Rep Power: 0

2005 Civic, 159,000mi

Got a Check Engine and the DTC was "Number 2 Cylinder Misfire" accompanied by minor engine shake. It occurs right after engine start and lasts only about a minute or less, after which it smooths out perfectly.

If I reset the code, it will out stay generally until the next day and a cold-start (though occasionally it will happen w/ a hot start as well).

My first thought was a plug or coil, so I pulled the plugs (all double platinum in great shape with perfect color), checked the gap, and swapper #2 plug into #1 cylinder (to see if the problem followed the plug, or stayed w/ the #2 cylinder).

I also swapped #2 coil into #3 cylinder, so that:

If the problem was the PLUG, the misfire would now show #1, and if the problem was the COIL, the misfire would now show in #3 cylinder.

RESULT: The misfire remained with Cylinder #2

This made me think the problem may be the injector, so I swapped #2 injector with #3 injector, thinking if it was the injector, the misfire would follow the bad one to #3.

No such luck, and the Misfire DTC Code remains #2, and the engine has the same little shake, usually after start.

In my mind, that leave wiring problems or some mechanical problem specific to #2 (compression, sticky valve, etc), but the engine is ordinarily SO smooth and there are no ticking noises even when unsmooth that I tend to doubt a mechanical problem.

Anyone with suggestions, or a similar problem that was found?

Thanks,
​​​​​​​Bob

 
Rep Power: 518

Post unghosted.

Is it losing antifreeze?
Pushing antifreeze out of the radiator and slowly overfilling the reservoir?

Cold miss for only a minute makes me think antifreeze may be getting into the cylinder. Head gasket would be the usual cause, and it should be having those associated symptoms if you look closely.

If you don't find any of the above, check valve clearance adjustment.

 
Rep Power: 0

No coolant use I can see (or white steam in the exhaust), but of course, it would only take a few drops that leak past a questionable head gasket into #2 that when it vaporizes causes a bit of a stumble to set the DTC.

If it's the head gasket, I suspect it's at the very beginning of its failure where the weep might be very little.

That occurred to me, but I tried to pass it off as a bad dream.

I have a coolant pressure test kit, so maybe when the engine is still hot, I'll carefully relieve the pressure, hook up the tester, and pressurize the system to 15psi or so and see if it holds pressure. I also have a video probe I can lower through the spark plug hole to see if after pressurized for a long time if there is any coolant sitting down in the cylinder on the piston.

The valve clearance is easily checked as well.

I have a cylinder leak-down compression gauge, but it won't work with deeply recessed plugs like on the Honda unless I make an adapter, which is doable as well, I guess. Left long enough, I guess it woiuld make bubbles in the cooling system if there were a leak.

A cranking type compression gauge (which will thread down the deep plug well) likely wouldn't reveal a tiny weep in the head gasket.

Thanks-

 
Rep Power: 518

Pressure will always drop as the engine cools. You would have to pump it up several times throughout the cooldown period..




You could wait til the engine is actively missing immediately after startup, shut it off and remove the spark plug and check if it is wet at that time. Figure out what it is wet with, fuel or coolant... (If you let it run until it stops missing you won't find the evidence on the spark plug)

A breach has to be quite large before compression is affected. Most of these in the early stages or failure will slowly pressurize the cooling system until enough liquid is displaced out of the radiator to cause noticeable symptoms like overheating or loss of heater output.

To pressurize a cylinder I use a whip from my compression tester kit. Just remove the Schrader valve core from the end and connect to air line (the shops quick couplers just happen to fit)

See 2nd video in 1st post in this thread link:
https://www.civicforums.com/forums/3...reference.html

 
Rep Power: 0

Thanks-

Yes, I could remove the whip end of my compression gauge and hook it directly to air source and look for bubbles.

Since it only misfires for a few seconds (much less than a minute), if the engine is cool there may be evidence on the plug if I get it out immediately (they all looked absolutely identical when I removed them, so whatever evidence might eventually might stay if the leak were significant, plus it never seems to do it when I am in a position to have tools and time at-hand).

There is no air (combustion gas) in the radiator that can be observed, but maybe leaving 150psi hooked up longer term would show.

Yes, w/ regard to the pressure tester, I would have to pump it up as the engine continued to cool. I guess I could use a regulated air source to continuously pressurize it (compressor through regulator to cooling system cap adapter) then just check #2 with the video camera, though I suspect at this stage the heating/cooling cycle at this point may induce any HG leakage (undoubtedly it will get worse if ignored), so doing it on a cold engine may not produce enough leakage to see.

When the engine is hot, there appears to be no significant leakage (yet) since there's no coolant loss, nor evidence of gasses in the coolant I can see.

There used to be (probably still is) available test strips to dip into coolant to look for chemical contamination by combustion gasses. Never used them, but if there is gas leakage, it's no doubt from #2 cyl.

The bubble test in the video was good, and we'll try that, however! That's much better test than just looking for coolant use at this stage.

Some good ideas, and I'll try 'em and report back for sure.

Thanks again-

 
Rep Power: 229

Since it is only at cold start i'd side with ezone on coolant leaking into the cylinder while parked, or, oil.

You mentioned there was no loss of coolant or exhaust gas in the rad.

I would suggest either using a bore scope in the cylinder before starting it; after leaving it overnight.

or

Use a long rod with a sponge secured to the end (maybe a very small sponge that will not get stick, secured with epoxy). Dip the rod into the cylinder after leaving it overnight and see if oil or coolant soak up.

 
Rep Power: 0

Now, the stumble is also after a hot start (car sitting 15min or more).

I think the bore (video) scope makes sense.

Also found the cooling fan sensor had a broken pin, so wasn't running unless the A/C was on. Jumping the pins at the harness connector does make the fan run, however.

I am helping a buddy with this job on his car, and he admits now the temp gauge has pegged on HOT a few times, even at 50mph on a level road, yet the radiator cooling fins are not clogged or compromised, leading me to think clogged radiator, partly closed thermostat, or broken blades on a water pump that has plastoc impeller (I've never seen a Honda WP with a plastic impeller, though).

At idle, the temp starts climbing even w/ the elec fans spinning, and when the top rad hose measures 230deg, the lower one reads only 130deg. I doubt the fans could cool 230deg water by 100deg if a full flow of water was flowing.

Also, if a full flow of 130deg water were entering the engine at the bottom, it shouldn't reach 230deg at the eng outlet I wouldn't think.

I am tempted to replace his radiator and thermostat (as well as the temp sensor) first as right now he cannot let go of his car long enough to do a head gasket job (he and his wife both have jobs and only the one car), and see if it can limp long enough to when he can do a head gasket.

Sound reasonable?

 
Rep Power: 229

Ya with those symptoms and especially on cylinder 2 which is on the far side from the radiator (which cools) but also in the middle of the block (which warps); you're probably looking at a blown head due to radiator fan switch failure, thermostat failure, or both.


Good news though it is possible to run these cars with a blown head gasket for over a year, at lease i could with mine before i rebuilt it.



-I would suggest leaving the fan jumped to ensure it cools as much as possible especially while idling at a stop.

-replace the thermostat because it's cheap so why not; do a proper air evacuation from the cooling system after replacing the thermostat and coolant.

-the radiator should be fine, but if you think it is clogged just test it by removing both radiator hoses and seeing if you can get tap water to flow through it.

-then make sure the owner checks the radiator level before every start.



***To evacuate the air from the coolant system (you probably know how but if not...)

-leave the cap off

-fill the radiator to full

-turn the heat selection to max hot

-but leave the fan at zero so it does not blow the hot air into the cab (this is done to open the heater core but not let the heater core cool the system)

-run the car until the thermostat opens up and both radiator hoses are hot

-top up the radiator until no more bubbles are seen.

 
Rep Power: 0

Thanks- It all makes sense.

Given that my friend is in a tough place financially w/ only one car for two people and two subsistence jobs, trying to stretch things until he has a back-up car (he's looking for a cheap Element) makes sense.

I can get him a radiator, thermostat and temp sensor all for under $100 and hopefully he can limp a while.

I have one of the evacuation devices that makes a complete vacuum in the cooling system, then you introduce the coolant, and there is virtually NO air hiding anywhere in the system, so no worries there.

On the radiator, it may or not be the thermostat or the radiator, but if I'm going to drain the system for $60 extra we might as well put a radiator in for good measure given its 10+ years old.