ECMlink Speed Density Switch No Start

GoldNugget90 · Aug 14, 2026

1990 Eclipse GSX switched to speed density and now won't start. Briefly idles low and rough for a second before cutting out. I did boost leak check and found no issues, had and fixed fuel leak at AFPR fitting and oil leak at turbo feed line. Mechanically no other observed issues.

Car ran perfectly with NO SPEED DENSITY & Walbro 255 rewired, AFPR, MBC, O2 housing back exhaust 3", DSMLink at all stock parameters, and stock 37 psi fuel pressure, AEM boost, wideband O2, water temp, oil pressure gauges

Car now won't run WITH SPEED DENSITY & after MHI 16g upgrade, FMIC install, new intake pipe, catch can, 1200CC injectors high-z, resistor pack delete, ECMLink SD cable, GM IAT, Omnipower 4bar, EGR block off, emissions delete
It was a lot of upgrade at once along with the SD conversion but I'm reasonably confident of correct installs. Literally any info is helpful, I'm struggling.

Log 1 is first crank attempt
Log 2 is second crank attempt
Temporary FMIC mounting solution

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

curt-s · Aug 14, 2026

Have you tried artificially inflating your start/idle VEs just to see if it will run?

SixBolt_16G · Aug 15, 2026

You should post the actual .elg log file so that we can see your other settings. PDFs dont tell the full story.

GoldNugget90 · Aug 15, 2026

curt-s said:
Have you tried artificially inflating your start/idle VEs just to see if it will run?

No not yet. I wasn't sure the best place to start troubleshooting, but I'll give this a shot and see what happens. Thanks for the starting point/suggestion.

SixBolt_16G said:
You should post the actual .elg log file so that we can see your other settings. PDFs don't tell the full story.

Added to original post

1990TSIAWDTALON · Aug 15, 2026

I looked the settings over and didn't notice too much, but what I did notice, I took a picture of. It looks like you need to charge the battery, I am surprised it ran at 8.? Volts. You've probably been using the battery alot. Charge it up good and hot. :thumb:

The global was a bit off for 1200s but I figured that was your choice. A/Fs look rich but maybe that's you. TPS volts are way below minimum of .63. Just some stuff I see. Maybe you have your WB welded into a different place and are still using the front O2 also. I just saw it needed a pin assignment is all.

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

GoldNugget90 · Aug 15, 2026

1990TSIAWDTALON said:
I looked the settings over and didn't notice too much, but what I did notice, I took a picture of. It looks like you need to charge the battery, I am surprised it ran at 8.? Volts. You've probably been using the battery alot. Charge it up good and hot. The global was a bit off for 1200s but I figured that was your choice. A/Fs look rich but maybe that's you. TPS volts are way below minimum of .63. Just some stuff I see. Maybe you have your WB welded into a different place and are still using the front O2 also. I just saw it needed a pin assignment is all.

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

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

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

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

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

You're right, I have been connected to the battery without a start, I'll get it charged before proceeding. I'll check voltage also, however, I've read a little on this and .48-.52v may be OK for a 1g? The global fuel is as ECMLink calculated, and my best guess at a deadtime. My wideband is in the rear O2 location (downpipe), front O2 is stock. I've updated the PC-side to reflect stock.

1990TSIAWDTALON · Aug 15, 2026

I have three 1g's. All run Ecmlink and all are set at .63v at idle. Sometimes that will go up .01 when the motor starts.
I would suggest this for the TPS as it makes manual adjustments much easier.
DSM TPS Bolt Mod for easy adjustments | DSMtuners.com https://share.google/Bxo1b9SDYc34RPc8U

Stapl3 · Aug 15, 2026

You're going to have to work backwards.

Install the old intake pipe on with the MAF. Plug the MAF into the stock connector. Go on the MafComp tab and change it from SD to the MAF.

If no luck there, I would put the old injectors back in with the old fuel settings.

Tony1094 · Aug 16, 2026

I think you also need to check the box on your ECU input tab for locking intake temp and baro pressure.

steve · Aug 16, 2026

Tony1094 said:
I think you also need to check the box on your ECU input tab for locking intake temp and baro pressure.

According to ECMTuning those default to locked when using SD but it's a great idea to be explicit about it to remove any confusion.

Tony1094 · Aug 16, 2026

What brand injectors did you get as well? Have a data sheet for them? Looks like you have stock injector voltages on the direct access tab and then added 400 global. Did you get those numbers from somewhere specific? May be drowning in fuel? Very hard to tell when your wideband is delayed like mine on startup.

GoldNugget90 · Aug 16, 2026

Stapl3 said:
You're going to have to work backwards.

Install the old intake pipe on with the MAF. Plug the MAF into the stock connector. Go on the MafComp tab and change it from SD to the MAF.

If no luck there, I would put the old injectors back in with the old fuel settings.

Was trying to avoid this step, but can if necessary. Different valve cover with it now going to a catch can so this becomes a big step backwards. Swapping back to old injectors isn't too bad. Will be a few days before I can get back into the garage.

Stapl3 · Aug 16, 2026

GoldNugget90 said:
Was trying to avoid this step, but can if necessary. Different valve cover with it now going to a catch can so this becomes a big step backwards. Swapping back to old injectors isn't too bad. Will be a few days before I can get back into the garage.

Forget the PCV and all that. Just see if it will fire up with the MAF plugged in and the old intake pipe connecting it to the turbo.

GoldNugget90 · Aug 20, 2026

Tony1094 said:
What brand injectors did you get as well? Have a data sheet for them? Looks like you have stock injector voltages on the direct access tab and then added 400 global. Did you get those numbers from somewhere specific? May be drowning in fuel? Very hard to tell when your wideband is delayed like mine on startup.

Made an adjustment to what I saw on ECMTuning, for 1150s with 330 msec, and adjusted. I've been playing with the global deadtime a bit, I think I was around 200-225 in the log posted here, where it was actually running but poorly and high idle. Wasn't until after that I saw that a few key values were not being captured. I've since not been able to replicate. Will try the reversion back to MAF this weekend as Stapl3 suggested when the garage isn't so hot, and boost the TPS volts.

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

Tony1094 · Aug 21, 2026

I think the numbers aren't working out due to your base fuel pressure. Pretty sure the FIC 1200s are only rated 1200 @ 43.5psi.

If you want to try out some numbers before you take everything apart:
Global fuel: -59.4%
Global deadtime: 0

On the direct access injbattadj tab:
7v 2850
9v 2020
12v 1200
14v 930
16v 775
19v 540

GoldNugget90 · Sunday at 7:12 PM

Tony1094 said:
I think the numbers aren't working out due to your base fuel pressure. Pretty sure the FIC 1200s are only rated 1200 @ 43.5psi.

If you want to try out some numbers before you take everything apart:
Global fuel: -59.4%
Global deadtime: 0

On the direct access injbattadj tab:
7v 2850
9v 2020
12v 1200
14v 930
16v 775
19v 540

Finally got a few minutes, tried your approach. Definitely a step in the right direction so thank you for the recommendation. Couple logs attached with those adjustments. Still won't quite stay running. "-xx" in the logs are the global fuel adjustments. What's the more worthwhile effort, continuing to toggle the global % or injbatteryadj?

Stapl3 · Sunday at 9:38 PM

GoldNugget90 said:
What's the more worthwhile effort, continuing to toggle the global % or injbatteryadj?

At this exact moment, Injbatteryadj. Actually I would leave the 'shape' in Injbatteryadj alone and change only deadtime on the fuel tab. Deadtime and injbatteryadj is one in the same. The only difference is injbatteryadj changes per volt. Adding 100 to deadtime on the fuel tab is the exact same thing as grabbing all sliders on Injbatteryadj and raising all of them 100.

Global % is determined by a WOT pull at peak VE. Injbatteryadj is used at idle and kinda at cruise.

It is a bit of chasing your tail setting the car up. You have to get the car 'normal', which you are doing. Then you have to do a WOT pull to determine if your global % is correct. 99/100 times you will have to do a change with the global. Once your global fuel % is changed to correct you will have to go back to the injbatteryadj and re-adjust that.

The sign of a properly calibrated fuel system on a DSM is when WB=AFRatioEst, at peak VE (~5500rpm), and your mafcomp slider is at zero or your SD table value is 100. If you can achieve that, you are gold. You can then do whatever you want to injbatteryadj to get idle and cruise so WB=AFRatioEst, and basically whatever you want to the MAFComp sliders / SD table to get WB=AFRatioEst. Your objective at this point should be getting the car good enough to hold idle, drive, and do that so needed WOT pull to a minimum of 6000rpm. That should be your only goal as you sit. And I know it is daunting as the car is questionable at idle so a WOT pull seems insane, but it is a required part of the process.

Tony1094 · Sunday at 10:06 PM

Stapl3 said:
At this exact moment, Injbatteryadj. Actually I would leave the 'shape' in Injbatteryadj alone and change only deadtime on the fuel tab. Deadtime and injbatteryadj is one in the same. The only difference is injbatteryadj changes per volt. Adding 100 to deadtime on the fuel tab is the exact same thing as grabbing all sliders on Injbatteryadj and raising all of them 100.

Global % is determined by a WOT pull at peak VE. Injbatteryadj is used at idle and kinda at cruise.

It is a bit of chasing your tail setting the car up. You have to get the car 'normal', which you are doing. Then you have to do a WOT pull to determine if your global % is correct. 99/100 times you will have to do a change with the global. Once your global fuel % is changed to correct you will have to go back to the injbatteryadj and re-adjust that.

The sign of a properly calibrated fuel system on a DSM is when WB=AFRatioEst, at peak VE (~5500rpm), and your mafcomp slider is at zero or your SD table value is 100. If you can achieve that, you are gold. You can then do whatever you want to injbatteryadj to get idle and cruise so WB=AFRatioEst, and basically whatever you want to the MAFComp sliders / SD table to get WB=AFRatioEst. Your objective at this point should be getting the car good enough to hold idle, drive, and do that so needed WOT pull to a minimum of 6000rpm. That should be your only goal as you sit. And I know it is daunting as the car is questionable at idle so a WOT pull seems insane, but it is a required part of the process.

Not to hijack a thread or anything at all but I'm still working on tuning my own car on SD and getting it dialed in, I just want to ask about what you said about peak VE. I've been adjusting my global fuel and SD scaling around trying to get BoostEST and my MAP sensor to read +/- 1PSI of each other, I was under the impression if they matched your scaling was accurate, is that not correct?

Tony1094 · Sunday at 10:14 PM

GoldNugget90 said:
Finally got a few minutes, tried your approach. Definitely a step in the right direction so thank you for the recommendation. Couple logs attached with those adjustments. Still won't quite stay running. "-xx" in the logs are the global fuel adjustments. What's the more worthwhile effort, continuing to toggle the global % or injbatteryadj?

I would try what Stapl3 suggested and maybe try adding 50 at a time to global deadtime and see how it behaves. I might be crazy but the -59 log looks like it was going to idle surge if it kept running. Mine did that as well, had to drop idle portion of my timing table to 5* to get it slow down, but I'm still chasing a cold idle surge myself after an FIAV bypass install :cry:

Stapl3 · Sunday at 10:30 PM

Tony1094 said:
I've been adjusting my global fuel and SD scaling around trying to get BoostEST and my MAP sensor to read +/- 1PSI of each other, I was under the impression if they matched your scaling was accurate, is that not correct?

BoostEst is just a tool where the equation assumes 100% VE, so it is a good sanity check vs. actual boost when using a MAF and used at as close as possible to 100% [peak] VE.

The SD table is telling the ECU what % of VE you are at. If you continue your method of looking at BoostEst while using SD, you will realize over time every cell ends up at a value of 100.

GoldNugget90 · Monday at 7:10 PM

Tried the global deadtime adjustments, from zero up to 100, at various different global fuels. This log is at 100 ms and -59.4%. Pretty terrified to go try and drive it in this state, but at least it stayed running >1 minute until I shut it off. Have to install a coolant overflow tank before hitting the road, but any other red flags or suggestions before going for the 6k rpm test?

1990TSIAWDTALON · Monday at 7:46 PM

I don't know if its been said but I sit in my car, idling, and play with the global first to try to get my LTFTs and STFTs close to zero. They never are so I reduce or add global a bit and look at my Ecmlink log that I am running and look at LTFTs. I adjust to get it close, or within +/- 5% then I fiddle with deadtimes for STFTs to bring them within 5% +/-. A note here, YOU CAN RUN NEGATIVE DEAD TIMES. It should idle pretty good and drive ok but the WOT will not be correct. I haven't got that perfected when I mess with my settings yet.
Marty

SixBolt_16G · Monday at 8:10 PM

Your Front O2 sensor went to the bathroom at 7 seconds and never came back to the party...?

1990TSIAWDTALON · Monday at 8:20 PM

Thanks @SixBolt_16G

GoldNugget90 · Monday at 8:29 PM

SixBolt_16G said:
Your Front O2 sensor went to the bathroom at 7 seconds and never came back to the party...?

Haha, Irish goodbye'd right outta there. Kinda weird, I have about 7 logs from the same time. 5 of them start fine and then taper to zero volts, guessing it's getting way too lean.

The Central Hub for DSM Community and Information

ECMlink Speed Density Switch No Start

GoldNugget90

Attachments

curt-s

SixBolt_16G

GoldNugget90

1990TSIAWDTALON

Attachments

GoldNugget90

1990TSIAWDTALON

Stapl3

Tony1094

steve

Tony1094

GoldNugget90

Stapl3

GoldNugget90

Attachments

Tony1094

GoldNugget90

Attachments

Stapl3

Tony1094

Tony1094

Stapl3

GoldNugget90

Attachments

1990TSIAWDTALON

SixBolt_16G

1990TSIAWDTALON

GoldNugget90

Similar threads

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds