want it to be accurate, you need to set it up as linear wideband and narrow the operating range.
 

Tony J

Proven Member
32
6
Oct 2, 2020
Belleville, ON_Canada
So far I’ve had the innovate mtxl, the lc2, and the aemx wideband. All 3 grounded to ecu ground. The aem is the closest to reading accurate but is still off a tad. It’s my belief that If you want it to be accurate, you need to set it up as linear wideband and narrow the operating range.
Right now the only thing I notice is a misread of about .4 from ecmlink to gauge but at wot its onto of eachother so I'm not worried
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic 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

Top