Engine Trouble, backfiring, no start, please help
 

Incoming wall of text. I don't know what could be important and what is superfluous, so I'm including it all just in case.

I just did a lot of work on my manual '88 S4 (replaced the engine wiring harness, replaced the T-belt and water pump, replaced the power steering pump, replaced the radiator, and had the bottom of the block welded where the car had been bottomed out to re-accept the alternator).

I finished the steering pump last night, and I hooked everything up and resolved to test it this morning. When I came out, I started the car (needed a jump, old battery was shot) and I got low oil pressure warning. Turns out I forgot to connect the oil cooler hose to the cooling tank in the radiator and oil was pouring everywhere. I immediately shut it down, resealed the union, and cleaned up the oil. I ended up pumping about 5 quarts onto the concrete pad, but replaced it with fresh. Finally, I started the car up, and it seemed to go fine, the temperature was running right where it should, so I took it for a drive round the block. The car still worked okay (although the power steering was tough, I think I need to re-center the steering rack after replacing the pump), so I drove it to the gas station since I was almost empty. Turned off the car, filled it up, and that is where my problem started.

The car, of course, wouldn't start, but my experience had been that the battery was a goner, so I got a jump at the station and attempted to drive it home. Immediately upon starting it up, I heard a very loud rattling noise coming from the engine. The engine ran exceptionally poorly, and with my RMB sounded more like a Harley than the normal "muscle car" sound I am used to. I pulled otu of the station and had to stop about a hundred yards down the road to make a left, and the car died again. My dad and I pushed the car onto the side street, called my mom to come get us and give us a jump, and we repeated the process. At this point the car was running even more poorly, and to keep it running I had to continually give it gas, even at the stoplight. Of course, it died at the light anyway, and we had to push it to the side and re-jump. We tried this once more, and it died again maybe 200 yards down the road. I called a tow truck, and brought it home. I discovered a small fuel leak on the passenger side in the engine compartment, as evidently the hose that I replaced wasn't sealing properly, so I put a clamp on it to keep it from leaking while I sorted out the rest of the problem. I

I replaced the battery (although I was sure this wasnt the actual problem), and after tightening up the leak, I thought I'd give it a go, and when I did, the car massively backfired through the airbox, blowing out the metal screen material around the top of the MAF sensor. I wasn't sure what to make of that, so I disconnected the fuel and ignition computer and cranked over the engine to try to vent out the cylinders (I reasoned that if the valves were bent, I'd already done the damage and I wasn't likely to do any more by trying to flush the cylinders out in this fashion). Upon reconnection, the car still smoked from the MAF, and the car wouldn't catch and fire up.

I am sure that there is enough oil in the car, though I swear that sound sounds like hydraulic lifters with insufficient oil in them. I'm afraid that I can't remember exactly what the oil pressure was reading, it may have been a little low, but the idiot light hadn't come on. This makes me doubt any oil starvation problems. I think I read somewhere that rough starts getting progressively worse can be caused by a bad fuel computer, and so I might be tempted to chalk the poor running up to that (although I haven't yet tested that diagnosis), but I still don't see how this would cause my rattling noise. I had run the car a few times up on jack stands while repairing it (I'd had an overheating problem that turned out to be a bad coolant temp sender), and I *never* heard this rattling noise, and driving *to* the gas station, I never heard it either. Up until I turned the car off to fill it with gas, she ran like a champ. Any ideas?

One more thing. Idk if it's important, but in the past when the car had warmed up sufficiently, if I let it sit idle, I would wind up having low oil pressure and a low (~600 RPM) and fluctuating idle, with corresponding fluctuation in oil pressure, though always low enough to set off the idiot light. I am not sure how long I have had this problem, since I bought the car and the oil pressure gauge was pegged high. While I had the engine out, I went ahead and replaced the sender, and the problem was fixed. The car has always had rough idle problems, so I assume that I've also had this oil pressure problem (terrifying thought).

I've also just remembered that I pulled off the air pump when I pulled the engine, and I was unable to turn the pulley by hand (I'm not sure I could turn it with a lever and the pump in a vice). I assumed it was broken and left it off upon reassembly, figuring that if it was broken, it wouldn't be functioning properly and would just be draining my efficiency without actually cleaning up the emissions, not that I care since I get an exemption every year anyhow. Upon reflection however (assuming the stiff pump is the normal condition), I suppose the O2 sensor in the cat could be leaning out the fuel mixture since it isn't getting sufficient air, and a too lean mixture could be causing the rough running. Thoughts on this?

Sounds like a timing issue!

First thing to do is check your work. DONT START the ENGINE
First check the routing of the ignition wires on each cap.
If you find any that are mixed up then fix these and start the engine, If not THEN
remove both of the top timing covers and check the timing marks also see if the belt is running correctly.
Check that you have secured both of the cam to pulleys and then refit the rotors.
Please report what you have found

This engine has run multiple times perfectly well. This is a problem that has developed (somehow), and I am sure that I did the timing right the first time. The engine would not have run as many times as it did (recall that I ran it while I was checking up on an overheating problem, at least 10 separate times from cold to fully warm). I will check the cams to see if the engine is in time, but I figured the easy quick-test would be a leakdown test. As I said, the engine has been run under it's poor performance condition, so if any valves were going to be bent, they already are. I figure if the car passes a leakdown test, then the valves couldn't be bent (although I suppose the engine could still be a cam tooth off, though that would involve some slippage of a *brand new* belt that was previously installed correctly, so I am a bit dubious at that)


Thanks for the prompt response, and if anyone has any ideas, I'd be glad for your opinions/expertise. = )

just recheck your work, you will find the problem

Well, I did the leakdown, and i found the problem. The passenger side cylinders are all within +/- 5% of each other, and the driver side are all within +/- 3% of each other, unfortunately the driver side are all about 30% "leakier" than the passenger side. This leads me to believe that the timing cam on the driver's side has skipped some unknown number of teeth (at this point) and the timing is now out of whack. I don't know how close the timing is, but I suppose if it was only a couple teeth out of line on the cam, the valves might not be bent, but just slightly open at TDC. I've just thought of a way to test this, so ill BRB

Well, it's late and I'm tired. I'll have to repeat the leakdown test tomorrow to gauge the extent of the valvetrain damage. More info to follow, thanks for all the support.

Listen to Mr Merlin Line up your crank to TDC open your cam covers and look for the mark(s) turn only clockwise. If it slipped you need to find why before you run it or the belt could slip to the point where your valves will collide.

My initial thought was plug wire order, but then it wouldn't have driven ok to the gas station.

Sounds to me like the cam sprocket on at least one side has jumped a tooth or three on the belt? Possibly due to cam bolt backing out, or something else in the loop not being tight enough (tensioner, broken bushing on a roller, failed water pump etc.)

Certainly do not try to start the engine, or turn it over using the starter motor again before checking timing of both cams, belt tension and path - if you're lucky it hasn't yet jumped enough to bend any valves.

Yeah, I will do this. I will say that I think there is a small chance that the head gasket is blown on the driver's side, since those seem to be the leaky cylinders. My guess is that, since I didn't recognize the problem immediately for what it was and jumped the car and drove it for about a minute, the valvetrain damage has been done, however, I most certainly won't try cranking the engine in case, as you say, the cam hasn't slipped enough to bend valves. Thanks for the input. I have made a spreadsheet (saved in pdf format) of what each valve is doing when each cylinder is at TDC and firing. If anyone is interested, email me at veltman.thomas at gmail.com

usually the head gasket doesnt BLOW in these car as you would then have coolant into the cylinder, since the individual cylinders all are surrounded by coolant passages. . .
The more likely possibility of headgasket failure comes from the head gaskets leaking coolant out the sides along the head /block surface.
That said if you have a boosted engine then it could blow out the cylinder to head sealing area, and then you will have leaking coolant into the cyls

Hmm, then its almost certainly a timing issue. When I turn over the engine by hand, it feels jerky (it rotates easily for 15-20 degrees, then there is a sticking point, and when you apply slightly more pressure, it rolls over and makes a "clunk" noise which sounds like its coming from the valvetrain). I had thought that the lifters were hydraulic in this car, but I can't see how they would lose their oil so rapidly to compress and result in any audible noise. Is this further evidence of bent valves, or is loss of lifter oil pressure typically so sudden (or even related to a timing issue at all)?

Can anyone describe the smell of burning coolant? When the engine backfired through the MAF, I got a strong whiff of somethign that didn't smell like normal gas or even oil. Now that I think about it, it might have smelled like burning antifreeze, but I can't be sure. I suppose I could check the oil and see if there was water in it?

Coolant has a very sweet smell when leaking or being burnt.

Do you have the plugs out when turning it?

On the leakdown test you gave the % between the cylinders, but not what the leak down % was.

When I suspect one has gotten out of time I pull the cams, lock down the crank and do a leak down, most will show a 90% or higher number if it has bent valves.

You can do this with out taking the cams out, just ensure that the cams on on the base circle on each cylinder when testing each cylinder.

I have never seen a head gasket fail to the point that it took out all the cylinders on one bank.

The clunk does not sound promising.

Good luck

Well, the reason I can't quote absolute leakdown is that my gauge is set up to show relative leakdown, not absolute.

Also, engine rotates the same way with plugs in or out, but in this case, plugs were out