Engine trouble diagnosis

[Northboundtrain]

Just got back from helping a guy diagnose hard starting issues with his '85 Jetta 1.6 n/a.  He bought it several months ago, drove it about 50 miles, and then was sideswiped.  The car had damage to the fender, hood, and bumper but no apparent structural or internal damage.  He drove it home from the accident scene and then the car sat for four or five months.  

Then he fixed the body damage and tried to start it.  It started very hard, ran a little while, and then wouldn't start after that.  He was able to pull start it once or twice, then he called me.

First, I checked the timing to make sure it hadn't jumped.  The cam was off a couple of degrees I would estimate, and the pump was at approximately 1.30 mm.  I got the timing right and we tried to start it.  It sounded like it wanted to go, but it wouldn't quite catch.  There were some little puffs of smoke coming out the intake while we tried to start it.  The starter turned the engine over quickly enough, and the battery is good.  Then I did a compression test.  200 psi in one cylinder, 270 psi in another.  We didn't bother with the other two.  

So obviously bad compression = won't start.  The question is:  what happened?  He said the car started fine and ran well before the accident.  I'm guessing that the pump and cam were thrown off their timing from the impact.  Would briefly running the engine at 1.30 mm with the cam off a degree or two ruin the rings or otherwise damage the cylinders?  Would it overheat the head/valves/pistons and cause warping?

He is trying to figure out what is likely wrong and what to do with the car without fully tearing down the engine.  

Thanks for any help.

[tylernt]

If the cam was off a little bit, the valves could have been tapping the piston tops. The drive home could have bent the valves slightly, leading to poor compression. The puffs of smoke out the intake could indicate one of more bent intake valves. I'd say that's enough to warrant pulling the head off.

[QuickTD]

At that altitude (boulder) 270psi may not be that bad. If you've been cranking it for a while the cylinders can get washed down from all the fuel and lower the compression as well.

 Sounds more like a gummed up transfer pump or other fuel supply problem. I wouldn't yank the head just yet. Put a hand pump or primer bulb on it and try again. Often the transfer pump vanes will free up with some running.

[vwmike]

Another thing to think about is that after it sat for a while some rust could have formed in the cylinders which would alter your compression reading but return to normal after some driving. It's hard to say really - just thought I'd throw that out there.

I'd double check all of the timing and if it's on you might put the injectors back in and put the lines back on but leave the lines loose at the injectors. Use the primer bulb as suggested and spin the pump with an impact wrench or something until fuel comes out of the unions at the injectors. Then tighten them up and put the belt back on. Leaky return lines can cause the system to bleed out and result in an extreme hard start condition. Trust me, it can take a long time to get it started after that happens.

[Northboundtrain]

I did crack the lines and verify that there is fuel to the injectors.  The pump is also newly rebuilt by Bosch, so I doubt it is a fuel supply issue.  I didn't even consider that the cold start lever could alter the timing reading, and I don't even know if it was in or out when while we were cranking the engine (duh!).  I guess I'll have to go back and double check that.  

But, if the lever was out, and I retarded the timing to 1.05mm with the lever out, then it still should start if all else is okay, yes?  

If fuel washing down the cylinder walls has lowered compression, then is there something that can be done to improve compression while trying to start it?  I was thinking of pulling the injectors and pouring a little oil into the combustion chamber, like you would do to diagnose the cause of bad compression in a gasser (and like you're not supposed to do when diagnosing a diesel because the engine might start!).

Thanks for the help so far

[chrissev]

I did crack the lines and verify that there is fuel to the injectors.  The pump is also newly rebuilt by Bosch, so I doubt it is a fuel supply issue.  I didn't even consider that the cold start lever could alter the timing reading, and I don't even know if it was in or out when while we were cranking the engine (duh!).  I guess I'll have to go back and double check that.  

But, if the lever was out, and I retarded the timing to 1.05mm with the lever out, then it still should start if all else is okay, yes?  

If fuel washing down the cylinder walls has lowered compression, then is there something that can be done to improve compression while trying to start it?  I was thinking of pulling the injectors and pouring a little oil into the combustion chamber, like you would do to diagnose the cause of bad compression in a gasser (and like you're not supposed to do when diagnosing a diesel because the engine might start!).

Thanks for the help so far

yeah don't put oil in the cylinders.  Diesels can run on engine oil if there is enough of it in them.  Regarding the compression, the engine has about half of what it needs.  I don't know about the issues that the other people were talking about regarding the engine having poor compression if it sits for a while (rusty cylinder walls?) or if there is diesel in the cylinders, but I do know that with that amount of compression you will have a hard time getting it going.  There won't be enough heat being generated for the diesel to ignite.  If you have set the timing wrong because the cold start was out when you did it that might add to the problem.  You're going to need to figure out a way to get around 400psi per cylinder again if you want that thing to run.

[QuickTD]

The compression is definately low, no question there. The thing that bothers me is why? What has changed since the car ran well? Engines don't usually wear out from sitting 4 months and don't bend valves while parked either. This is what leads me to believe something is just stuck or gummed up. A stuck valve could cause the low compression in one cylinder and explain the puffing out the intake. The engine should fire on a cylinder with 270psi if it has good glow plugs and is inside a heated building. I'd recheck the timing and have another go. A small amount of oil in the cylinders may help temporarily restore compression and get it to fire. It won't run way on a few drops, just don't check the compression with oil in the cylinders, your guage can be destroyed if the engine fires the oil.

[Black Smokin' Diesel]

This may sound dumb but when you did the compression check on one cylinder, did you make sure the valves were closed?

Seing how you didn't notice if the cold start was in or out it thought I'd ask  :wink: I think you timed the pump with the handle out and that's why it won't start. It's odd that the engine lost it's compression just by sitting. Highly unlikely IMO.

How can a sideswipe accident cause the cam and pump to jump on the timing belt?

[jtanguay]

i had a head-on collision (hit someone turning left who didnt bother looking)

it wasnt a high speed collision (unless 40-50km/h is with rainy weather)

the engine ran fine before the accident, but when I tried to start it after...  same symptom puff smoke.. it wanted to start and i would have kept cranking if the engine coolant didnt leak out

i would definitely say check the timing belt or even change it.  if your motor was running when you collided, and it shut off on its own... its a possibility

[tylernt]

Engines don't usually ... don't bend valves while parked

I think the intake valves got bent on the drive home after the accident. I can't think of any other reason for puffs of smoke coming back out the intake, but would love to hear other ideas...

[tylernt]

Hey here's an idea. Use a 17mm wrench on the crankshaft pulley to turn the engine over by hand. If it's silky smooth for two full revolutions of the crank, your valves should not bet hitting the pistons. If there are certain spots where the crank tries to stop or is harder to turn, your pistons might be hitting the valves.

[jtanguay]

Hey here's an idea. Use a 17mm wrench on the crankshaft pulley to turn the engine over by hand. If it's silky smooth for two full revolutions of the crank, your valves should not bet hitting the pistons. If there are certain spots where the crank tries to stop or is harder to turn, your pistons might be hitting the valves.

that could work.. just need to make sure you're not getting into the compression stroke

[tylernt]

that could work.. just need to make sure you're not getting into the compression stroke

Oops, good point. Pull the glow plugs out first.

[Master ACiD]

diesels cant pop out the intake unless there is something wrong with some of the valves. i would pull the head and have a look.

if the cam was off a few degrees, it wouldnt hurt anything. i have ran with my cam + and - 20 crank degrees from zero without the valves hitting. i have actually ran with the cam off a few teeth and still no valve problems, it just runs like crap and becomes very loud.

also these engines dont really bend valves in the traditional sence. because the valve is basically paraell to the piston, the valve wont bend the face like most engines. instead it tends to form a s curve like a snake while keeping the valve face level.

[Northboundtrain]

Thanks again for all the help.  My plan at this point is to double check the pump timing and reset it if necessary.  Then squirt a little oil into the injector ports and turn the engine over by hand a few times to work the oil around the cylinder walls.  Then give it another go.  If it still doesn't start, then I will assume bent valves and go from there -- maybe do a leak down test on the cylinder that's puffing smoke out the intake (I remember the number two in particular being the worst).

Any idea what 270 psi on a cold engine would translate to on a properly warmed up engine?

Thanks for taking an interest.