Engine got wet, now won't start.

[Michelle]

I recently removed all of the front end parts from my second-generation Legacy to repair damage done during a minor collision. There was a lull between periods that I was working on it, and in that time, it rained on the exposed engine. The battery was not in the car when it happened, but upon re-installation, the car will not start. It's cranking over with full strength, and I know I'm getting fuel. The issue seems to be something ignition-related. The service manual doesn't seem to help much with diagnosis for ignition-related troubles. I am poring over it today in an attempt to resolve this, as I'd like to be able to drive the car to the body shop soon. I just thought that I'd post this here to see if any of you might know where to begin looking. I don't mind general answers, but know that I have checked fuses and have checked anything seemingly obvious already.

Thanks a lot for the help!

[Hinyo]

Start by checking your fuses and make sure it is all well dried out. I had a ECU fuse I swapped with a bad fuse on accident one time that lead to infinite crank. Than check fuel by removing the main line and putting it into a container to see if it pumps out. check spark by pulling a spark plug out and finding ground to see if it arcs. can anyone else chime in?

[meepers]

I would start at fuses, unplug the coil connector, spray it with some contact cleaner and let it dry completely before reinstalling it. I would recommend pulling the cam sensor and cleaning it and the connector as well. No cam/crank signal will make for no start. Water really shouldnt cause any issue unless you have corrosion or it killed the coil or ignitor. I pressure wash my engine regularly, but I almost always do it with the car running and blow everything off with compressed air while it runs to get all the nooks nice and dry and not get water spots.

[Michelle]

It's sat in the Texas summer heat for a couple of weeks since it was rained on, so I feel that anything needing to be dried out would have by now. I'll look into both of those, though, as the coil connector is bone dry. An excuse to by the MSD coil wouldn't hurt if it did turn out to be bad, I presume.

Do any of you have any experience getting code readers to work on these cars if I can only get it into "accessories" mode and not actually start it? I know my boyfriend had an older scanner at one point and it would behave differently for different types of cars. We can't seem to find it, so I thought I'd ask before he goes and buys a new one.

[meepers]

Moisture at this point wouldnt be your problem, but you could very well have corrosion/oxidation on those connectors if water got in and was trapped until it dried.

It all depends on the scanner... My old Mac scanner reads my 98 with no problem. I am unable to get the new fangled Snapon Solis scanner that uses resistor keys to read it.

[Michelle]

In the meantime, I got a new Bluetooth scanner to see if I could read the issue in accessories or when cranking. No dice. I also had Jarrett remove and clean both the crankshaft and camshaft position sensors. The connectors on them were perfect, at the metallic surfaces were somewhat grimy. Both were cleaned well and reinstalled. Still same thing. The engine is cranking, we're just not getting spark. This has been verified with a brand new spark plug plugged into the wire held outside of the engine while cranking. Also, all of the fuses seem to check out okay. Any that look questionable are pulled out to verify. So far, all have been intact.

Any additional thoughts?

[Michelle]

Here's a video we made this afternoon when trying to diagnose stuff. Jarrett noticed a noise coming from the IACV on the throttle body. We looked around and can't seem to find many that have this same type of IACV, so that may be one of those weird phase II oddities. Can anybody tell me whether or not the IACV should make this noise in a no-start condition?

[nepbug]

I've never heard that sound before. Maybe clean the IACV. If that doesn't work I've got a known good one in my garage.

Have you tried a checked the coil yet? The FSM tells you how to test it.

Sent from my Nexus 7 using Tapatalk

[Michelle]

This is more of a "thinking out loud" post because I've been reading up on lots of the IACV-related Subaru posts here and elsewhere on the web. While it seems pretty common for the air control valve to cause no-start conditions, I can't see any reason that they would cause the car to get no spark. I plan to remove and clean the IACV when I get home today, but just thought I'd bounce that thought off of someone if there were others trying to avoid doing work at their job today, too.

Didn't see that part in the FSM. I'll go back and look through it.

[1692]

I asked a mechanic friend of mine that worked on my old 96 Legacy wagon, and he had suggested all the things that have been suggested so far in this thread. I think nepbug has a good suggestion with the coil though. I know I have never heard that sound in the video before, and honestly have no idea what it is.

If you haven't done so, I'd call around to some mechanics that specialize in older Subaru's, and see if one can give you some insight that none of us have thought of. With everything you have already checked... I'm stumped!

[nepbug]

Here you go, this is for my 02 Impreza, but should be the same for you.



Edited to add:
Well actually I'm not sure what car you have, you might have the three wire version. In which case, look in your factory service manual under the engine, then under ignition to find the info. If you're three wire and find you have a bad coil, you can do the MSD Neon Coil upgrade

[Michelle]

Thanks for that! I rechecked my FSM (a digital copy I bought on eBay) and found a procedure in the earlier year model folders, but nothing in the 1999 when I checked initially. The 1999 section I have must be supplemental instead of standalone.

Anyway, it had a 3-pin coil instead of the 4-pin coil you've attached. Helpful seeing as my car has a 4-pin coil as well.

I didn't get to work on things tonight as I was called in to work and ended up playing in a couple of hockey games while I was there. To any of you suggesting things, devoting your patience or even the silent readers, thanks for the help. I'll get on it again tomorrow. I'll go pick up a cheap multimeter if I can, too.

[Wiscon_Mark]

If you need a known good spare coil, I have one for sale, here: https://sl-i.net/FORUM/showthread.ph...-added-6-2-15)

[Michelle]

I just bought a new multimeter on Amazon and it should be here Thursday. I'll let you know if I need a coil by then. I'm just hoping it's not an ECU that I need.

[Michelle]

Okay, I've tested for resistance across all of the pins and the spark plug terminals of the coil. The only place I'm getting any reading whatsoever is between 3 & 4 on the coil primary connector. Nothing at all at the specified pins, and nothing across the spark plug wire terminals. Is this an indication that I'm an idiot with a multimeter, or that the coil is for sure bad?

[nepbug]

What multimeter do you have? Can you take a picture of how you have it configured when testing the coil?

Sent from my VS985 4G using Tapatalk

[Michelle]

I was mostly kidding about the ability to use a multimeter. At least I hope so anyway. That would be embarrassing.

[Michelle]

To make this even more confusing, I'm now getting resistance across the coil between 3-4 and 1-2 once I've started using alligator clips to help me out. I'm not, however, getting any resistance between the pins on the primary connector. Even the ones that I was getting previously are no longer giving a reading. It must have been the electrodes touching in the close confines.

Also, I bought a new coil. O'Reilly Auto Parts had it in stock so I figured I'd try it. $113.00 later and it still doesn't start. I thought I was stumped before. I'm beside myself now.

Does anyone know where the ECU fuse is located? I can't seem to find any evidence of it in the 1999 Legacy's FSM.

[r3v_v3ng3]

Any luck?
Yeah the obd2 scanner won't be able to help you out since the battery was out and the computer had no power to store it.
You can buy a cheap spark tester at Harbor Freight for like $5 to see if your getting spark from each cylinder.
Check and make sure your timing belt didn't snap.
Compression test
Other than that it would be the sensors (cam/crank trigger) to check

Also clean that ground on top of the intake manifold (left side by the coil) it might of corroded.
and if your getting fuel, your plugs is probably fouled and cylinder flooded since there is nothing burning the fuel being injected. So check the plugs too.

page 18-20 on FSM 99 RHD on the wiring diagram (fuse location)

[Michelle]

The car is running! Not well, but enough that I have been able to receive codes.

Monday we got the engine to start and run. Don't ask me how. It just did. The idle surged for a bit and eventually leveled off to just run roughly.





I shut the car off and went to have lunch. I came back about 2 hours later and it won't start again.

On Monday night we get it to start and run again, but still not sure how. It idles about the same as it did in the first video I included and is VERY sensitive to me pressing the gas. A tap of the pedal causes the RPMs to climb to 4000rpm almost instantly. Jarrett did manage to retrieve codes, though, but there were too many to count.

There was a slight hiss near the IACV so the engine was killed and the IACV removed. The old o-ring was only halfway there so it was replaced with a spare nitrile o-ring. It didn't change much when the engine was restarted, but there was a reduction in the number of codes that were present. This could have been due to us clearing them and rescanning and completely independent of swapping the o-ring on the IACV. However, all 5 codes that the ECU was throwing were IACV-related.

The codes were cleared again and the car restarted. I opened the heater to circulate coolant. Once the temperature was high enough for the thermostat to open we added more coolant. The idle was a LITTLE better than it was in the first video I posted at this point, but still rough by all accounts. The main differences were that the car was slightly less sensitive to the accelerator pedal and idled at around 1500rpm. It would also buck and nearly stall when placed into gear. My assumption that this is due to an increased load. I cleared and rescanned for codes and got one code (P01507). Still an IACV code. Nothing appeared wrong with the IACV and it ran perfectly fine beforehand so I doubt that this is a true issue.

Jarrett went back to Dallas Monday night so everything else is just me again.

Now that it was idling somewhat I decided to drive it around the block. It had very sluggish performance and difficulty applying the brakes. The performance I could have guessed based on how it was idling and the brakes weren't much more confusing. The pedal effort was made more difficult by the low vacuum at the intake manifold. I pulled the booster hose off of the intake manifold noticed that vacuum was very weak.

A technician friend came over a bit later and discovered that injectors 1 and 2 are not getting a pulse. He's here now and we're looking for issues related to that going back as far as to look at the wiring at the ECU connectors. As nothing has changed in regard to wiring I don't see how that could be an issue.

Questions:

Do you want me to have him read voltage at the coil connector and TPS still?
Is there anything else this evidence would point you toward?
The technician friend can come over fairly easily and has lots of testing equipment. Is there anything I should have him check, specifically?

[nepbug]

Checking the coil and the TPS should be quick and will tick things off the list.

Have him test the plug wires too. The FSM has the acceptable values listed.

I still would suspect that IACV a bit, something seems off there. Have you cleaned it recently?

After that, I would check cam and crank position sensors.

Sent from my VS985 4G using Tapatalk