(800) 853-2651
SHOP NOW

Subaru timing belt question . .

J

JDemaris

Well-known Member
Yes, not a tractor - but at least it's an engine.

I guess my question can relate to any engine with two separate cylinder banks and a timimg belt (V2, V4, V6, V8, flat-opposed Boxer, etc).

I just bought a 2002 Subaru Impreza with the 2.5 engine. So, it's a flat Boxer four - two cylinders per bank/side. One cam per head - so two cams total but called a SOHC engine.

Ran perfect but had 110,000 miles with no service work ever done. Still had original spark plugs, timing belt, water pump, etc.

OK, no big deal, right? I've done many, many timing-belt jobs - but this time I have a problem.

I installed a new belt and idlers along with new water pump. I thought I check timing marks correctly several times before putting back together.

Started it, and it now sounds like an old Chevy I stuck a hot cam in. Idles lousy, but smooth out at higher RPMs. I drove it and a drive-speed, it seems to run fine. But . . now a engine-check-light has come on. I checked the codes and it says "misfire on cylinders 1 and 3."
Note that 1 and 3 are both in one bank. So, if the timing belt was off a tooth for the one bank - it would effect only that bank and two cylinders.

I hate to second-guess myself, but also tend to believe in "cause and effect." It ran fine before, and now does not. Only other things I did, was pull out the old spark plugs - after this problem - and found the gaps at .090". I put in the new plugs gapped properly at .044". It did not fix this problem. I was a little worried that I might of screwed up a plug wire since this thing is a "hemi" with long plug boots that go into the head. They come out hard. But, I can't see two going bad all-of-a-sudden, and they check out at 5000 ohms each, which ought to be fine.

Anybody ever driven an engine with one bank slightly out of time (valve timing, that is) ?
 
if you dont have the wires crossed on 1 and 3 i'd say cam is a tooth off on that side...its not that hard to get em off by a tooth even if they look rite...almost gotta be a 3' tall midget to get in that engine compartment.
 
Its the most probable cause... remember paralax error when viewing the timing marks. Look them square on.

I've only done a couple of the older 2.2 sohc and it was with the whole front end off.
 
My x wife had the very same car. It was a solid dependable auto but when it came time for a timing belt I sent it to the dealer.

I dont farm out many projects but I remember looking under the hood and making the decision.

Considering that the car had only needed tires and oil changes in 5 years of hard driving the $830 I paid the dealer was money well spent.
 
My first thought would be like yours since you were in the engine I would be going back and verify everything as well. Having said that, I have never determined how a mis-fire is actually sensed. In my daughter"s VW Jetta it seems to be only related to ignition signals. She had a CEL on with a cylinder 2 mis-fire reported. Changed plugs with no effect, went on to wires still not getting improvement. Now codes are also reading a cylinder 4 misfire. Find out that in this ignition system cylinders 1 and 3 "share" a coil as do 2 and 4. Replaced the "coilpack" and CEL went away and mileage improved from 25 mpg to almost 27. Your situation just reminds me of this so I would remind/recommend that you go through ignition.

YMMV,

Kirk
 
Yes, I had thought about a possible coil problem. There is just one coil with four outputs. Seems unlikely though, that this problem would begin just after I put in a new timing belt. I did check spark at the coil output with the engine idling - and it can jump a 1/2" gap with a blue bolt of lightning.

I reluctantly searched through several Subaru forums - and found one post that's kind of interesting - if true. The guy there states that most aftermarket belts for my engine come with the left-cam mark one-tooth off i.e. marked wrong and puts just the one side of the engine out of time. If true, that IS my problem. My belt is Cloyes brand, not OEM from Subaru. Will having one cam, one tooth off cause a mis-fire code for those two cylinders? I have no idea at this point.

Only way to know for sure now is to pull it back apart and actually count the teeth between marks and check.

I have put in many timing belts with zero problems - but this is the first belt I've ever seen that comes with timing marks printed right on it -to make it "foolproof" I guess. I wish now it had no marks - because then I would of checked timing the old-fashioned way.
 
I'm not so trusting of other people's work. Also, putting a new belt and water pump in a non-rusty car is only a three hour job. If rusty, it takes more time.

I'm not saying there's anything wrong with sending the job out for those that choose to. Just not for me. There are enough things in life I can't do myself, so for the things I CAN, I do them.

I'm probably not up on the latest in regard to what shops charge, but I do wonder what exactly that $830 paid for? Belt alone, or replacement of all moving parts associated with the belt-drive.

I suspect a dealer would install all new idlers, a new hydraulic tensioner, water pump - regardless if needed, or not. From Subaru, I suspect that would be $600 just in parts alone.
If I was running a repair shop, I'd replace all those parts also, unless I could send out with no guarantees.

I checked all moving parts very carefully and replaced three questionable idlers also and put in a new belt and water-pump. Old water pump seemed perfect, but with over 100K, a new $35 pump seems like a good investment. Total parts cost around $150.
 
I guess this kind of gets back to the question I was asking before I started rambling. What is sensing a mis-fire? I was told that on my daughter's Jetta it was only ignition based so mis-fire codes are always ignition somewhere (on her car anyway). I don't know if that is accurate or not. Are the sensor(s) actually recording successful ignition/combustion or ignition only?

Hoping somebody will chime in and enlighten me.

Kirk
 
I've done a bunch of those timing belts over the years. Here is the problem. You need to use a timing belt with the marks painted on the belt. The marks on the plastic covers are not accurate. Hope this helps.
 
I was (and still am) wondering the same. I do know that error codes are not always universal, and can mean different things with different make cars. There isn't a separate oxygen sensor for each cylinder, so I'm wondering how a non-ignition problem could be sensed for just two cylinders? In theory, I can understand sensing a ignition trigger not working, or a fuel injector not firing. But, if valve timing causes - lets say . . . poor combustion on just those two cylinders - how would the computer know? That's rhetorical since I don't have a clue at this point - especially with an OBDII system.
 
That's what I did, and now wish I hadn't. I've never gone wrong by using hard-marks on the engine and gears. This time - I used the marks on the new belt, and I've been told now that many new belts are one tooth off on the left-cylinder-bank.
 
Give us an update when you find it.

I am still leaning toward ignition, but I am only a shade tree mechanic. If I had to make a living that way, I would be starving. Most of my repairs take about 3-5 times the book time values.

Good Luck,

Kirk
 
(quoted from post at 16:02:12 11/15/10) Yes, not a tractor - but at least it's an engine.

I guess my question can relate to any engine with two separate cylinder banks and a timimg belt (V2, V4, V6, V8, flat-opposed Boxer, etc).

I just bought a 2002 Subaru Impreza with the 2.5 engine. So, it's a flat Boxer four - two cylinders per bank/side. One cam per head - so two cams total but called a SOHC engine.

Ran perfect but had 110,000 miles with no service work ever done. Still had original spark plugs, timing belt, water pump, etc.

OK, no big deal, right? I've done many, many timing-belt jobs - but this time I have a problem.

I installed a new belt and idlers along with new water pump. I thought I check timing marks correctly several times before putting back together.

Started it, and it now sounds like an old Chevy I stuck a hot cam in. Idles lousy, but smooth out at higher RPMs. I drove it and a drive-speed, it seems to run fine. But . . now a engine-check-light has come on. I checked the codes and it says "misfire on cylinders 1 and 3."
Note that 1 and 3 are both in one bank. So, if the timing belt was off a tooth for the one bank - it would effect only that bank and two cylinders.

I hate to second-guess myself, but also tend to believe in "cause and effect." It ran fine before, and now does not. Only other things I did, was pull out the old spark plugs - after this problem - and found the gaps at .090". I put in the new plugs gapped properly at .044". It did not fix this problem. I was a little worried that I might of screwed up a plug wire since this thing is a "hemi" with long plug boots that go into the head. They come out hard. But, I can't see two going bad all-of-a-sudden, and they check out at 5000 ohms each, which ought to be fine.

Anybody ever driven an engine with one bank slightly out of time (valve timing, that is) ?
Click to expand...

Sum'n does come to mind,,, you are not gonna like what I am gonna say,,, I did a sub once,,, replace the cam seals,,, found out the cam rockers will fall out if the cam is moved just a tad out... I had to remove the rocker cover and put'em back in place... On another sub got the job all done and was put'n up my overpriced tools and found one new cam belt and one old belt :oops: Yep installed one of the old belts :twisted:

Did you get the instructions with the component kit,,, you have to follow them to a tee,, the auto tensioner is not really a tensioner,,, its a slack adjuster that takes up the slack of a PROPERLY installed belt over the belts life span,,, last thang you do is release the tensioner,,, at that time the belt show have already been set to spec ...


Cornfuses me to... probably sum'n about counting teeth also...
 
Here"s a web site that may help you. Hal


http://www.scoobymods.com/my01-rs-timing-belt-change-t3837.html
 
Keep us posted as to the final outcome. My wife"s 2003 Outback has 84,000 miles and I"ve been thinking of taking on the job of changing the timing belt also. Avoiding pitfalls is easier than digging out of them. LOL
 
Last one I did ,,I remember ,you had to line up one cam,then turn engine over 180,or 360 ? I can't recall for sure . But it did exactly what your running into now . Reread timing proceedure .
 
Last one I did ,,I remember ,you had to line up one cam,then turn engine over 180,or 360 ? I can't recall for sure . But it did exactly what your running into now . Reread timing proceedure .
 
I own one and had the WP fail (toothed idler stopped ripping off rubber teeth) The key to me getting it correct was found on page 2A-28 of the Haynes manual (page may vary).
The grief was keeping it on the point of the cam on an open valve at the perfect timing location. Be careful to keep all slack out of the drive side of belt. It is a crash valve engine, do not assume it can be turned with impunity. Be careful to not over tighten the mount bolts, Mine was stripped on disassembly and needed to be helicoiled. Jim
 
JD, Not to second guess your work, but did you start to wrap the timing belt opposite the tensioner? You can be off one tooth and affect engine vaccum enough to lower the total intake charge and cause the cylinder misfire codes and the rough idle. I've seen it on Chrysler's 3.5 v-6. After timing belt replacement did you turn the crank 720 degrees and verify mechanical time? I would start there first. Hope this helps. Gerard
 
Mr. John, my wife has a 94 Impreza. I went to look at something on it as a favor to you, and as it turns out you did the favor for me. I had left the park lights on and would have had a dead battery in the A.M. Now,to my answer. My subaru has a camshaft sensor on the left side cam. If there is one on the crankshaft, [and I don"t know if there is] the computer could compare the inputs and possibly determine a misfire.All this said, I agree with everyone that you are off on your timing marks. I hope this helps. I have learned a lot from you. Steve Norwood. Texarkana, Texas.
 
I noticed a sensor on the driver's-side camshaft, and also a sensor behind the crankshaft sprocket.

I've been told that with these sensors, and also a knock sensor somewhere, the computer can compare all that data - along with fuel injector pulses for each cylinder - and figure out a lot.

I tore it back apart last night (took less then half an hour). I will repost today if I fix it.
It appears that the right-cam is one tooth off and left cam is fine. This because the timing marks on the belt are incorrect. The cam that is one tooth off runs on cylinders #1 and #3 - the ones that now give an error code as "misfiring."

Right now - the issue seems to be mis-information. From all I can tell, the tooth count on the timing belt on the SOHC 2.5 is different from other similar engines (like 2.2s) - and the info is wrong in the factory service manual. Subsequently, new aftermarket timing belts are marked wrong.

I did a lot of searching and finally found two people who had the exact same problem. So, I took the original belt and examined with a magnifying glass. Sure enough, I found the old OEM timing marks and they do NOT match the new belt.
 
Took me one hour to take apart and fix. The problem WAS the belt being one tooth off on one side. The timing marks on the belt are wrong. Also - seems the information in the OEM factory service manual is wrong - but since I don't have one myself - I'm going by what others are saying.

I went to three Subaru-only repair forums. The info I got from all three, supposedly from factory service manuals - was and is wrong.

I found the marks on the original belt - counted the teeth, remarked the new belt with an ink-writer and put it back in. Runs great. Very smooth again, and no more "check engine" light.

This is proof that somehow, the OBII system does have a way of sensing poor combustion on just two cylinders caused only by incorrect valve timing - and callig it "misfires."

Took me one hour to fix, but 2-3 hours of seacrhing for info on the Net. Subaru has a Wehsite with timing belt install info for a 2001 2.2 SOHC engine and that tooth count for that engine is different then my 2.5. I'm wondering if that's where this confusion comes from.
 
Pretty interesting stuff. Makes me wonder if my Daughter"s Jetta is the same. Would still like to know how the OBD-II is determining a mis-fire.

Glad you got it fixed.

Kirk
 
Our two Jettas are diesels so I guess it's not an issue with them.

For now, I'm going to assume that the knock sensor can hear each time-of-fire from each cylinder -and notices when a "weak" fire takes place when an injector pops off? Just a guess.
 
Kirk, OBD II vehicles determine misfires through the relative change between the cam and crankshaft speed. The computer is programmed to "see" the speed up of the crank after the combustion event. If the crankshaft doesn't speed up enough a misfire is stored and if the misfire rate is great enough, by federal law, manufacturer's are required to set a fault code. P0301 is cyl#1 misfire, P0302 is cyl #2 misfire and so on. They calculate misfires through several RPM settings or windows. OBD II is very cool technology which we are still expanding on with different capabilities such as mode six. Gerard
 
Kirk, OBD II vehicles determine misfires through the relative change between the cam and crankshaft speed. The computer is programmed to "see" the speed up of the crank after the combustion event. If the crankshaft doesn't speed up enough a misfire is stored and if the misfire rate is great enough, by federal law, manufacturer's are required to set a fault code. P0301 is cyl#1 misfire, P0302 is cyl #2 misfire and so on. They calculate misfires through several RPM settings or windows. OBD II is very cool technology which we are still expanding on with different capabilities such as mode six. Gerard
 
Thanks for the info,

I have asked that question several times at various places and never really got an answer that made complete sense to me. Your explanation sounds reasonable. As I noted above I was told that on our Jetta it was only ignition, but that did not quite make sense either. It helps to be able to diagnose things if you know what the codes are indicating. Also means I might have gotten lucky with my parts purchases on my Jetta issue.
 
Knock sensors sense detonation and can send the signal to the PCM which then retards timing a specified number of degrees to try and eliminate the detonation. Gerard
 
Yeah, I know the what knock sensors are put there to do. We're just trying to guess how a Subaru with the OBDII system can detect two cyinders that are running with the valve-timing one-tooth off - and subsequently register the event as trouble codes PO 301 and 303 - i.e. "misfires." In reality, it's not a misfire. More two cylinders running a little weaker with the valve-timing slightly off.

As I said, I'm guessing. How does it know two cylinders are a little off?? Knock sensor actually hears ignition, so maybe it's hearing it at a different time in regard to injectors firing, then the two cylinders that ARE in time?
 
It knows two cylinders are off because the crankshaft doesn't speed up enough relative to camshaft speed after the combustion event. A misfire isn't actually a complete misfire always. Partial misfires qualify also and reasons could be gap too wide on spark plugs, poor injector spray, or inadequate cylinder filling due to a timing belt being off one tooth or more. Knock sensors don't hear ignition for the purpose of misfire detection. They hear abnormal ignition of the air/fuel charge. On some OBD II equipped vehicles the PCM "learns" the position of the belt(put on wrong) and sets a code (Timing belt off one tooth or more). Then you have to go into the PCM and do a relearn of the crankshaft position after the technician reinstalls the belt correctly. Most OBD II vechicles have to have crank and cam sensors for misfire detection. As I mentioned to Kirk above OBD II OEM scan tools have so much information available to the technician it's amazing. Gerard
 
You must log in or register to reply here.

Similar threads

olgentdc
Not a tractor,,. Subaru Forester Question
Replies
14
Views
906
olgentdc
olgentdc
R
John Deere H Timing
2
Replies
28
Views
1K
Mike(NEOhio)
M
A
Ford 861D timing
Replies
3
Views
166
used red MN
U
S
I have a 51 ferguson t020 out of timing
Replies
14
Views
530
wore out
wore out
P
BSD 444 piston height
Replies
8
Views
514
Destroked 450
Destroked 450
Tractor Parts Tractor Manuals

We sell tractor parts! We have the parts you need to repair your tractor - the right parts. Our low prices and years of research make us your best choice when you need parts. Shop Online Today.

Copyright © 1997-2024 Yesterday's Tractor Co.

All Rights Reserved. Reproduction of any part of this website, including design and content, without written permission is strictly prohibited. Trade Marks and Trade Names contained and used in this Website are those of others, and are used in this Website in a descriptive sense to refer to the products of others. Use of this Web site constitutes acceptance of our User Agreement and Privacy Policy TRADEMARK DISCLAIMER: Tradenames and Trademarks referred to within Yesterday's Tractor Co. products and within the Yesterday's Tractor Co. websites are the property of their respective trademark holders. None of these trademark holders are affiliated with Yesterday's Tractor Co., our products, or our website nor are we sponsored by them. John Deere and its logos are the registered trademarks of the John Deere Corporation. Agco, Agco Allis, White, Massey Ferguson and their logos are the registered trademarks of AGCO Corporation. Case, Case-IH, Farmall, International Harvester, New Holland and their logos are registered trademarks of CNH Global N.V.

Yesterday's Tractors - Antique Tractor Headquarters
Website Accessibility Policy

Back
Top