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 ExtremePSI
Please Support Fuel Injector Clinic

2G Misfire on #2 - P0302, Nothing left to change! Help a 1996 Eclipse GSX

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.

Scottinalaska

Probationary Member
29
2
Jun 28, 2017
Anchorage, Alaska
1996 Eclipse, automatic, completely stock
Code: P0302 - Miss on #2 cylinder.

All four NGK plugs and wires are new.
Fuel pressure: in spec at 45psi at all RPMs
#2 Spark plug is black, appearing rich condition while other plugs are more white in color showing good burn.

Replaced Coil that operates #2 and #3
Cleared codes and drove. Miss on #2 remained.

Swapped fuel injectors #1 and #2.
Cleared codes and drove. Miss on #2 remained.

Swapped ALL new plugs putting #2 plug into #4.
Cleared codes and drove. Drove fantastic for 2 miles, then miss started again. Miss on #2 remained.

Swapped plug wires (new) between #2 and #3.
Cleared codes and drove. Miss on #2 remained

Compression check: #1: 180, #2: 160, #3: 190, #4: 145

90 octane gas. Fresh oil change.
Is there something I have missed? Compression isn't even. Is oil causing a misfire? Ideas? Thanks!
 
Based on your compression numbers, you need to do something. Could be a headgasket, could need a rebuild. But clearly, you need to spend some money and time unfortunately.
 
Whenever you get the cylinder misfire do you get any smoke coming out of your exhaust? Also drive the car and when you get the cylinder misfire stop the car and pull out the spark plug and check for anything in the cylinder like oil. Have you noticed your coolant getting lower? that could also be a sign.
 
Do a PROPER compression test, throttle wide open and crank the engine until the compression gauge stops moving in each cylinder, all spark plugs need to be out also not just the one your checking, your compression is not likely causing your misfire in this case.
 
Part you still need to check is the injector resistor.

If there is a bad resistor it can miss some times In one line (which in this case is #2) and cause the miss fire

It will be hard to test it which it miss sometimes and not all the time, so change the resistor will be the way to go.

If the harness has a lot of leverage swap injector connector from #2 to #3 and see what it does.

If the resistor is your problem then you will have a miss fire on #3.

Do that and report back.
 
I dont mean to hijack this thread or anything. I had a question of my own about checking that resistor pack. I was throwing an injector code a while back and my first thought was to check that. The code seems to have cleared itself up amidst all of my poking and prodding and I never got around to checking the resistor pack yet in still curious. It's a pain to get to the resistor pack just for probing a couple terminals and the harness is too short. Since the injectors are low side switched at the ecu if I recall correctly and I believe power is fed to the resistor from the battery through a relay. Is it possible to probe the positive side of the injector and the negative side of the relay to test the resistor for that corresponding injector or is that not how they are wired?
 
The injector resistor is easy enough to unbolt and bring up top to check, a misfire like this could be a failing ecu though.
 
Do a PROPER compression test, throttle wide open and crank the engine until the compression gauge stops moving in each cylinder, all spark plugs need to be out also not just the one your checking, your compression is not likely causing your misfire in this case.

Thanks, Motomattx. I redid the compression check. The number 4 is now at 155 (10psi higher). The rest remained the same - #1: 180, #2: 160, #3: 190 #4: 155
Also swapped #2 and #3 fuel injector harnesses as MiguelMCV suggested. Now I have a P0300 random misfire instead of just misfire on 2. Running pretty rough! Here are the plugs though. Where # 2 usually looked dark, now #3 looks dark after 20 miles of driving/missing!
You must be logged in to view this image or video.
 
Part you still need to check is the injector resistor.

If there is a bad resistor it can miss some times In one line (which in this case is #2) and cause the miss fire

It will be hard to test it which it miss sometimes and not all the time, so change the resistor will be the way to go.

If the harness has a lot of leverage swap injector connector from #2 to #3 and see what it does.

If the resistor is your problem then you will have a miss fire on #3.

Do that and report back.
Just posted a result above. Thanks for the advice. I was able to swap the injector wires. So, the injector pulse isn't timed like a spark so it doesn't matter which harness is on which injector?
Scott
 
The injectors are batch fired twice per 4 stroke cycle iirc, even though they were claimed to be sequential in advertising. Pulsewidth changes the same for all 4 based on how much oxygen hits the O2 sensor.

The injector harness swap proves to you that the problem is either in voltage or resistance to that injector. Check the resistor yet?

No, Delta. I haven't checked the resistor(s) yet. That part is new territory for me. Need a little guidance. thanks.
 
No, Delta. I haven't checked the resistor(s) yet. That part is new territory for me. Need a little guidance. thanks.
Can it be as simple as installing another resistor pack at this point? Or am I possibly running into an ECM that has a driver melted inside it?
 
So, I located a known good Resistor Pack, bought it and installed it. It ran fantastic for the first couple miles. The car had sat for a week. Then the missing began again, check engine light came on again. I am thinking that maybe it could be the ECU. But I am also thinking perhaps since it seems to be heat related, I should do another compression check when it's hot. Maybe the numbers will go whacko. No smoke though.
 
Did a compression check when it was hot and missing. Readings were all about the same as before. All held compression for the 15 seconds I watched before I bled it down and went to the next hole. I decided to just throw in some fresh plugs - NGK BPR6es, the usual. .28 gap, etc.
It ran Fantastic for a couple miles. Flawless, actually. And then, the miss returned slowly getting worse for the next mile. Code P0300 - random misfires. I am afraid I am going the ECU route now. I am out of options.
 
You didn’t swap to a 1g CAS did you?

If you did then that’s your problem. If you didn’t then do one last try.
At the back of the coils swap the wires ( blue/black stripe and blue), once you swap the wires then swap the firing order at the spark plugs,
Swap wire 2 to wire 1 and swap number 3 with number 4 and see what it does.
If you get a mis fire on number 1 then your coils are bad.

If you do have a 1g cas then your only option is to go back to a 2g sensors or use a potentiometer to fix it, otherwise you will need ecmlink to cure your problem.
 
Miguel, Thanks for your detailed reply. If I went to Oreilly's and got a 1996 CAS, I assume it was a 2G sensor.
Periodically, especially when it is dead cold, it runs fantastic for a couple miles! Does this alone rule out an incorrect CAS?
Now, for the next question if we have the proper 2G CAS,
Recommendations for a new ECU? or repair? Or upgrade to Link? It's just a factory automatic GSX for the customer. thanks.
 
Post a picture of the engine bay. The 95-96 CAS is located behind the intake cam sprocket, 1G and 97+ CAS are on the other side of the head.

If it only happens when the engine is warm, it's probably because it's going into closed loop and the fuel trims are affecting fueling. Your misfires will read as a lean burn by an O2 sensor and dump extra fuel as soon as you hit temp for closed loop. Since this is for a customer, should we assume you are a shop with a scantool that can view live data via the obd2 port?

Also, I am going to assume that you meant 0.028" on the plug gap. That missing zero can make a really big difference.
 
Last edited:
Haha! Yes, I missed a zero. That would be SOME gap on the plugs, Delta!
I just verified a couple of things. First, it is a 1997, not a 1996. Since it has been over a month since I installed the CAS, I no longer have a good record of what the part number was. Interestingly enough, when I called Orielley's to check which part number they sold me (Each year is different!), the don't show either sensor being sold in the last year!
Thus, I am left to conclude either I was sold a completely different part number OR more likely I think, I bought a 1996 model one since the price listed is what I recall paying and their records are wrong (perfect fit on the right side of the engine though, like a 97 should be.. .
I gave the car back to the customer to drive for the week so I could have a break!
Here are a couple photos. I am willing to try a different sensor - recall the original reason I replaced it was because it was a no-start condition since it no longer triggered the fuel pump to keep pumping after crank mode.
yes, I can check fuel trims, etc on my scantool. I just didn't.
scott
 

Attachments

  • Eclipse timing belt side.JPG
    Eclipse timing belt side.JPG
    268.9 KB · Views: 161
  • Eclipse CAS.JPG
    Eclipse CAS.JPG
    356.7 KB · Views: 187
That is indeed a 97 CAS in your picture. The stock pos plastic bpv/bov still being present is also a pretty good indicator that the car is otherwise rather unmolested.

Is P0300 the only DTC? When you replaced the CAS, was there code P0340 or was it P0335?

I wonder if you reset it now, whether you would still get the random misfire or a cylinder specific code.

The next thing I would do at this point before resetting is check fuel trims after the customer has driven it this week. Of particular interest are LTFTs. There are 2 Long Term Fuel Trims active on a 2G factory ECU, low airflow and mid airflow. These are the STFT values that were consistently seen long enough for the ECU to move them into a more permanent correction. So STFT might be sitting at -5%, but LTFT might be at +12%, meaning that the ECU is actually adding 7% more fuel than the fuel map in memory calls for.

After noting LTFTs, I would scroll through the scantool to the front O2 sensor and make sure it's still switching. If it's squatting, straining and dropping a constant 0.01v full lean and the LTFTs are both at +25%, you're either actually getting a lot less fuel than you should or you simply have a bad O2 sensor.

I would also verify the CTS is ballpark accurate. If it's been idling for 10 minutes and still says the temp is -69F, ...well there you go.
 
Last edited:
Post a picture of the engine bay. The 95-96 CAS is located behind the intake cam sprocket, 1G and 97+ CAS are on the other side of the head.

If it only happens when the engine is warm, it's probably because it's going into closed loop and the fuel trims are affecting fueling. Your misfires will read as a lean burn by an O2 sensor and dump extra fuel as soon as you hit temp for closed loop. Since this is for a customer, should we assume you are a shop with a scantool that can view live data via the obd2 port?

Also, I am going to assume that you meant 0.028" on the plug gap. That missing zero can make a really big difference.
I just want to tell you I admire your responses and recommendations. I was planning to add some input here but you have hit anything I could mention.
 
Great news! Car returned due to a major leaking crank seal. Timing belt is now due at 120K anyway so both belts and all related components will be replaced this week. This gives me a bit more time for diagnosis too.

O2 sensors are active and working fantastic in tandem.

Coolant Temp Sensor (CTS) starts at 7 degrees Celsius and rises to 101 degrees C. Good.

LTFT – 1.6% - 3.9%

STFT – 0 – 1.6%

Ignition timing advance for Cylinder #1 actually displays anything from 7.5 to 17 degrees! Maybe this is normal. I was just never looking at that before.

This is all at idle.

When I started it up cold, it promptly evidences a miss just by listening. Code P0300 (random misfire) is present. But no other codes. Originally, I got a cam code (P0340) when it wouldn’t start and got towed in.. Replacing the cam sensor fixed the no start condition but it was missing within a couple of miles. Now it misses right away.

Condemn the ECU yet?
 
While I am changing the timing belt and all components, maybe I should toss in a new crank sensor. It's the one electronic component, aside from the ECM, that I haven't changed. Good idea?
 
Yes, replace it if you will already be in there. I wouldn't be surprised if you find the mechanical timing to be off too.
The timing belt was due. The crank seemed almost a whole tooth forward due to belt stretch. The only places I can see oil possibly leaking is the crank and the camshafts. Below the cams it is dry as a bone. The crank shows a bit of seepage. But just ABOVE the crank a drip is seen here and there. And when running it dripped from under the lower cover a drip every 5 seconds! I'd hate to miss a mystery seal thinking just those three will fix it.
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.
 
So after a successful timing belt install, perfect tension, new crank sensor (only one) etc., the car started up and ran flawlessly for about 8 minutes. Smooth idle, settled down after a minute. But then the miss showed up again.
I feel I am left with getting the ECM rebuilt or another one. It is a NONeprom one. It looks fine inside but as I understand, they seem to have only so much life. Any recommendations? Or just spend 150 bucks and have a rebuilt one programmed and sent to me? like
https://www.fs1inc.com/1997-mitsubishi-eclipse-2-0l-engine-computer-pcm-ecm-ecu-programmed-plug-play.html
or off to ECMtuning?
thanks
You must be logged in to view this image or video.
You must be logged in to view this image or video.
 
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