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 JNZ Tuning
Please Support Fuel Injector Clinic

Resolved Starts, idles, dies shortly after. New motor

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.

Street9s

Proven Member
64
15
Apr 26, 2017
Pasco, Washington
New motor, 2.4 block cam gears have been adjusted properly. walboro 255, 875cc injectors. Ecmlink.

First startup went well. Idled fine for about 6 or 7 minutes while I let it warm up and check for leaks. Shut her down. Then fired it back up to take it for a drive, clutch didn't disengage. Fixed that. Started it up, went to go break it in and it sputtered and died.

Now what it does is it starts up, idles like it's perfectly fine, then dies after a few seconds. Attempts to start after that just stumble and don't really get a idle.

I've tried: cas, checked ecu, added an afpr.

I have other mods and I just don't feel they pertain to the scenario. I've got quite a bit of experience with these cars as well as cars in general. I just don't know what could be causing my problem. The part that really stumps me is that it ran so well initially, then just took a crap.

Any questions, feel free to ask and I will be happy to answer.
 
That did it. Now it dies with any throttle input. TPS voltage is way off. Time for a new one.

TPS volts/range is important for sure....but that is more about squeezing every ounce outtta your setup. TPS wouldn't cause it to die with throttle input.

Dying with any throttle input still sounds deadtime/global related to me....especially since you changed it and it fixed the first problem. You have swung to far the other way now.

Deadtime is extremely important at idle (and cruise later on) due to the fact it is a large % of the total fuel injector pulse width. Mine is so twitchy that if I change it even 5ms it will die....5ms the other way and it'll run but dies on throttle input.

If you can post a log of it running (warmed up if it'll stay running that long) and let us a take a look we can see where you're at.
 
Here's the log. I've tried about 2000 different values for the deadtime global. When I max that out, I upped the entire scalar graph. Things seemed to get better when i got into the 3-4000 ms range. But there's always a blip at 1500 or so, but beyond that it revs freely. Also, if I hold down the throttle slightly it will surge, and I put that in the log.
 

Attachments

  • log.2017.08.05-01.elg
    86.7 KB · Views: 26
Once the afr est is matching the actual afr, I should have a proper dead time or at least close, correct? I also assume the greatest change in actual afr will occur when I am close as well. Does that sound right?
 
After further troubleshooting I have discovered that I am only getting ~1.16v at the 5v source for both the TPS and the MAP. I checked the board for a burned sensor ground, it turned out fine. I'm also getting accurate air temp and coolant temp, so the sensor ground is definitely not the cause. What could cause less than 5v at the sensors?
 
After further troubleshooting I have discovered that I am only getting ~1.16v at the 5v source for both the TPS and the MAP. I checked the board for a burned sensor ground, it turned out fine. I'm also getting accurate air temp and coolant temp, so the sensor ground is definitely not the cause. What could cause less than 5v at the sensors?

Hmmm.....that is weird. The MAP would mos def cause an issue.
 
Since the only thing responsible for the 5v system is the ecu, i decided to swap in the ecu from my other car. Have a full 5v now and it revs freely provided I give it a lot of dead time. In the 1300us range. Is this too much? I can't seem to get it to lean out or change the add at all really using the VE table. If I bring the latency back down, it will do the same "blip" thing as before.
 
Since the only thing responsible for the 5v system is the ecu, i decided to swap in the ecu from my other car. Have a full 5v now and it revs freely provided I give it a lot of dead time. In the 1300us range. Is this too much? I can't seem to get it to lean out or change the add at all really using the VE table. If I bring the latency back down, it will do the same "blip" thing as before.

Yea, for 875 injectors that does kinda sound like a lot.

What was the voltage deadtimes for those injectors? Are they the right in ecmlink in direct access?

Can you post a new log from the new ecu then?
 
They're set at what my buddy sent me in his tune file. Then I tweaked the global dead time from there. I'll check fuel pressure and set the scaling table to whatever delphi has for the specs. Meanwhile I have to send an ecu out..
 
They're set at what my buddy sent me in his tune file. Then I tweaked the global dead time from there. I'll check fuel pressure and set the scaling table to whatever delphi has for the specs. Meanwhile I have to send an ecu out..


The global fuel % scaling slider calculator thingy gets you close....but you need to manually dial it in to get deadtime/global % right.

You need to play with the deadtime and global fuel % while watching: AFREst vs WB PLUS BoostEst vs MAP PLUS AirflowPerRev. While watching all these get the AirFlowPerRev to 0.25g/rev (i think but dont remember off hand....check ECMLInk wiki for the actual value) using the MAFComp slider (or VE table if your SD) AND global fuel % AND the global deadtime. It takes some back and forth between all values but once dialed in makes the rest of tuning much more accurate.

Once AFREst matches WB PLUS BoostEst matches(closeish) MAP PLUS AirflowPerRev is in the right range your deadtime and global fuel% will be calibrated. Should run a gajillion times better once the deadtimes are very close.


That being said it sounds like it was a voltage issue with the MAP sensor via the ECU 5V out sucking balls. :)
 
That's some dam good advice. I suppose on my other car things were close enough that it didn't matter. It made 645 just the other day. I'll get those ironed out provided it's not too searing hot out when I get home..
 
Fixed. Swapped in the ecu from my other car and I got 5v at the tps and map sensor. No need to replace the tps now... IC101 seems to be the culprit. Thinking about repairing it myself. Although it'll only be 40$ to send it in.

Thanks for the help!
 
Meh, I'd send it in. They fixed my driver and had it back to me in like 5 days from me shipping it.

Wow, that's awesome. I asked them and they said it would be the minimum charge to get it done. Plus they'll probably go through the rest of it and fix that if need be as well.
 
Right on. I've ordered 2150 injectors from them before and they didn't include the pigtails as they said it would. So they sent me the PNP harnesses for no charge at all. Seems like a great company to deal with. Nice to have that for our community.
 
Fixed. Swapped in the ecu from my other car and I got 5v at the tps and map sensor. No need to replace the tps now... IC101 seems to be the culprit. Thinking about repairing it myself. Although it'll only be 40$ to send it in.

Thanks for the help!


Awesome.

Can you mark this thread answered then? So many open threads out there... :)
 
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