Hi, I have a 97 Civic with 1.6 engine. It has very low compression, test done cold ( no start ) on three cylinders ( #2 = 40, #3= 60, #4= 70 psi), and #1= 135. I put a tablespoon of oil into the low holes, and it didn't raise the numbers much( 8 psi on one , zip on the other two). I'm thinking valve job ( or bad head gasket, but no evidence of oil in radiator or coolant in the oil ) is most likely. It just seems a little odd that one cylinder would be that much higher than the others. Just a little background, the car was running decently( amazing I'd say given those numbers ) until a few days ago when I attempted to start it and it coughed and then refused to do anything but crank after that. I haven't done a leakdown test, should I? Any ideas welcome.

 

check timing? even your highest reading is still really low

 

did you own this car while it ran fine?

overheating and head lift will cause low numbers across the board

 

Did the timing belt slip, strip, or break?

 

The valves are not sealing properly would be my guess. Along the lines of what RonJ was thinking. Timing belt failure and a couple cylinders tore themselves up?

 

92shade: I put #1 at TDC and it looks like it's right at or close to where #1 plug is at dist , so it's not going to be too far off, I think. I know 135 isn't ideal by any means, but if they were all near that I'd be a happy camper. The relatively huge diff between #1 and the rest just seems strange to me, but I don't have a lot to compare it to, maybe it isn't strange.

Linked: Yes, I've had the car for about a year and a half. Never overheated, lost coolant,etc. , nothing abnormal that I could detect. What is head lift, never heard the term, is that detonation?

RonJ@HT/ EJ8 Man: Timing belt hasn't broke( camshaft still turns), but it may have slipped, will be pulling the covers tomorrow to inspect. I'm hoping there's no serious damage, but should have a better idea in that area when I can inspect the timing belt. If it did slip, I think it probably happened during the first failed start attempt a few days ago, and fortunately ( I hope ) the engine did not actually fire.

Will update after yanking the covers ( and maybe the head ) tomorrow.

 

when a car over heats, the head is made of aluminum and low melting point, so if a engine overheats to a high enough temp, the head can warp and not seal the cylinders

 

Ok understand, learned something new, but again if it ever did overheat, then my temp gauge is lying ( possible ), and I would think that there would have been some boiloff of the coolant, am I missing something there?

A bad thought occured to me that the reason that #1 is still functioning relatively decently, is that the other cyls croaking(after belt slip ) shut the engine down before the firing order got back around to #1. Don't know how that theory holds up, but it's all I've got at the moment. Will know more tomorrow.

 

sounds like belt skipped a tooth or two. shutting it off before it hit back on number 1 cylinder would be quite hard to do if possible. seems weird u have compression in one cylinder and almost nothing in the rest. it sounds like u need to pull the head and inspect anyway. head gaskets are cheap at autozone and so are timing belts. the belt could be stretched some too. the 135 ur getting in cylinder 1 is because the engine is cold and not warm as it should to do the test. cross ur fingers no valves were bent

 

Agreed it's farfetched that the engine would shut itself down fast enough to spare #1, but it never really started just hiccuped, and it's all I have to go one at the moment that might explain the readings I'm seeing. The service records show the belt being changed around 50K and 8 years ago. So, mileage is ok, but age is a little long in the tooth, should have inspected it sooner in hindsight. Oh well, I'll get my chance to do that now. Good point on the engine being cold, generally speaking unless the numbers are really low, the relative readings cyl to cyl are more important, to me anyway.

 

Well I got the valve/upper timing covers off, and surprise, surprise the belt is fine, no missing teeth no obvious signs of wear. If it skipped a tooth I don't think it's the belts fault. There is a little deflection( maybe 3/8”) between the cam sprocket and the water pump pulley, not sure if this is too much or ok. Trying to get an accurate gauge of the alignment marks on the cam sprocket is a real b*tch. The “UP” mark is up, and I measure from the top of the cylinder head right? Is there some trick I'm missing here? I have a bad feeling I'm going to be tearing the head off, but at this point I don't see any alternative.

 

 

Thanks for the pic, it's the same one I have. Amazing how much easier it is to use those marks when you use the ones on the backside ( facing the engine ), which I found by accident. They should tell you this in the manual which of course they don't. Anyway it looks like I'm hosed with the cam pully marks lined up with the head, the crank indicator lines up with the red ( timing ) mark on the crank pulley instead of the white(TDC). Looks like we have a winner ( or loser in my case ). Note to self, never, ever buy another car with an interference engine, without a doubt one of the most brain dead engineering feats of all time( appropriate for a race car maybe but for a passenger care ridiculous ). Thanks for all the assistance, I've got a job ahead of me.

 

I totally agree. I should have mentioned it with the diagram I posted.

The next step is to remove the crank pulley, install a new timing belt with the proper mechanical timing, and then compression test the cylinders.

Best of luck. Here's hoping there's no engine damage.

 

Thanks, I've got my fingers crossed, but I'm prepared for the worst. I'm calling Autozone in the morning to see if I can borrow a pulley puller and get a belt/tensioner. I'll post the results, good or bad.

 

Something does bug me a bit, if the mechanical timing is off by that much, why is #1 as high as it is versus the rest? Something doesn't add up, or it does and there is damage to the valves on 2,3 &4. Oh well, will find out tomorrow I guess.

 

After some delay finally got the crank bolt off. The belt is virtually pristine, no evidence of damage to gears or idler. There was approx. 1" deflection on the long run of the belt. Still unclear as to why the belt slipped.

 

I don't think there should be that much deflection.

 

I'm trying to find a spec for how much deflection is allowed? Anybody know?

 

Finding conflicting info. From .5" to 1". Do you have a helms manual?

 

I was using a Haynes, nothing there on timing belt deflection. Then I just found an online reproduction of the Honda Service Manual for that year, nothing in there either that I could find regarding deflection on timing belts. If I'm reading correctly, it seems that if the timing belt is not worn, cracked,etc., and the tensioner is set/operating properly, then that everything takes care of itself, as in you don't measure anything.

 

an inch is pushing it ... if u buy a new belt and install it u will barely get any and they're only 20-30 bux at auto zone.

 

Measuring belt deflection is unnecessary if you tension the belt exactly as described in the service manual.

 

apexi_rsx: Yea, I was planning to replace it anyway when I started, it's just that the belt is in such d*mn good shape I almost hate to toss it. New one's $34 at Auto Zone, no point in taking a chance even though this car is on Craiglist as soon as it's running again.

RonJ@HT: That's the conclusion I came to, thanks for confirming it.