Gear marks vs pulley marks for timing????

After doing a leak down test I went to verify the timing. I found #1 TDC with a compression gauge and aligned the pulley mark with the zero on the belt cover.

 

I found that the crank gear marking to be about 60 degrees to the left, in stead up straight up. The cam markings were also about 60 degrees to the left of their timing marks. I verified the pulley was still on the zero mark and it was.

 

So if I move the crank and cam gear markings to their timing marks the mark on the pulley is no longer in line with the zero/TDC mark on the belt cover.

 

What the heck am I not understanding about timing this engine??????

 

EJ22-2.2L. Two pictures included.

 

Thanks, dmend

The mark on the block you need to line up is on the crank position sensor. You need to line up the crank sprocket mark with the mark on the block to be in time. That is NOT TDC. Subaru doesn't use TDC, it uses mid stroke or something like that, so if the engine is at TDC, it'd make sense that the timing marks don't line up. Also take the rubber seal off the block to see the timing mark.

The timing mark on the accessory pulley is for setting/checking spark timing and is at TDC as compared to the scale on the cover. The marks on the timing sprockets are for setting/checking cam timing. This is done with all the cylinders part way down their stroke so you don't bend a valve. It helps to call them assembly marks as compared to the accessory pulley having a timing mark.

 

If you go to do the timing belt you'll find out why when the driver's side cam jumps off the peak of the cam lobes and turns about 90 degrees. Most of the 2.2s won't actually hit the piston when this happens, but in the twin cam engines you can get the intake valves hitting the exhaust valves. If there had been a desire to modify the 2.2 engines later for better efficiency they could do so without having to change the timing procedure whereas if cam timing was done at TDC then they would have had to change the cam timing procedure.

Edited July 27, 2019 by doublechaz

I set the timing using the cam markings and crankshaft mark to line up with their respective marks. I replaced the intake gasket and all the vacuum hoses are new. The engine starts but barely runs and smokes. The vacuum gauge suggests late valve timing. I pulled the cam covers and both marks line up with the cover marks. I'm assuming the crankshaft is still correct. Also, I am not getting any OBDII codes. I verified the tooth count on the timing belt so I have the correct belt.

 

If there is no vacuum leak and the timing is right. Could the ECM be getting a signal from a component to cause late valve timing?

 

Pictures are after I rotated the engine four times, that's why the belt marks aren't visible.

 

Thank you,

dmend

Those assembly marks look right.

 

Does it still pass compression test? Easy check is to pull the injector connectors for all four so it won't start and crank it. Listen for the sound to keep even tempo and sound. Then if you doubt you can do a real compression test.

 

What color smoke? Black at the tail pipe?

 

Does the "barely run" sound even, random rough, or rhythmically rough? If it is even or random then that would indicate that one cylinder is not radically different from another.

 

How does the vacuum gauge behave?

It starts quickly, idles very rough and rocks/vibrates (like ignition wires are crossed), but they aren’t, and will barely stay running. Opening the throttle is like half the engine wants to accelerate and the other half wants to die. Previously, while idling I pulled each ignition wire and replaced each and found strong spark and Cyl 3,2,&4 there was no idle change. When I pulled cyl 1 the engine died.

 

Cyl 1 Cyl 3 Cyl 2 Cyl 4

Cold Compression Test 190 200 200 200

Leak down Test #1 20% 15% 15% 15%

Leak down Test #2 15% 15% 15% 15%

 

Vacuum gauge shows 7-9 in. Hg. and drops towards zero as the throttle opens.

 

Smoke is blue gray and holding a paper towel to the exhaust catches a lot if soot like debris.

 

Listened to injectors with a stethoscope and they were strong and consistent.

 

Background info:

Was running fine and one day on startup it began blowing massive amounts of white smoke. It was moist and sweet so I figured a head gasket blew. Engine has 180,000 miles on it so pulled the heads and changed head gaskets, but used same intake manifold gaskets (they were new earlier this year). Engine still was smoking but less coolant and more oily residue in tailpipe. Pulled the heads and had them resurfaced, new head gaskets and installed, also used same intake manifold gaskets. No coolant burning, but very sooty smokey exhaust, this is when the smoke was clearly blue/gray. It hasn't ran right since. Machinist said he had to take off 6/10,000. Replaced intake manifold gaskets.

 

Thanks,

dmend

That does sound like something radically wrong with cam timing. Is it possible the cam sprockets got swapped during all this? Only one of them has the nubs on the back to trigger the cam sensor on the driver's side. Our gen II OBDII system seems to not be smart enough to check sanity of signals, just open circuit type faults. When I was three teeth jumped I didn't get codes even though the computer was teleporting backwards and forwards in time as the conflicting crank and cam signals came in.

 

I don't have much hope that this is your problem because I'm not convinced that the ECU would fire the injectors and coils at all if there was zero cam signal.

 

I don't have good advice on an easy way to check this. To finally solve my problem I got a $30 oscilloscope kit, built it, and then looked at the waveform of the cam sensor, crank sensor, injector impulses, etc. You might be able to pull the cam sensor and see or feel in there for the nubs, but you would have to bar the engine over very carefully by hand. Any other way would rip the end off your finger. Maybe a good multimeter would be able to show you something from the cam sensor while you crank the engine.

 

The best would be a multi channel scope so you could see the relation between crank/cam/injector/spark. You might also need a correctly working gen II 2.2 to get a correct signal pattern to compare to.

Another thought, can you carefully look from above, and from the side with an inspection mirror to make sure nothing got trapped between the head and intake? That vacuum level is terrible. I get about 15 inches at idle, and I only have about 24.5 inches to work with so someone in a normal location should probably be looking at about 18 inches at least.

 

If you know someone local who could loan the use of a smoke machine you could be sure of vacuum quickly.

I"m pretty sure the cam gears didn't get mixed. But, you never know.....will check it out.

Your feedback has given me a path to pursue, better then throwing darts. Thank you.

 

Now, back to the garage....

dmend

Can you come up with a multimeter? If you can get it connected to the cam position sensor somehow and you crank it over it might show some activity on a good sensor/gear combo. I'm sure it won't read any activity if the sensor is dead or the gear is wrong. Better than taking a bunch of stuff apart if it shows something. But it might be a dead end test if the meter isn't quick enough to react to small signals.

The cam gears are on the correct sides. "L" is on the driver/left side head. I pulled the cam sensor and located the tabs that change the magnetic signal just to make sure they were there.

 

At this point I can't do the scope thing, but I like the kit idea-its on my radar....

 

Pulled both the crank and cam sensors. Showing 2.07 and 2.08 ohms. on the DVOM. Specs being 1K-4K ohms. When passing a bolt past them the ohms changed, though the readings weren't consistent, i.e., 1.92, 1.91, 1.65, 1.87, 1.15, 1.35, 0.77. It is very likely due to my inability to position the bolt exactly the same distance each pass. So, I tried placing the bolt on the face of crank sensor with similar results, inconsistent ohms readings.

 

SO...thinking back about setting the timing I think the belt on the crank gear and the crank mark is off. I found a picture in Haynes manual that clearly shows the crank mark in line with the case mark. I've been setting the crank mark at 12 o'clock.

 

All though I've counted the teeth on the belt I've never been able to get the center/crank belt mark directly under the mark on the case, always just to the left, about 1/2 tooth distance.

 

My next move is to go back and reset the timing belt and crank.

 

dmend

I really didn't expect this but the timing marks look dead-on. So if the timing is right, the crank and cam sensors are working, and as of yet I haven't found a vacuum leak, a rough idle could be caused by bad coil? Tomorrows task.

 

dmend

That is definitely the teeth behind the cam sensor.

 

In the previous shots the cams are lined up right for sure. The perspective on the crank makes it a little harder to tell. Perhaps it is one tooth off, but I didn't think so looking at it.

 

A coil problem could cause trouble. That one is easier to check with the DVM.

 

The cam and crank sensors I would say the main thing to check for is a pulse of voltage on the DVM while you move the bolt past the sensor face.

Edited August 26, 2019 by doublechaz

Everything checks out with the DVOM. Cam and crank sensors, a new subaru TPS, Fuel injectors are sound strong and consistent, fuel pressure regulator is good, IAC, Barometric Atmospheric sensor, fuel pump, EGR valve works, strong spark at all spark plugs. Under the assumption the valve timing is correct....

 

What's left is the issue of a vacuum leak and exhaust smoke.

The vacuum gauge at 400rpm idle reads 8 hg in. Crack the throttle and it will drop towards zero. With new intake gaskets installed with copper spray to help ensure sealing. New vacuum hoses and zip tied for sealing. I’ve sprayed carb cleaner, bundled incense sticks to create smoke and can’t find a leak source. I’ve even pressurized the intake manifold and did not find a leak. The question is: what else can cause low vacuum? If vacuum is caused by the pistons drawing in air and moving out exhaust then if the heads are leaking coolant into the cylinders causing poor combustion (low, rough idle and smoking) then there would be less vacuum in the system.

 

I had the heads resurfaced back when all this started and the machine shop said he took off 6 ten thousands and said there are still some dips. I think my next move is to pull the heads and check them for flatness and the engine case also. Maybe my worst case scenario will be only getting new/other heads and the case will be fine.

 

Thanks,

dmend

Bent or burned valves can lower compression which could be like low vacuum, but I think it would be missing pretty bad if that was the case (also I see your compression and leakdown test say it can't be this). I don't know of a way that the valve system could be messed up in a way where the valves barely open so I think that is a dead end.

 

I don't think it could be that the timing sprocket or cam sprockets got mixed up with another engine so the marks are wrong. By that I mean I think ours are all the same as long as you are staying within SOHC or within DOHC.

 

I saw a video once where everything was right on the face of it, but the cam sprocket locating pin was sheared off so the cam was not in alignment with the sprocket. I don't think that would work on our cars since the SOHC has two cams and having both like this is inconceivable. With just one like that you would be back to missing real bad. (I did get good compression tests once when one of my cams was jumped 3 teeth. It ran like crap, shook real bad, blew black smoke. I don't know what the vacuum was like then.) Maybe one cam pin could be sheared and lead to your symptoms after all.

 

I suppose if the woodruff key was missing on the crank you could have everything look right but the pistons would be off compared to the rest of the system. If the keyway in the crank or in the timing gear was really terribly wallowed out you could have it misaligned enough to do this, but no one would put it back together looking like that. The woodruff is just out of frame in your crank picture. It is unfortunately on the bottom when you are using the assembly marks.

 

I can't be sure from all the way over here, but I think the smoke you are talking about is unburned fuel from things being so totally borked.

 

I wish you were close enough to drive to. This thing is really a mystery.

Edited August 21, 2019 by doublechaz

Eureka!!

I think I found the problem. The right side cam is excessively worn on the intake lobes. Deep groves in them. The exhausts aren't too bad. The right side intakes not as bad but still some groves. The exhausts are about the same as the left. Won't be able to put a micrometer on them until they are extracted. I'll post the numbers.

 

dmend

Wow! That sucks. But it explains the crazy behavior. Turns out there is a way for the valves to barely open even when things are lined up right. Edited August 26, 2019 by doublechaz

By right cam do you mean passenger side? And is yours a phase 1 or phase 2 2.2? Cause I have a spare cam if you want it

Yea, happy and sad at the same time….

 

Intake specs: 1.2596 to 1.2635

Mine intake is: 1.2520, 1.2514, 1.2275, and 1.2514

 

I’ve been perusing the forums, parts stores and dealerships and it is not looking good.

Dealership says camshafts are N/A, can’t find anyone for a new grind camshaft , haven’t seen a performance cam for the 2.2L.

 

Surefire and ATK engineering make remanufactured heads but have Better Business Bureau reviews that aren’t glowing.

 

I don’t want to put 2.5L heads on this engine. Any chance you know what Subaru models and years would have interchangeable camshafts or heads with a 1995 legacy, 2.2L, SOHC, EGR, non-turbo?

 

This engine was originally paired with an ATM. What does the Manual Trans head cylinder have different than the ATM head.

 

Regards,

dmend

Thanks for the offer of the camshaft. As you can see I'm in need of both sides.

 

dmend

95 has two exhaust ports in the head. 96 up to somewhere had just one. You could swap, you would just need the exhaust Y pipe to match the heads.

 

I would look for a junkyard or other parts car with Auto and pull the heads out whole.

 

Perhaps Setnev will see this and answer. He seems to know every possible combination of mismatches that will bolt right up and work fine.