Injectors being emo (maybe?)

When I start my car cold, it's fine. However, if I turn it off and it is off for more than about 60 seconds, I get check engine and wicked shakes (which combined with the feeble acceleration and reek of gasoline tells me I have at least one cylinder dead). If the car remains off for several hours (at least 5), it gets cold enough that it's fine again. It doesn't seem to matter if I drive 5 miles to work or 50 miles to Denver, it's fine until I turn it off and start it warm/hot.

The problem seemed to start after I washed the engine bay prior to the recent Colorado meet. It may just be coincidence. My initial thought was that some water had gotten somewhere it oughtn't have, so I pulled the spark plug wires and blew that area out with an air compressor. Didn't fix it.

I checked the CEL code and it says injector (code 41). I disconnected the control/sensor wires (which is a PITA, btw) and blew that out with air and then applied some dielectric grease to the contacts. Didn't fix it.

I put about 3/4 can of Seafoam into the gas tank with some fresh gas to clean the injectors (or dehydrate, if that's the problem), and the rest into the crank case. Didn't fix it.

I'm gonna clean the MAF sensor soon, but other than that, I don't know what else to do besides replace injectors, and I'd rather not drop money on stock ones when I'm gonna be putting in bigger ones in the foreseeable future after I get all the upgrade stuff together. Unless the ECU is lying to me and the problem is not injectors. My roommate suggested it might also be spark plug wires...the plugs themselves are only about a year old; NGK as everyone seems to recommend for our cars.

If I do need new injectors, I might as well tough it out until I can get the big FMIC and air intake installed (currently sitting in the garage), move battery (which needs replacement anyway) and install bigger injectors. I am also supposed to be able to grab a Walbro 255 fuel pump from a dead Eclipse if/when I need it.

So ultimately, my question is twofold:
1. Any other ideas for diagnosing/fixing this problem? Anyone north of Denver got a datalogger I could use?
2. What upgrades do I need before installing bigger injectors (if I need to replace them)?

PS - Car is completely stock, has the usual 1g exhaust mani leak, otherwise in great condition.

 

[toybreaker]

Could be water got in the cas, and corrosion has set in.

Also, did you blow out the transistor connector?

I would also pull the plate it bolts to, remove the transistor, and make sure water didn't enter the unit thru the little hole on the back of the transistor body.

Make sure the plug wire terminals in the coils are clean and dry.

Make sure to blow out the o2 connector as well.

So, yea, basically disconnect every single item, dry them, inspect, clean as required.

Logger is avaliable, but you'll have to come down to town.

 

[rgeier11]

My guess would be either bad injector or bad ECU.

 

[atc250r]

I had a similar problem and it ended up being the CAS plug. I removed the injector plugs and soldered on new ones but it didn't help. I carefully took the CAS plug apart, disassembled all the individual connectors, and soldered new wires onto them. That may not be your problem but it sounds like it is CAS related so I'd try swapping another in. When I swapped mine with a known good one it still did it so I ended up doing the connector.

John

 

[atc250r]

Oh, and I'm confused by the "Injectors Being Emo" title. Did they start to cry alot, wear dark clothes, makeup, and get a funny hair cut like this dork?

/ubbthreads/images/graemlins/grin.gif

John

 

[Bimmubishi]

Like this

click

 

Does the CEL code come back if you reset the ECU? You will only get that code if there is an actual electrical path issue with the injector or wiring, or damage to the part of the ECU that checks for continuity of the injector circuit. The ECU checks to make sure that the 12 volts that feeds the injector is present on each injector wire between firings. If it doesn't see one it flags a Code 41. If it happened just after you washed the car, it could simply be oxidation on one of the injector plugs. Take them off, look for oxidation, clean them, and put them back on.
I've also seen this be caused by a bad injector wire between the injector and the ECU, but that was only after I had swapped out a couple hundred ECU's for testing, which broke the wire inside the insulation. /ubbthreads/images/graemlins/frown.gif

 

Resetting ECU entails simply disconnecting the negative terminal on the battery, right?

What is CAS? And which transistor?

Injector plugs appeared to be shiny and clean...didn't see any corrosion or oxidation anywhere, and put the dielectric grease on to ensure that that wouldn't happen later.

I will try resetting ECU when I get home from work, then if it persists, check CAS and the rest. I'm mostly baffled because it's fine when it's started cold, and generally runs fine until it is turned off (it once went back to normal while acting up, and once or twice crapped out after starting and running ok, mid-drive).

PS - that pic looks more like a fat goth kid than an emo kid. Never trust a fat goth...saw that on a T-shirt once, hehe.

 

[Andy_S]

CAS = Cam Angle Sensor

 

[atc250r]

You've missed the point. How the heck are your injectors being "emo"? /ubbthreads/images/graemlins/confused.gif

 

They're being emo because sometimes they work fine and sometimes they sulk and pout. I guess bipolar might be more accurate, since emo kids are sad and pouty all the time.

Resetting ECU did not fix it.

 

What I would do next is disable one injector at a time with the logger, if you have one with that feature, to see which injector it is. Actually, now that I think about it, the way I found my bad injector wire was to unplug the connector from the ECU and ohm out the 4 injector wires to GROUND with the car turned OFF. You should get similar readings on all 4 of them, about 13 ohms if I recall.
1G ECU harness
Of course, this will only work when you are actually having the problem, so you will need to get the car hot first. Don't forget that it could be a problem with the injector resistor pack. Perhaps you fractured one of them when you shot water on it? /ubbthreads/images/graemlins/dunno.gif

 

Hey, I am not sure if you are still having this problem but if you are able to drive up to at least Thornton I could help ya with my logger. I am sure GMP would let us use his garage if we needed to lol!

 

Thanks for the offer, Cody. PM me your phone number and beer preference; I can probably come down on Sunday if that works for you (and Gary if we'll be using his garage...shouldn't be TOO involved as long as we have a decent set of tools handy). If I don't figure it out before then, that should definitely help track down the source of the problem.

Also, I know where the coil is, but where are the CAS and the transistor in question located? I guess I should just unplug other stuff attached to the ignition and fuel systems and check it like I did the injector plugs...bound to come across it that way! I'll apply dielectric grease to these as well, especially if it looks sketchy.

I'm still baffled by the problem, in that it doesn't matter how long it runs: it's only bad on a hot start, but not from getting heated to operating temp after a cold start. You'd think it'd crap out once it got hot.

 

Okay, so I went down to Cody's place and we hooked up the data logger and could not re-create the problem...and it hasn't done it since. WTF? Maybe it worked itself out?

I'm still going to inspect a couple more connectors to make sure there is no corrosion. I also wonder if it's more loose wiring (like the kickpanel fusebox) that's been jarred out of its heat sensitivity.

Thanks for aid in this diagnosis...I wish I could say for sure what caused it, although now that I've posted, it'll probably start doing it again.

 

Did you check the connector to the resistor pack on the firewall? It had to be either there or at one of the injectors, IMHO.

 

So it's started up again, after a few months of no problems...definitely not water-related, and I'm venturing not even necessarily anything in front of the firewall.

Here's my hypothesis: My kickpanel fusebox is notoriously loose (IE, my power locks, wipers, radio, seatbelts, and clock all sh*t out on me because the fuse clasps are loose...have to wiggle the "Room" fuse to get it all back on). If the ECU runs anything through this box, it may be disrupting the signal from time to time, such as temp or knock or something, which would then trick the ECU into thinking the injectors were acting up and thusly throw CEL and put fuel mix out of whack. Or maybe that looseness would be losing power to one or more injectors, leaving them shut (and legitimately throwing a warning from ECU). Engine temp still seems to be the operative factor here...and it has, on occasion, kicked out of being screwed up and back to full power.

Fusebox is the likely culprit, and needs to be replaced anyway, so I will start there. I've not reviewed my CD o' Galant diagrams about what goes thru that box, but I would venture something related is in there somewhere (not on the easily reached fusible links on the bottom, but perhaps in the upper, recessed area).

Input appreciated.

 

[toybreaker]

The problem that you're seeing with your fuse box has is that the little buss bars inside have been subjected to an overcurrent situation severe enough to anneal/soften them to the point that they have lost thier gription. (heh...love dat woid)

moral to the story don't substitute fuse ratings

Someone in years past put a few fuses of the wrong rating in when they were having some issues, and well, you see the result.

I thought you'd be having issues with it further down the road when I noticed it last summer. The volkswagons I grew up on had these kinds of problems a lot, and I learned to be scared of tampered/improper fuses, as the prince of darkeness was always waiting to catch me with a date in the rain to smote me with electrical inactivity. /ubbthreads/images/graemlins/banghead.gif

If I remember correctly, it was on the lighting side of life that there was some 30's in there instead of the proper rating, but it's been awhile since I looked at it.

I found a servicable fuse box in the parts pile, but I think it would be wise to carefully inspect all the plugs and harness pins before just blindly installing it and trusting everything to be golden.

I should be back in town in a day or two, and I'll give you a call then.

 

My roommate's dad (an electrical engineer) suggested something similar when we first pulled out the fusebox and discovered the problem...excess current weakening the metal...heh, yeah, gription...nice. In any event, I'm not sure why the previous owner would have done that, unless he was not the original owner (because I know for a fact that he knows better than to do that). I didn't know that current (other than the effect of heat) could anneal metal, but I figured it could behave sort of like an electromagnet and the flexion would work-harden the contacts...but that'd mean broke fuse grippers, not excessively malleable ones.

Sounds good...I will be sure to verify proper amperage for all fuses upon installing a replacement box, and examine the harnesses. Thanks!

Jeff: do you think if the sensors aren't making it to the ECU that it could be sending false signals to injectors and/or that bad fusebox could also prevent ECU from sending signal at all to one or multiple injectors? Given the trouble with the interior fusebox, and the possibility of ECU going thru it (still haven't looked, but I'm guessing it does), is that a more likely culprit than the resistor pack? I'll try to look for it if I have time this weekend...I put it off because the problem stopped for a spell.

 

Well, I finally got my own data logging gear, and either I'm doin it wrong, or the ECU is going bonkers when this problem occurs, as it's throwing all kinds and counts of error codes at me in Tunerstein. I'ma set up MMcD to run on a Palm emulator and try that method when I get a chance.

Anyway, at this point, I'm suspecting my relay box as being more the culprit of this than the injectors themselves, but we'll find out when it's time to tear out the box and inspect the wiring harnesses. And the ECU /ubbthreads/images/graemlins/frown.gif