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 STM Tuned
Please Support ExtremePSI

2G No start 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.

Jon Lane

15+ Year Contributor
405
31
May 1, 2004
SW, Florida
'98 2G Tsi. New engine on original everything - wiring, fuel, coils, etc. Coils wired correctly. Car cranks and gives a mild misfire.

Crank sensor installed carefully so I assume it's right. Need to re-check cam sensor.

Before I tear into this much more what would a mild misfire about once a rotation indicate ? If either sensor were off, would engine get spark?
 
There's no indication either spark or fuel are a problem - I smell fuel and purged the system prior.

The question is if a mistimed sensor will still fire. Will the computer signal the igniter if one sensor is off?
 
Try swapping plug wires 1&4 for 2&3, swap them at the coils, maybe you have them 180* out for some reason.

Right you are. Last night I realized the coil pack had been inverted for the Evo II manifold. Rerouted the plug wires and am much closer to a fire. Also verified injectors are in order.

Now I have a stumbling start, as if two cylinders are firing and the other two are not. Got one chuff back out the intake. Still searching for answers. I have a second igniter I could try too.
 
Ignition is triggered by the crank sensor only, if you have spark at all then the crank sensor is good. The fuel injectors are triggered by short/long pattern of the cam sensor and by whether there's any misalignment with the crank sensor. In the absence of a CAS signal, the ECU will attempt to guess the correct alignment one time per ignition switch power on. If the cam sensor signal is intermittent there will be an all or nothing cranking each time the ignition key switch is twisted, which is a tell-tale sign of a bad 2Gb CAS.

Your checklist should be:
Verify cam timing
Verify injector resistor pack and injector leads
Verify CAS signal
Verify coils
Verify power transistor
 
Good advice, all. Thank you.

Having corrected the plug order I replaced the igniter. After a short crank I got it to fire and run, however it seemed like a third cylinder had to come online after the first two. It needed a little throttle. The fourth cylinder is still offline. I let it idle 5-10 secs to verify good oil pressure and switched off.

It feels to me like the ECU has to recognize everything and is slowly getting each cylinder up to snuff. Given that the car responded to the plug wire problem and possibly the second igniter, I'm thinking of letting it idle and see if that fourth cylinder activates.

This seems to verify everything that's on a shared system - sensors, fuel, timing, ECU. The unknown would be coils and igniter. Since both igniters and the coil set were taken from the car when running last, I feel they should be fine too. Maybe the last injector isn't primed yet. All four exhaust runners felt warm but obviously that's not a valid check if one cylinder is offline.

Think I'll check head bolt torque and give it a coolant fill and try again.

Ignition is triggered by the crank sensor only, if you have spark at all then the crank sensor is good. The fuel injectors are triggered by short/long pattern of the cam sensor and by whether there's any misalignment with the crank sensor. In the absence of a CAS signal, the ECU will attempt to guess the correct alignment one time per ignition switch power on.

Pertinent info.

If the cam sensor signal is intermittent there will be an all or nothing cranking each time the ignition key switch is twisted, which is a tell-tale sign of a bad 2Gb CAS.

I had a bad cam sensor once. Car would start, run three seconds, die. An aftermarket sensor failed almost right away. A second OEM sensor has been running ever since.

Do any of these symptoms cause you to doubt the cam sensor?
 
Last edited by a moderator:
Pull the plug out of the cylinder that is still off line and see if it is dry or fuel soaked. I bet it is dry and the injector isn't getting it fuel, but we will see.
 
Right again: Cyl #1 is dry. What are options; pull that injector and soak it in spirits overnight?
 
Do you have access to a noid light?

Yes, clean it. You can also unplug the injector and with the engine off, use a little 9V battery and a pair of alligator clips to put power to the pins on the injector to check if it's stuck or not. You'll hear it click if it's opening. Just don't leave the alligator clips on it for more than a couple minutes or you could damage the injector.
 
I'd pull the injector, and do as Zack mentioned with the test. I also do that test and clamp a 6" pc of hose to the top of the injector, fill the hose with carb cleaner spray and then hose clamp it to my air hose nozzle. When I touch the battery to the injector pins with a couple pc's of wire and clips, I put compressed air to the hose full of cleaner and make the injector SPRAY. If it doesn't spray or looks very odd in pattern (I blow it against a pc of cardboard to look at the pattern) then it needs cleaned several times or sent in for service. :thumb:
That is also how I "pickle" injectors that I am taking out of service, only I use WD40 in the hose and make it spray.
 
Injector is out. Electrical test clicks the valve cleanly. I have the spray end soaking in spirits overnight.

If it clicks should it spray or are the two separate things?
 
It should spray. It is either extremely clogged or dead. You are verifying the electrical part but the mechanical part isn't cooperating right now. Put some of your cleaner into the top part of the injector also, the clog could be before the solenoid (the electrical part that clicks) although I doubt it.
 
No, you'd have to have a noid light to verify signal at the injector. The pulse widths are way too short for most VOMs, even top of the line Fluke meters. That's why I asked if you had access to a noid light in post 14. You can rent them from some parts stores for free with a deposit/refund system.
 
...you'd have to have a noid light to verify signal at the injector.

I see; didn't know that. Amazon has single lights for ten dollars - do you know which type is compatible? For example, there's a Bosch2 type with what looks like the same flat connectors.
 
Noid light says this injector circuit is live.

Soaked injector a while and checked it by drawing a vacuum through it while giving it 12v. By that test it's fine - no flow without voltage. With voltage it clicks and goes open. Cyl #1 still off.

No idea why this cylinder looks dry.

I'll try another igniter. After that maybe a coil pack?

UPDATE: Second (original) power transistor igniter is shot. Seems to try running 2 cylinders. That's why it's not in there.

Plug wire is good - measures 178 ohms.

Part number MD189747 looks like the 2G transistor, or J722T, which also applies to the 1G? Do these things ever drop just one cylinder or is it always a pair at a time?
 
Last edited:
Drop a long dowel (screwdriver, socket extension, etc) into the plug hole for #1 and slowly rotate the crank with a breaker bar until you have true top dead center on #1. Then verify the cam sprocket timing marks are lined up.

Timing verified.
 
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