I just replaced the head gasket in my 1992 Civic. Started it up before putting everything back together (just a second since it had no coolant, exhaust wasn't connected, etc.) and it ran fine. But now after putting everything back together and filling the cooling system, it won't start. Turning the key, the starter motor runs, and the engine turns with virtually no resistance - so smooth that at first I thought the starter motor was just spinning without turning the engine.

What's the best way to go about diagnosing the problem? I don't see anything obviously wrong - no visible leaks, etc.

3 Answers 3


If it is spinning with no resistance, you don't have any compression. You should be able to test this pretty easily as you'll probably be able to spin it over quite happily with a spanner on the crank pulley. I can't see why that would suddenly happen when refilling the coolant system though, unless there was a problem with the new gasket installation.

What caused you to need to replace it in the first place? That might give some clues.

  • The gasket was busted at the exhaust side and spewing coolant out the exhaust pipe. Could a timing problem cause complete lack of compression? I forgot to tighten the timing belt tensioner all the way until after the first time I started it with coolant in the system, so perhaps the pressure of having to turn the water pump could have caused it to slip? Commented Oct 9, 2012 at 13:42
  • A timing problem can very easily cause a lack of compression, so that's probably your first suspect. Hopefully you didn't hear a nasty clattery noise the first time you tried to start it (valves open too late can easily hit pistons and bend...)
    – Nick C
    Commented Oct 9, 2012 at 14:05
  • In my experience (having been in a 4g63 powered Eclipse both times it had timing belt tensioner failures on startup), you'll never hear the impact. The engine just spins freely and won't catch... :-( Commented Oct 9, 2012 at 14:16
  • If the belt did slip and the timing is badly out of sync, how can I get it back in sync? Obviously I could just take the head back off and manually inspect the position of the pistons, and check to see whether the valves were damaged at the same time, but it'd be nice to avoid doing that if there's a chance the valves aren't smashed. Commented Oct 9, 2012 at 16:43
  • And if the valves are smashed, could I just replace the whole head with a head off a 1988 Civic? I think it's the same engine and I've got one sitting around that doesn't work.. :-) Commented Oct 9, 2012 at 19:04

The other answers and my suspicion based on them were right: the problem was lack of compression due to the timing belt having slipped. I took out the #1 spark plug and brought it to TDC, and found that the cam sprocket was off by 90 degrees from where it should be. After centering all 4 pistons to avoid interference with the valves, adjusting the cam sprocket to the "UP" position, and then moving #1 back to TDC, the engine runs again. I can't confirm yet that it's not damaged; it lacks power, but that could be from exhaust fumes in the engine compartment (2 of the studs that hold the exhaust pipe to the manifold broke taking it off) and/or old, dirty/corroded spark plugs.

  • 1
    Well done. A compression test should be able to tell you if there is any damage (and if you can change a HG, a comp test should be a piece of cake for you), but a blowing exhaust and corroded plugs will definitely sap power. 90 degrees out is quite impressive though, I'd have thought that if it were an interference engine that's be enough to totally destroy the valves...
    – Nick C
    Commented Oct 10, 2012 at 10:28
  • My guess is that because the belt wasn't tightened out of sync but just slipping, (1) it stopped slipping when it lost compression, and (2) if it started to hit a valve, that might have been just the resistance it needed to stop slipping. Or I could have just gotten lucky. I'm not sure the valves aren't bent/damaged at all, but it's at least running which is a good sign. Commented Oct 10, 2012 at 13:03

If the timing belt did slip due to lack of tension with accessories on it, the lack of compression may well be due to some smashed valves... I believe the Civics are all interference motors. See if you can confirm or deny that before you start pulling the head. Anyone know of an easier way to check the valves? Are there any strong indications of damage from just pulling the valve cover?

  • That's what I'd always been told too, but when I opened it up, the top of the pistons have grooves that seem to match the valves exactly. It seems like they wouldn't get damaged even if the timing was off. Commented Oct 9, 2012 at 14:04
  • It depends on how far the valves are pushed, more than just having grooves. For example, the Toyota 5S-FE is non-interference until you put in aftermarket cams (294 grind I believe?) that push the valves just the little bit further and makes the engine interference despite the cutouts. Some cars are like that from the factory. Commented Oct 9, 2012 at 14:18
  • The first comment here youtube.com/watch?v=HWnMxJp97-o makes me hopeful: "Timing belt is a couple teeth off. Ask me how I know... lol" Commented Oct 9, 2012 at 23:03
  • That reminds me... I had the timing belt jump one tooth on my 5S-FE powered car. It went from running great to running like it was only on 2 cylinders. Top speed of about 25mph... :-) Commented Oct 10, 2012 at 11:21

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .