The Central Hub for DSM Community and Information

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. This is where the DSM platform history is documented and archived. Log in to help us in our mission, and to remove most ads from the browsing experience.

Need Help With DSMlink Cruise/Idle Tune

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

zckry

10+ Year Contributor
78
0
Nov 1, 2008
North, Texas
Guys, I am getting really really frustrated trying to tune my GSX on closed loop.
I have been following the directions perfectly on the DSMlink Tuning Guide (http://www.dsmtuners.com/forums/articles-tuning-ecu/262460-basic-dsmlink-tuning-guide.html) yet I can't seem to either get my fuel trims right or my airflow right.
After doing another log last night, I found my timing is super high on cruise.
And I'm getting small counts of knock because of the timing (not to mention the motor feels like crap).
I haven't messed with any MAF comp values above 150 because the guide never mentioned it, I don't know if I need to or not.

I need your guys' help because DSMlink is still learning process for me.

I attached the log I had last night, mostly cruise with 2 WOT pulls near the end.

Mods:
Galant VR4 6-bolt motor swap
Ported E3 exhaust manifold
B16G Turbo
2.5" intercooler piping
Spearco FMI
Injen intake
Forced Performance MBC
3" straight TBE
ACT 2600 PP
GReddy Oil Cooler
Fluidine radiator
580cc PTE injectors
Walbro 255 fuel pump
Stainless fuel line from pump/tank to rail
AN fittings at fuel rail
Paxton FPR
DSMLink v2
AEM Uego wideband

Any help is greatly appreciated.

I'm also running the stock 2G MAF
 
First off, you said you never adjusted your MAF comp beyond the 150HZ level, but they're ALL above zero across the board. You should probably reset that to zero first, then adjust your global and deadtime to get your trims more in line. Secondly, click the ALS/Knck tab, and on the knock sensor control, activate above 2500, and TPS above 25% or 30%. You don't have to worry about knock at 12% throttle.

After you have the MAF comp set to zero, check that your deadtime and global are set to the reccomended starting point for your injectors, and follow this portion of the guide:

Closed Loop Tuning (Cruise/Idle):
1. Setup DSMLink to Log the following (if available):
a. TPS
b. STFT
c. LTFT Lo
d. LTFT Mid (2G) or LTFT Hi (1G)
e. MAFRaw
f. Wideband
g. Boost sensor (GM 3 bar, AEM, etc.)
2. Turn on vehicle and let warm properly.
3. Drive normally, cruising for at least 10 minutes - Do not go WOT
4. Let car idle for 5 minute before shutting it off - Stop logging
5. Choose your method of closed loop tuning by determining first if AirflowPerRev is at an acceptable value at idle. If AirflowPerRev is not acceptable, begin tuning using Method 1 until it is at an acceptable value. If AirflowPerRev is acceptable, proceed to Method 2.


Method 1 – Airflow Metering:
1. Turn on vehicle and let warm properly.
2. Begin logging. Drive normally, cruising for at least 10 minutes - Do not go WOT
3. Let car idle for 5 minute before shutting it off - Stop logging
4. Compare LTFT for idle (Lo) and cruise (Mid) operations – Strive to get as close to 0% as possible.
5. If either is greater than +/- 5 from 0%, proceed to step 6. If not, and your happy with your %, you may stop.
6. If LTFT Lo is negative, decrease the 50 Hz slider in the amount of that value. (i.e. If LTFT Lo is -10%, decrease the 50 Hz slider from 0% to -10%).
7. If LTFT Mid is positive, increase the 150 Hz slider in the amount of that value. (i.e. If LTFT Mid is 10%, increase the 150 Hz slider from 0% to 10%).
Or: (From method 1) If LTFT Lo is greater than LTFT Mid, increase dead time. If LTFT Lo is less then LTFT Mid, decrease dead time.
8. If adjustments are made, return to step 1.


Post up a log again after this is done.
 
Last edited by a moderator:
Ps. You did do this and make sure everything is within these parameters first right?

Before You Start:
• Verify that Base Timing is correct
• Verify that Base Fuel Pressure is correct (1g manual = 37 psi; 1g auto = 43.5;
• 2g = 43.5 psi)
• Verify that there are no boost leaks
• Ensure all other components are working correctly
• Swap plug wires to account for different CAS (97-99)
• Approximation of desired values: (Some of these must be obtained manually and some via DSMLink. Others are simply values to monitor and keep within given parameters.)

A/F Ratio (closed loop) 14.7
(open loop) 11
AirFlow (compare to specific turbo output)
AirFlowPerRev (at idle) M/T = .27 10% (.24 - .30)
A/T (neutral) = same as M/T
(drive) = .31 10% (.28 - .34)
CoolTmp x>179°, sufficient operating temp. to tune
if: x<206°, full timing
206°<x<224°, -1°
x>224°, -2°
IdleSw (car off / ecu on) 1
IntTemp 34°<x<84
ISCPosition 30 (at idle)
LearnedIdleAdj 140 (at idle)
LTFT Lo x&#8594;0, (0 5)
LTFT Mid x&#8594;o, (0 5)
Front02 cycles around .5 (.05-.95)
RawThrotPos 32
ThrotPos if: idle, 0
WOT, 100
TPSvolts .63 (when RawThrotPos = 32)
 
Sorry for the tons of posts, but I see that you have your global (-17) and deadtime (180) set correctly for your PTE 580 injectors, but that's just a starting point. After you turn your MAF comp to zero, you should be logging some fuel trims closer to zero, but usually you'll need to fine tune global and/or deadtime to make both LTFT low, and LTFT hi zero out.
 
I'd like to point out that by downloading the v.3 software and running v.2 logs, the CombinedFT function can be used to aid in the setting up of the airflow points/values.

It helped me immensely.
 
What I mean was, I got frustrated and unhooked the battery to reset DSMlink to all the original settings and haven't messed with the MAF comp since last night when I reset it.

I guess the original settings have the MAF set for a hacked 2G MAF even though it's not a hacked 2G MAF.

Do you mean I should set every MAF hz (50, 150, 250, 400, etc.) to 0 and start from there or do you mean after 50hz and 150hz I should change the rest to 0 (250, 400, 800, etc.)and start from there?

And yes, I have verified all of those minus the timing (I don't have a timing light).
The only boost leak I have is tiny one near the throttle body that't not going to make that much of a difference.
 
I'd like to point out that by downloading the v.3 software and running v.2 logs, the CombinedFT function can be used to aid in the setting up of the airflow points/values.

It helped me immensely.

I thought that was only available on the actual v3 chip, not the software?
 
Nope. the software will do it, you'll just have less MAF points to adjust with v.2
 
Sorry for the tons of posts, but I see that you have your global (-17) and deadtime (180) set correctly for your PTE 580 injectors, but that's just a starting point. After you turn your MAF comp to zero, you should be logging some fuel trims closer to zero, but usually you'll need to fine tune global and/or deadtime to make both LTFT low, and LTFT hi zero out.

I realize this, I just reset everything last night out of frustration so I can start from there and not my messed up idle tune I had going on.
 
Also just found out that the lower honeycomb is removed but that isn't considered completely hacked is it?
I thought it had to be all but the main honey.
 
Oh crap, I just noticed that your engine only went up to operating temp for a short time, (only 200 seconds in the middle of the log), then went cold again. You can't adjust your fuel trims unless the engine is warm enough. You need a thermostat.


It mentions this in the BEFORE YOU START section of the tuning guide:

(CoolTmp x>179°, sufficient operating temp. to tune)

You really should reread that guide, it's really good.
 
Oh crap, I just noticed that your engine only went up to operating temp for a short time, (only 200 seconds in the middle of the log), then went cold again. You can't adjust your fuel trims unless the engine is warm enough. You need a thermostat.


It mentions this in the BEFORE YOU START section of the tuning guide:

(CoolTmp x>179°, sufficient operating temp. to tune)

You really should reread that guide, it's really good.

Sorry, believe it or not, it's been freaking cold here in Texas for the past week.
And I have a decent thermostat.
I let my car idle for over 5 mins before I did that run.
I'm thinking my fluidine radiator is keeping it cooler.
And I've been reading that guide so much I practically have it memorized, I couldn't do anything about the coolant temp,
that was the warmest I could get it without idling it for over 10 mins.
 
Also just found out that the lower honeycomb is removed but that isn't considered completely hacked is it?
I thought it had to be all but the main honey.

No, that's not completely hacked, but it may let in slightly more air than you are metering. Fully hacked MAF's have the sides cut out too.

I would just set the MAF comp at zero anyway for now, and get the thermostat changed. As soon as you start to criuse, the temp drops to 140F and that's way too cold for the trims to adjust themselves. Once you have your temp in line, you can log your Long term trims at criuse, let it idle for five minutes, then fine tune your global and deadtime accordingly.

After you get your LTFTs all close to equal and zero, then you can look for certain HZ levels where the STFT is high or low, and adjust for that.

For example: I got my LTFT low and hi near zero by adjusting global and deadtime, but when criusing at 150Hz-200Hz, 11% throttle, 55 mph, my STFT stayed at +16%. At all the other speeds and airflow levels my STFT stayed at or near 0%.

The symptom: After a while of criusing at +16% my fuel trim would reset to zero and my wideband would show lean. The ECU would gradually insrease STFT to +16% again and the cycle would continue.

The cure: I raised my 150Hz airflow slider, so it would give more fuel at 150Hz and therefore the STFT did not sit at the maximum limit for long enough to cause the "reset" that was leaving me running so lean. My STFT still hovers around +12% at 150Hz, but since it isn't pegged, it still maintains a proper A/F.
 
140 degrees is still too cold in my opinion. You would need to use a lot of offset. I would try a hotter thermostat. What temp do you have in it?

Won't raising the offset make your fan come on sooner and exasturbate the problem?

Ps. Sorry about the made up word. What I meant was, "Won't that add to the problem?"
 
Okay, I'm about to do a run since it's hotter out today so no adjustment should be needed.
I'll wait until it get to sufficient operating temp and I will drop my MAF comp's to 0 across the board and put the log up after the run and we can go from there hopefully.
 
Eh, I like the cold weather for the most part (we don't get much down here).
Okay, here we go looks like I need to adjust my MAF comp up some now.
Does anyone know why my timing is so high still though?
I have everything set to 0 across the board as far as timing goes.
Got it to tuning temp and raised my knock sensor control colt :thumb:
Now help me :p
 
Ok I really don't see any knock to be concerned about, the only time I see any, (only .7 degrees knock comp.) is at 14% throttle, and I think you're safe to ignore knock below 25% throttle. I've read this in posts from the creators of link, so I set mine to ignore it below 25% throttle too. The timing map allows for higher advance at the lower TPS levels, so I think you're worrying about nothing on that part. As long as your base timing is correct, I would trust it to do it's job timing your spark under normal criuse conditions.

I see now why your MAF comp was up a little above zero, you aren't showing enough airflow per rev at idle rpm. I would adjust that across the board to compensate for your partially hacked MAF. Airflow per rev should read about between .24-.30 at idle.
 
Here's a quote from The creator of link:

"The knock sensor is totally ignore by the ECU if the conditions specified in the knock control dialog are not met. That's how the factory logic works as well. It's just not used below about 2250 RPM from the factory. Same thing with the DSMLink controls, only it's adjustable and includes a TPS check.

Thomas Dorris"

Another:

"I would not call using the ECU->Knock dialog (within reason) a bandaid. The factory 2G ECU includes the same logic already. It ignores knock retard below 2200 RPM or so. The DSMLink dialog just provides the ability to add a TPS comparison and more adjustment to the factory value. Your engine is rather different now from what the engineers originally coded this ECU for...I would expect to need some adjustments of a few things.

Thomas Dorris"

As you can see from your log, this knock is happening at 2000 rpm, which is below the factory preset threshold for compensating for knock. This effectively makes your car hypersensitive to knock.
 
Or maybe it's phantom knock?

I guess it could be, but there's one spot at 84.39 seconds that looks like it might be real knock. Still, at such low RPM and throttle, I would never worry about such a tiny blip of knock.

The designers of the car never intended us to worry about knock below 2250 rpm. So apparently it's expected right?
 
Yeah, I hope it's just phantom knock, guess I'll find out in the long run haha.
I'm pretty damn close to 0% for both Lo and Mid LTFT after MAF adjustments.
BUT I can't seem to get the AirFlowPerRev close to .24 or higher without making my Lo LTFT lower significantly (to -12.5%).
Log for it is attached, let me know what I should adjust to stay close to the 0% yet get a better AirFlowPerRev.
 
Add Value - Be Respectful - No Trolling - No Misinformation - Participate Often!
Support Vendors who Support the DSM Community

Build Thread Updates

Latest Classifieds

Back
Top