Car won't start: Troubleshoot

[ivymike1031]

(cross-post from af general: engineering/technical forum)
I thought that some of you might enjoy this, so I'm posting it.

The registration ran out on my girlfriend's car last saturday, and we couldn't re-register it until some paperwork showed up (problems with the IL DMV), so the car spent the weekend in the garage, except for a brief period on Sunday when I drove it out into the driveway to clear some space in the garage for some other projects I had going on (pulled back in a few hours later). Yesterday (Wednesday), we had the paperwork taken care of, and my girlfriend was ready to start driving the car. She reports that when she got in the car and tried to start it, the engine sounded normal for a second, then made a funny sound and started sounding really strange. I wasn't there while she was doing this, but I can confirm that when I got home, it sure sounded odd while cranking. The sound wasn't like it had absolutely no compression, but it also didn't sound like the starter was running against much resistance. The engine was clearly NOT in the mood to start.

The vehicle is a '97 Honda Civic dx, with 125,000 miles on it. The engine is a SOHC non-Vtec 1.6L D-series (D16Y8 if I'm not mistaken). It has been impeccably maintained, for the most part. I did succeed in getting the car to run again, and now it shows no signs that there ever was a problem. I'm not entirely certain (even now) of what the culprit was, but the one explanation that made sense to me was later confirmed as a common problem by a guy who has done engine development for 20 years, so I think we have a solid theory about what the trouble may have been. The point of this post is to allow y'all to take some stabs at it and see what you come up with. I did quite a lot of tests before I got the thing running, so if you'd want to check something in the course of diagnosing the problem, go ahead and suggest the check, and I'll give the closest thing that I have to the desired test results.

Beginning condition: the above information. Car doesn't start, and it sounds odd while cranking. Doesn't sound like it has no compression, but sounds like the starter isn't struggling much either. Car ran perfectly the last time it was started, which was 3 days earlier (no roughness in the idle, no hesitation on acceleration, no strange noises, etc...). The car has been garaged since the last time it ran. Temps in the garage have probably ranged from the high 50's to low 90's, with moderate humidity (I don't expect this to have a big bearing on your diagnosis).

Troubleshoot away....

[b16a3sol]

the first thing i want to know about is the compression test. with enough mentions to it, there has to be some results from that. then i would want to know about fuel and spark. after that i would move to timing. lets see where this gets us. the funny sound tells me that someting got dislodged, but i have no idea what kind of funny sound it was. im guessing to motor was cranking, rather than the starter just spinning against nothing, or a disconnected crankgear since it says the motor sounded odd while cranking. this is a strange sounding problem, but hopefully someone in here can guess correctly.

[ivymike1031]

There wasn't a particular sound that I can describe, it just didn't sound normal while cranking. There wasn't any kind of clicking, thumping, popping, etc., but it just didn't sound quite right.

* fuel rail was pressurized

* spark was present & strong at all four cyls

* valve timing (cam phase) was set correctly, and I re-tensioned the timing belt just in case

* compression test showed the following:
cyl 1: 90psi
cyl 2: 30psi
cyl 3: 60psi
cyl 4: 155psi
I should add that the compression test results were somewhat fishy, due to the poor quality of the compression test instrument. It is difficult to be sure that the device is properly seated & sealed, but I tried to be careful about getting it in there.

[b16a3sol]

compression appears to be the problem then. since it seems to be spread across more than one cylinder, it has to be something they all share. head gasket is what i would think it is, but it would be weird for it to just blow up like that. also a hole developing in the block would be unheard of. i dont think they would all have their piston rings fail at the same time, unless this was a problem that took a while to notice. something might have gotten jammed in the valve train, like bits of oil filter, but that would also be strange for them all to get stuck, while lucky number 4 got nothing. i really have no idea where to go with this one, and im glad you were able to figure it out. post some more clues so that someone can hopefully stumble on the answer by the end of the week.

[ivymike1031]

I think you're right about something sticking the valves open, but I'm skeptical about it being bits of oil filter (not sure how they'd get to a place where they could do that). It's also not a precise statement to say that I was able to figure it out, since I don't have definitive evidence of what the problem was.

As I completed the compression test on cyl #4, the car sounded like it wanted to start, and I went ahead and let it start with one plug out. I shut it off a moment later, put the plug back in, and fired it up. It ran great, like nothing had ever happened, with no check engine light, rough idle, or anything... The only explanation I could come up with was that somehow multiple valves had gotten stuck open. Checking valve lash would have confirmed that hypothesis, but I didn't bother doing it while the problem existed, and there was no point once the engine was running again.

I went into work the next day, and sauntered over to the desk of a guy who's been doing engine development for over 20 years, thinking that I'd stump him on this one. I described the problem, and he said right off the bat "sounds like some of your valves jammed open. I've heard that's a problem with some Honda engines." He then proceeded to tell me about a Chrysler engine with a similar problem that he'd done lab tests to identify. He said that in testing, they could pretty repeatably jam valves open in 5 out of 6 cylinders on a V6. He said that it was one of the wierder things that he'd ever seen, but sometimes carbon deposits in many cylinders would come loose all at once, and the valves would jam. After the deposits had built up sufficiently, they could shake large amounts loose simultaneously, usually after a few very short engine cycles (like when I moved the car out of and back into the garage).

Anyway... the problem no longer exists, so I suppose you could say that it has been solved. If a similar situation ever comes up, remember to check the valve lash (but DON'T adjust it if you find that it's way too large, unless you can be sure that the valves have seated properly).

[b16a3sol]

the part of the valve train i was thinking about was the rocker arm, spring, cam assembly. from there a shitty fram or other variant of oil filter could break up and either keep the rocker from following the lobe correctly or keep the spring from doing its job correctly. both of those parts are submerged in oil constantly, allowing this situation to happen, i would think.

thats pretty weird that all the deposits would shake loose all at once, except for in one cylinder. maybe its time to switch to chevron with techron . i cant really think of any way to prevent that other than the fishy route of a fuel additive. the cure also seems to be as simple as turning the motor enough times for the deposits to get pulled out of the system. outside of turning the motor by hand and spraying compressed air into the intake manifold with the exhaust valves open, i can see no other way of handling the problem.

good post, i learned a lot from it. keep up the good work.

[rx racer]

sounds like the battery is going dead. try jump starting the car.

[ivymike1031]

what are you talking about? The car cranked over just fine!

[rx racer]

HA HA I missed last couple posts