2008 timing marks don't line up with old belt but car ran fine?!

[chim chim]

I've got a real mystery here with my timing belt. I had just replaced my (TAC) throttle actuator control and PCM and the car was running fine. No codes, smooth idle, and good power. I pulled it into the shop to replace the timing belt (68K).
As I was turning over the engine with a wrench on the crankshaft to line up the marks, I got resistance about half a revolution through and the engine wouldn't continue to turn over by just using the wrench.
The marks weren't close to being lined up yet, so I made the fatal mistake of using the starter to roll over the engine to get the marks close. Well, I probably bent the valves in the process and it was a stupid mistake, but at the time it didn't even cross my mind that the marks wouldn't line up. I mean I had just driven the car.
Anyways, as I line up the camshaft marks with each other, the crankshaft is about 5 teeth off, roughly 90 degrees. I mean it's not even close.
And yes, I know the crank revolves twice for every time the cams rotate once.
Well, I've never had this happen before. How the heck did the car run? The original belt I'm taking off isn't broken nor missing teeth nor does it have any oil on it. It looks really good.
I'm the original owner of the car and have never had any trouble with the timing belt and this is its first replacement.
After I realized my probable mistake, I put on the new timing belt anyways. I lined up the timing marks properly after taking off the old belt. Once lined up and with the new belt on, the engine rotated freely all the way around with a wrench as it should.
I hoped I might have gotten lucky and tried starting it, but no, the car wouldn't start. It just turns over. I have spark and fuel pressure.
So my next step is going to be a compression test. Assuming my compression is next to nothing in one or more cylinders because of bent valves, I'll remove the head and examine the valves.
But how in the world was the timing of the original belt off in the first place when the car was driving fine. The only possible explanation I've heard so far, is that the crank bolt and the serpentine belt pulley held tension on the crankshaft timing belt pulley, and that when I loosened the crank bolt to remove the serpentine belt pulley, that tension was released from the crankshaft timing belt pulley and allowed it to move.
Any ideas. I'm at a loss.
Thanks,
chim chim

[avguy]

What method did you use to loosen the crank bolt, and were there any issues getting it to go?

[chim chim]

I used a 120v plug in, 1/2 inch drive, impact wrench with approximately 320 ft/lb torque. The crank bolt came off right away, no issues. I didn't have to sit there and rattle on it. But I see where you're going with this. Hmmm...

[avguy]

I've always suspected that using an impact tool on a free-wheeling crank bolt could cause something in the belt loop to jump. I have no proof of that, because I've always locked the harmonic balancer in place before removing or installing the crank bolt, so this is JMO.

But consider what's actually happening. A huge amount of 'instant' counter-clockwise torque is being applied to the entire timing belt loop, which jolts it in the opposite direction that it was designed to move. The engineer who designed the tensioner probably would not have felt the need to build it to handle a huge force in the opposite direction, and IMO it seems likely the tensioner could momentarily bounce, lose tension, and cause the belt to skip, until the tension returns to reengage the belt. Just speculation, and I have no plans to ever try to prove that theory.

But the reason is immaterial at this point for you, and best of luck with the valve job.

[chim chim]

I think you're onto something here. That would be my best guess at this time as well. Having gone through this experience, it is the only thing I can think of that makes sense. It would explain the car running fine beforehand and the timing marks being off when I tried to line them up. Thank you for you time and opinion on this matter. Much appreciated. chim chim

[zchandran]

I'm not understanding the problem. The number of teeth on the camshaft is not an integral multiple of that on the crankshaft. So, for example, suppose you had 40 teeth on each of the (equal) camshaft pulleys, and 30 teeth on the crankshaft pulley. If you rotated the camshaft pulleys one entire revolution so that their marks aligned again, your crankshaft pulley would be off by 10 teeth. Or am I being stupid and missing something?

Using an impact wrench is how every engine builder I know does it. The inertia in the camshaft would keep the force from going into the belt. I take my lawnmower blade off with an impact wrench all the time - the blade does not even move an inch. An impact wrench absolutely will not make your timing belt rotate.

[JB_Fraser]

Just did mine a few weeks back, 21 teeth on the crank pulley and 42 on the cam pulleys.
wouldn't work otherwise.

Cheers

[zchandran]

Lol, never mind... I work with a lot of machinery that uses belts, so forgot for car engines it HAS to be double...