24 replies to this topic

[bearbalu]

I removed the engine and replaced HG/Clutch/Timing belt on my 98 outback.
I am experiencing some engine shudder/shake - feel it when I am stopped. Car runs great otherwise. Idle rpm is fine too.

Any ideas where to start?

[f15xxx]

if you have hydraulic lifters they may have drained some fluid during the TB change depending upon how much you rotated the engine to find TDC etc. in which case the car may run rough and/or noisy for an hour or so. this is what my literature says and i'm in the middle of a TB change myself.

99 leg l sedan 2.2l 5-sp

[bearbalu]

Thanks. They aren't hydraulic....

[svxpert]

I removed the engine and replaced HG/Clutch/Timing belt on my 98 outback.
I am experiencing some engine shudder/shake - feel it when I am stopped. Car runs great otherwise. Idle rpm is fine too.

Any ideas where to start?

timing off?

[craigusoz]

Did you put the clutch pressure plate back in the same orientation on the flywheel as it was in when you took it off?

[bearbalu]

Yes, I did put back pressure plate in same orientation with flywheel - no clutch shudder that I can feel - depressing clutch pedal makes no difference.

Timing - I doubt it though it is not impossible.

Why I think timing is right - First, the car runs great. Second, I counted the teeth on timing belt(from crank sprocket mark to cam sprocket mark, and intake to exhaust cam sprocket marks) and they were exactly as per spec. As I understand it, only reliable way is counting teeth. The Right cam marks aligned perfectly with inside right timing cover....

The "not impossible part" - The marks on left intake cam sprocket did not align perfectly with left inner timing cover - seemed off by less than 1/2 tooth. If I had tried moving to next teeth, I would be off by more than 1/2 tooth and more importantly, the teeth count would have been off.
Mark on left exhuast cam sprocket was off by even less - perhaps 1/4 tooth. The left intake and exhaust alignment with respect to each other (the double etched marks on sprockets) seemed off by 1/2 tooth in relation to each other but the factory paint mark(not sprocket marks) going across left intake and exhaust were in perfect alignment.

I also suspect that a resurfaced head and an updated head gasket (Subaru redesigned them) with different thickness can make material difference to alignment with relation to inner mark, since they impact the distance between the crank and cam sprockets.

[Nug]

Did you do just 1 headgasket? Was that head milled?


Did you adjust the valves afterward?

[bearbalu]

I did both heads, both were resurfaced - I know that blows a hole in my theory since right marks aligned, but left didn't. I did adjust valve clearances by replacing shims. Machine shop replaced valve seals while heads were being resurfaced.

[Nug]

If you counted the teeth, and they were right, then I think the timing belt is installed correctly.

Compression check? Bad plug wire?

[bearbalu]

I did have a CEL scare after HG replacement.


http://www.ultimates...ead.php?t=27635

I did measure compression on 1 and 2 and it seemed to around 200psi after 6-7 cranks. I didn't do 3 and 4 since they are harder to reach with engine in the car. Spark plug wire resistance measures up okay too. The plugs seem okay. I checked the injectors too and they were fine.

Somehow the CEL problem went away mysteriously.

Any other ideas? Could it be mechanical - I redid the pitching stopper but didn't make a difference. Engine mounts seem okay too...

[bearbalu]

To ensure that timing belt hadn't skipped a teeth, I took out the left and right timing covers (center cover is blocked by crank pulley). The timing belt seems still fine - The marks on cam sprockets are still as I left them when I first installed it. 105.5 teeth count between left and right cams which is spec for my EJ25 DOHC - Since I can't measure from cranksprocket, this is the next best thing.

Spark plug wires - I did the experiment in dark garage - no arcs visible even when I spray with water/lysol!

[RodA2003]

Might want to check the compression on the other 2 cylinders. I had the same symptoms after doing the HGs, and I had to reshim the valves to make the compression good on all cylinders, then it ran fine.

[bearbalu]

I already reshimed the valves when I took out the heads to do HG.

I might try compression test on other cylinders as you suggested.

[bearbalu]

How can I tell if engine or transmission mount could be the reason for engine shake ? I can only see engine mounts (the one sitting on front cross member) visually, can't see anything else without first raising the engine... Could loose transmission mount cause engine shake? How would I check that?

[WAWalker]

Did you adjust valves after you put the heads back on the engine, or did you do it with heads off? If the valves are adjusted with the heads off the clearance will change when the engine is assembled.

[bearbalu]

With heads on.


Nor sure about other cars, but on DOHC EJ25, you cannot adjust valve clearance with heads off - one has to bolt heads onto short block, put in valve lifters and valve shims and then put the camshaft...

[ron917]

In a pushrod engine, that might be true. In an overhead cam engine, like the one we're talking about here, that's impossible. The relationship between the camshaft and followers will not change.

On the EJ25 DOHC, the head bolts are under the cam shafts, so it makes no sense to do the adjustment with the head off the block. Unless you don't mind disassembling the thing again to bolt the head to the block...

[99obw]

I adjusted the valves on the bench when I did the headgaskets. I just torqued the cams into the heads and checked the clearances. I won't say that the clearance couldn't theoretically change when the heads are reinstalled in the engine, but the change would be extremely small.

[NoahDL88]

could it be as simple as a bad spark plug or a wire that came loose?

[bearbalu]

The car is running great with no CEL. Plug wires are tight.
Engine shake is the only problem, feel it at idle. As I said Idle rpms are fine too...

Could the shake still be attributed to a problem in ignition or fuel system? Wouldn't a problem in those trigger CEL?

[99obw]

The car is running great with no CEL. Plug wires are tight.
Engine shake is the only problem, feel it at idle. As I said Idle rpms are fine too...

Could the shake still be attributed to a problem in ignition or fuel system? Wouldn't a problem in those trigger CEL?

I think it would, OBDII systems are pretty good at detecting a misfire. Have you double checked the engine mounts?

[bearbalu]

How does one identify engine mount problems?

The engine mounts that fit on cross member are tight - I torqued them to spec. It is hard to see if anything is wrong with engine weight on them - visually they look fine.

Is there a way of identifying mount problems short of lifting the weight off the mounts - which would mean I would have to raise the engine.

[99obw]

I would just examine the mount closely with the engine running. You should see movement but it should be very limited. The only engine mount failure I have ever diagnosed was very evident, with a couple of inches of movement when gears were changed, so my knowledge is limited to that experience.

Have you tried pulling one spark plug wire at a time to see if the nature of the shake changes? If you find a cylinder where pulling the wire doesn't change anything, that's the one you want to examine more closely.

If you haven't I would do a compression test on all cylinders.

[bearbalu]

I would just examine the mount closely with the engine running. You should see movement but it should be very limited. The only engine mount failure I have ever diagnosed was very evident, with a couple of inches of movement when gears were changed, so my knowledge is limited to that experience.

Have you tried pulling one spark plug wire at a time to see if the nature of the shake changes? If you find a cylinder where pulling the wire doesn't change anything, that's the one you want to examine more closely.


If you haven't I would do a compression test on all cylinders.

Nope I haven't tried removing one plug at a time. You mean try to find a cylinder with no compression or very low compression? Won't very low compression also trigger a CEL? If a cylinder has reasonable compression (even 40 to 50 percent of other cylinders), I would assume that removing the plug wire will change the nature of shake.

I am thinking about doing it on cylinder 2 and 4...it's hard to get that tube into spark plug hole with engine in the car!

[grossgary]

COMPRESSION TEST
i think you'll find something.