2G 1995 Eclipse GST engine sputters when free revving

phant0mxx · May 30, 2026

I've never seen this behavior before and I'm at a loss.
Engine free revs fine up until ~5000 RPM, where it sputters (video linked below) and I can make it go past 5000 RPM if I very slowly press the gas pedal instead of quickly pushing it.
It also drives fine when there is no load on the engine. As soon as you add load or try to build boost, nothing happens, there is no cut, no jerking, nothing. It doesn't build boost and it doesn't keep accelerating, sort of hangs the revs in place until I let go of the accelerator pedal.

Video:

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

In the video you can see I accelerate 4 times.
1st time I do an under 5k RPM acceleration.
2nd time I go past 5k RPM to trigger the issue and hold pedal very little, under 2 seconds.
3rd time I go past 5k RPM again and this time I hold the pedal to accentuate the problem. This time the RPMs started dropping while sputtering, but sometimes the RPMs hold while sputtering, but it never continues to increase in RPM.
4th time I go beyond 5k RPM by very gently pressing the accelerator pedal instead of jabbing it.

Engine idles okay, minus a small oscillation of 50-75 RPM.
ISC works and catches the RPMs before settling back to idle speed.

Parts robbed from a running car:

PTU (Power Transistor Unit)
Ignition coils
Spark plugs
Spark plug wires
MAF
Blow-off valve

I need to find a known good ECU to test now... The donor has a non-PnP standalone so I couldn't grab it to test.

Smoke tested and no vacuum leaks.
Fuel pressure idling is ~35psi, when accelerating it goes up to ~45psi.
Pressure does not drop while the engine sputters or when driving and trying to load up the engine.
Injectors were cleaned flow tested and rebuilt with new filters.
Ignition timing when locked is at 5º BTDC, and unlocked ~8º BTDC.
Physical timing marks line up correctly.

ECU part number is MD305162 E2T61675 5322 E
ECU components replaced were IC104 and IC107 due to an internally shorted ISC that blew both integrated circuits.
Used original Mitsubishi M5269L ICs and reconstructed burnt PCB traces from ICs to ECU header pins 4, 17, and 18.
Caps look good, no leaks or corroded solder joints nor traces.

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

I should probably also explicitly say that there's no CEL.

I am open to hearing any suggestions or comments at this point.

dwb · Sunday at 2:11 PM

What's the mod list?
Gas gauge looks like you're running on fumes.
What's your your base fuel pressure without a vacuum/boost source?
Try a boost leak test as opposed to a smoke test.
Any open ports between the maf and turbo?

chrisman287 · Sunday at 3:19 PM

Shouldn’t fuel pressure be higher at idle? I also second the gas you’re empty empty. Which plugs and gap? More info on the car?

phant0mxx · Sunday at 6:17 PM

dwb said:
What's the mod list?
Gas gauge looks like you're running on fumes.
What's your your base fuel pressure without a vacuum/boost source?
Try a boost leak test as opposed to a smoke test.
Any open ports between the maf and turbo?

No mods, 100% original.
Fuel was topped off, same issue.
Base pressure is ~45psi without vac line.
No open ports between MAF and turbo, all lines are installed and clamped.
The smoke test was performed like a BLT, intake was pressurized to 15psi, no leaks.

phant0mxx · Sunday at 6:20 PM

chrisman287 said:
Shouldn’t fuel pressure be higher at idle? I also second the gas you’re empty empty. Which plugs and gap? More info on the car?

FSM calls for the same values I have on the car.
Fuel topped off, same issue.
All original car, no mods.

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

Plugs are brand new BPR6ES, gapped to 0.028"

phant0mxx · Monday at 11:21 AM

Put the scope on the cam and crank sensors (ECU side, pins 88 and 89), signals looks good.
This screenshot is the signal while the car is stuttering/sputtering passing 5k RPM free revving.

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

phant0mxx · Tuesday at 12:25 PM

Just tried another ECU from a friend's car and the problem is still present.
Took it out for a drive and no change.
The engine stops revving at 3000 RPM while under load and it holds there sputtering.

phant0mxx · Wednesday at 10:16 AM

I just unbolted the exhaust at the O2 housing, let it warm up, and revved it.
Same problem.
My last hope was that the exhaust was clogged and it was building so much back pressure that it didn't let me rev.
I've exhausted all I can think of.

dwb · Wednesday at 10:42 AM

Has this car ever run properly for you? If so, any chance you have been running a higher ethanol content fuel like e15 and have you inspected fuel injectors for debris or buildup?
Do you have anyway to log it?
Show us a clear pic of the engine bay. Maybe we can spot something off.
What's the age/condition of the fuel pump and filter? Have you inspected the pickup inside the tank?
Have you checked for diagnostic codes?
Does it smoke or smell rich?
Does the fuel system hold pressure after turning off the pump?

phant0mxx · Wednesday at 11:22 AM

dwb said:
Has this car ever run properly for you? If so, any chance you have been running a higher ethanol content fuel like e15 and have you inspected fuel injectors for debris or buildup?
Do you have anyway to log it?
Show us a clear pic of the engine bay. Maybe we can spot something off.
What's the age/condition of the fuel pump and filter? Have you inspected the pickup inside the tank?
Have you checked for diagnostic codes?
Does it smoke or smell rich?
Does the fuel system hold pressure after turning off the pump?

It has never ran good, all fuels locally have 12 to 15% ethanol.
Injectors were cleaned, flow tested, and rebuilt with new filters and orings.
I don't have any kind of logger, maybe I should invest in one, and the ECU from the car doesn't connect to OBD2 scanners because its one of those with OBD2 port but OBD1 protocol. I still have my friend's ECU in possession which is a true OBD2 ECU and can connect, so I'll let the car run with that computer and see if I pull any codes.

Bay pic:

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

I am not inclined to think the fuel pump is bad because I've tested the fuel pressure and it is within spec. Car idles and drives normally until I put some load into it and that's when the symptoms appear. Fuel pressure holds true under all conditions.
It does not smell rich.

Which logger would be recommended for me to get? Evoscan?

pauleyman · Wednesday at 11:28 AM

phant0mxx said:
It has never ran good, all fuels locally have 12 to 15% ethanol.
Injectors were cleaned, flow tested, and rebuilt with new filters and orings.
I don't have any kind of logger, maybe I should invest in one, and the ECU from the car doesn't connect to OBD2 scanners because its one of those with OBD2 port but OBD1 protocol. I still have my friend's ECU in possession which is a true OBD2 ECU and can connect, so I'll let the car run with that computer and see if I pull any codes.Bay pic:
You must be logged in to view this image or video.
I am not inclined to think the fuel pump is bad because I've tested the fuel pressure and it is within spec. Car idles and drives normally until I put some load into it and that's when the symptoms appear. Fuel pressure holds true under all conditions.
It does not smell rich.Which logger would be recommended for me to get? Evoscan?

All US 2g cars should be obd2

All stock yet you show a pic of a non recirculated BOV. What else do we not know about?

steve · Wednesday at 11:54 AM

pauleyman said:
All US 2g cars should be obd2

All stock yet you show a pic of a non recirculated BOV. What else do we not know about?

MD305162 isn't a USDM ECU and I bet the car isn't either.

phant0mxx · Wednesday at 12:12 PM

pauleyman said:
All US 2g cars should be obd2

All stock yet you show a pic of a non recirculated BOV. What else do we not know about?

Brother, I have the original blow off valve sitting on the passenger seat.
If you follow the thread I stated that I replaced the BOV because I thought it was opening under load, that's why I put that aftermarket one in with a stiff ass spring in it.
Why do you think that I am hiding stuff? I want to drive the car again, hiding shit would only hurt me.

Edit: I just looked at the thread again and see that I didn't mention it here... I must have mentioned the BOV to my brother or a friend and though I said it here. My bad.

steve said:
MD305162 isn't a USDM ECU and I bet the car isn't either.

Correct. It's a Brazil delivered Eclipse.

I've been pulling my hair out and overlooked the fact that ECU pin 92 has 0ohm to chassis ground.
It seems that there's a short to chassis ground somewhere in the sensor grounds, maybe that is what is causing my problems. I'll inspect further into this now...

Edit2: Not a sensor ground problem. I had partially disconnected the ECU plug, that's why I had a "short" to ground.

pauleyman · Wednesday at 1:07 PM

phant0mxx said:
Brother, I have the original blow off valve sitting on the passenger seat.
If you follow the thread I stated that I replaced the BOV because I thought it was opening under load, that's why I put that aftermarket one in with a stiff ass spring in it.
Why do you think that I am hiding stuff? I want to drive the car again, hiding shit would only hurt me.

Edit: I just looked at the thread again and see that I didn't mention it here... I must have mentioned the BOV to my brother or a friend and though I said it here. My bad.

Correct. It's a Brazil delivered Eclipse.

I've been pulling my hair out and overlooked the fact that ECU pin 92 has 0ohm to chassis ground.
It seems that there's a short to chassis ground somewhere in the sensor grounds, maybe that is what is causing my problems. I'll inspect further into this now...

Edit2: Not a sensor ground problem. I had partially disconnected the ECU plug, that's why I had a "short" to ground.

I wasn't being accusatory. After 30 plus years of this we've just seen too many descriptions of "its all stock" only to find it isn't.

Sounds like youre getting somewhere on the chassis ground possibility.

Are you in brazil?

phant0mxx · Wednesday at 1:50 PM

pauleyman said:
I wasn't being accusatory. After 30 plus years of this we've just seen too many descriptions of "its all stock" only to find it isn't.

Sounds like youre getting somewhere on the chassis ground possibility.

Are you in brazil?

I understand, sorry about my outburst.

Unfortunately it wasn't the sensor grounds, I had partially disconnected the ECU and it caused my incorrect reading. I checked again with the connector fully off and it checks out fine.

I am living in Argentina now and there seems to be a mix of Brazil, Europe, and US cars here.

I let the car warm up with the US spec ECU and connected an OBD2 scanner.
I can see live data now.

Car is in closed loop once fully warm.
Injector 2.3ms
LTFT 12.5%
STFT 5.5%
AFS 4mg/s
VAF 37hz
Baro 101kpa (does not move, bad MAP or MAF?)
MDP 105.8kpa (does not move, same as above?)
IAC position 0 steps (this could be a bad I chip on my friends ECU, doesn't affect revving)
Timing ~9btdc (when I rev it up it jumps to 30btdc and then when it sputters it retards back to close to 10 btdc

phant0mxx · Wednesday at 2:27 PM

Okay, MDP sensor was bad but installing a new one still did not solve anything.

New MDP sensor reads ~63kpa at idle and moves with throttle input.

pauleyman · Wednesday at 3:13 PM

I cant comment on US vs non US cars but I'm curious if the pin outs are the same. Stands to reason they would be or there would need to be different harnesses. Not sure why there would be obd2 and non obd2 ecus.

flintoid · Wednesday at 3:18 PM

how new are your plugs and wires? how are the plugs gapped? Have you measured them?

High rpm break up sure sounds like it's your plugs / ignition. Just because they were taken from a running car doesn't mean they'll work in yours. You scoping your crank sensors sure makes it seem like it's an ignition fault. And your wire harness around the PTU sure looks wild. I wonder if there's some sort of electrical loss around that.

phant0mxx · Wednesday at 3:34 PM

pauleyman said:
I cant comment on US vs non US cars but I'm curious if the pin outs are the same. Stands to reason they would be or there would need to be different harnesses. Not sure why there would be obd2 and non obd2 ecus.

Afaik, pinouts are identical and the car behaves exactly the same with both ECUs. My best guess is that outside the US OBD2 wasn't mandatory from 1996.

flintoid said:
how new are your plugs and wires? how are the plugs gapped? Have you measured them?

High rpm break up sure sounds like it's your plugs / ignition. Just because they were taken from a running car doesn't mean they'll work in yours. You scoping your crank sensors sure makes it seem like it's an ignition fault. And your wire harness around the PTU sure looks wild. I wonder if there's some sort of electrical loss around that.

Installed brand new BPR6ES plugs and gapped to 0.028".
Wires are also brand new.

The high RPM breakup only happens free revving, when driving, the breakup occurs at 3000rpm.

I know it looks messy because I pulled it all out to inspect and pin test and then didn't route it like it was before. I'll clean it up once the problem is resolved.

All swapped parts from the running car were returned and they are confirmed working.

flintoid · Wednesday at 3:39 PM

How are your coils? have you swapped those out? Are they grounded properly?

I'm hyper focused on ignition because I had this very same issue and it was plugs/wires for me. Free revving it would break up close to redline. Then under load it'd break up at a lower RPM, but sometimes I could "creep it up" by easing into it.

Have you boost leak tested everything? (I know, this is a very basic question, but I swore I had a boost leak until I replaced my plugs and wires and was CONSTANTLY boost leak testing my car hoping I'd find something)

steve · Wednesday at 4:02 PM

phant0mxx said:
IAC position 0 steps (this could be a bad I chip on my friends ECU, doesn't affect revving)

This indicates that the ECU was unable to lower the idle speed to it's programmed setting and has attempted to close the ISC 100%. You mentioned a bad ISC blowing the drivers in the MD305162 ECU. Have you measured the coil resistances of the "replacement" ISC? Made sure you don't have any boost/vacuum leaks that could cause the idle to be high? Adjusted the BISS?

phant0mxx · Wednesday at 4:29 PM

flintoid said:
How are your coils? have you swapped those out? Are they grounded properly?

I'm hyper focused on ignition because I had this very same issue and it was plugs/wires for me. Free revving it would break up close to redline. Then under load it'd break up at a lower RPM, but sometimes I could "creep it up" by easing into it.

Have you boost leak tested everything? (I know, this is a very basic question, but I swore I had a boost leak until I replaced my plugs and wires and was CONSTANTLY boost leak testing my car hoping I'd find something)

I have swapped coils and measured the resistance of primary and secondary coils, all were within spec.
By grounded properly you mean bolted to the intake manifold? If so yes.
To me it sounds more like a "soft" injection cut rather than a "hard" ignition cut, but I could be wrong.
The ignition timing does become retarded when it starts to sputter. It jumps to 30btdc normally and when sputtering starts it drops to 10-15btdc and revs climb down. (all this free revving)

I have boost leak tested everything once more and the TB shaft and brake booster (between the booster and brake master cylinder) have leaks, but I wouldn't attribute this problem to those leaks because at no point in time I am able to make positive pressure from the turbo before the sputtering begins.

I appreciate the comments, keeps me engaged and takes away some of the anger caused by not being able to find the root cause of this problem LOL

steve said:
This indicates that the ECU was unable to lower the idle speed to it's programmed setting and has attempted to close the ISC 100%. You mentioned a bad ISC blowing the drivers in the MD305162 ECU. Have you measured the coil resistances of the "replacement" ISC? Made sure you don't have any boost/vacuum leaks that could cause the idle to be high? Adjusted the BISS?

Battery has been out a couple of times today, so maybe I need to re calibrate the ISC position.
I have measured the resistance of the replacement, its at ~43 ohm on all 4 coils, it's an aftermarket black top ISC.
Boost leak tested once again, described above.
I initially adjusted everything to make it idle smooth, and it worked fine. All out of whack now it seems.
I did use the Brazil ECU and not the US ECU to adjust.
Anyways, I can redo all that once the revving problem is finally resolved, idling high (although it is idling at 850 rpm) is not the cause of being unable to reach redline and causing ignition timing to retard and make the engine sputter.

An additional comment on the problem, I noticed that sometimes when I rev it up and hit the threshold where it sputters, sometimes the throttle becomes unresponsive, no matter how much throttle I give it it will not react, and sometimes it will just die coming down from the sputtering.

flintoid · Wednesday at 4:55 PM

have you tried swapping the MAF? The throttle being unresponsive and sometimes killing it seems like it could be a bad MAF signal. Along those lines, have you checked the signal from the TPS? Is the voltage a linear increase or does it have dead zones?

pauleyman · Wednesday at 5:43 PM

Hiw about this. Get the car somewhere dark and watch for acing issues. Yiu could have a crack on a coil tower and didnt spot it. Wires are almost certainly ruled out. I had a similar problem once and it was arcing at the spark plug boots but anywhere it can jump you are losing ignition in a cylinder

phant0mxx · Wednesday at 6:13 PM

flintoid said:
have you tried swapping the MAF? The throttle being unresponsive and sometimes killing it seems like it could be a bad MAF signal. Along those lines, have you checked the signal from the TPS? Is the voltage a linear increase or does it have dead zones?

I've tried 2 other MAFs, one spare one I had and one from a good known working car.
Using the US spec ECU and watching the live data I don't see any erratic MAF related signals.
TPS also outputs linear voltage.

If I hold the pedal until it sputters and continue to hold it, I can trigger a code 31(knock sensor) on the Brazil ECU and I can read the flashes on the CEL by jumping Pin 1 of the OBD2 to GND.
I tried doing it again on the US ECU and it didn't trigger anything. Strange... Do I have 2 bad ECUs? The donor of the US spec ECU was parted out and I can't reinstall it to check and I don't know anyone else with an Eclipse that will let me take parts off to test.

pauleyman said:
Hiw about this. Get the car somewhere dark and watch for acing issues. Yiu could have a crack on a coil tower and didnt spot it. Wires are almost certainly ruled out. I had a similar problem once and it was racing at the spark plug boots but anywhere it can jump you are losing ignition in a cylinder

I just closed the garage and turned all lights off. No sparking unfortunately

The Central Hub for DSM Community and Information

2G 1995 Eclipse GST engine sputters when free revving

phant0mxx

Attachments

dwb

chrisman287

phant0mxx

phant0mxx

Attachments

phant0mxx

Attachments

phant0mxx

phant0mxx

dwb

phant0mxx

Attachments

pauleyman

steve

phant0mxx

pauleyman

phant0mxx

phant0mxx

pauleyman

flintoid

phant0mxx

flintoid

steve

phant0mxx

flintoid

pauleyman

phant0mxx

Similar threads

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds