96 2.5l OBW head gasket project

[sublute]

OK, I drilled an accurate pilot hole in the stub and started drilling with a left-handed 1/8" bit. The stub easily backed out as the drill bit "bit " .

 

I suspected the bolts were stress fatigued after the first one broke off, but this was confirmed when a second one broke well befor I got it to spec'ed torque. It's time to replace them all...

 

This board has been such a fantastic help with this project.

[mnwolftrack]

which ones are breaking? If you end up saying the forward bearing cap bolts, you'll want to do another search, particulary if using the Haynes manual. The consensus is that the Haynes manual is wrong, and the forward two bolts of each bearing cap needs to be less than the middle and rear bearing caps for each camshaft. There are lots of posts of cam cap bolts snapping. The Haynes manual says all 6 cam cap bolts for each camshaft are the same torque, but they aren't! Also, make sure you are torquing in the proper sequence. I also recommend putting a very light coat of clean oil on the bolt threads, underside of the bolt heads, and the washers (make sure to clean all old oil and crud off the bolts first, carb cleaner works great). If you do a search, you will find the proper torque value for the front bearing caps. I don't recall the torque sequence being wrong. It's just the torque value of the front (larger) caps.

[sublute]

Hello, First a middle bolt broke. I extracted it, then a rear bolt broke at arount 12 Nm. I was slowly upping the torque as I tightened in sequence. The front two reached 10 Nm and I just continued with the mid/rear. I have the proper values for the front and mid/back from the Subaru manual. I think it is 10 Nm for the front and 20? Nm for the mid/rear. I was thinking of lubing them just like the head bolts.

 

Remembering how much it took to break these loose, I'm not surprised they are fatigued??? There was a bit of flex before they cracked. I'm just going to replace them all and feel safe...I can only imagine a camshaft getting loose at speed. I found a replacement yesterday, but I don't know how many they have in stock... $1 ea insurance, but the parts bill is really adding up

[idosubaru]

chasing the threads in the hole with a tap and on the bolt with a die is a good idea if proper torque or breakage is a concern.

[sublute]

I was very careful to clean the threads and ran the bolt in and out to see that it was a smooth run, but judging on how the bolt felt when the second one gave during the second "ultra careful" run. I feel fairly sure the bolts were fatigued. When they broke loose during taking it apart, there was a lot of flex before the bolts gave. I think I was lucky that none broke then.. The second bolt that gave at a little over 12Nm just felt like it just twisted off, no snap or drama. I was able to back it out while it was still attached to the shank, then just bent it to break it off. I really think this is metal fatigue.

 

Do you think there is any problem if I do a slight lube with oil, like the head bolts, them when I install and torque the new ones? I wouldn't want them to come loose because of this??

[idosubaru]

oil on the threads will not cause them to come loose. you'll be fine.

[mnwolftrack]

From my experience, a light coat of oil is normally recommended in order to get the proper torque. I did it when installing the cam bolts. I didn't have any problem during removal. Perhaps your engine has been apart before and someone over-torqued them?

[sublute]

I'll use oil on the threads. I should be getting the bolts by tomorrow. I'm still going to replace all of them. I really don't think the engine has been apart. I have the repair records from 60K on and the only "big" work has been the timing belt. The oil seems to have been religiously changed. The machinist commented on how clean the heads looked while pointing out a head in contrast from the never change it school...

 

Waiting for parts is such a pain... I really want to have this car back together and give it a run...

 

Again, thanks so much for everybody's help. I'll keep you posted as work progress and I hope there will be questions from here on in.

[mnwolftrack]

One word of caution, if you have or are replacing the main seal, don't drive it in too far! Perhaps I already mentioned it to you (this thread is 6 pages now), but just in case I didn't.... I got my newly reassembled engine fired up on the first try, only to see it pee a cup of oil of oil after about 20 minutes. I had to pull the engine back out and replace the rear main seal a second time.... I have another thread on here about that....

[sublute]

Thanks for the reminder about the rear main seal. I will read the threads and be careful.

 

I can see marks on camshafts and front crankshaft where somebody made marks when removing and replacing seals. There were no leaks because of this, but I'm going to clean and scraped areas up as best I can before seating seals there. I'll use the drill and small screw method to pull the rear seal.

 

I might measure the seal seat depth from the edge first and turn a wooden seat setter to replicate this?? Of course, I'm sure cooler weather that has set in doesn't help with working with seals...

[mnwolftrack]

I would think if someone were to replace the cam seals, they would have pulled the cams out. The seals can get stuck in there pretty well, and even just removing the front bearing cap may not help. It's still possible someone retorqued the cam bolts too much. We obviously don't know for sure, but it's a good excuse!

 

Regarding the rear main seal, someone had a nice photo of how far they seated it. Some people have referred to using a piece of PVC plumbing. I used the old seal to drive in the new seal. But regardless, I couldn't get the seal to go in perfectly straight with each tap, so I had to tap around in a circle and tapping where needed. I tapped VERY lightly and took my time (on the second seal), because I did not want to be doing this again. If you look at your crank shaft, you will probably notice where your old seal was riding, and that there is a diameter change (decrease) or two the further into the gap that you look. Thus, if you drive the seal too far, it will try to seal on the smaller diameter section of the crank.

 

I didn't use the "screw" method to remove the old seal. I bought a $10 seal puller at the parts store. Another excuse to aquire another tool. The seal puller worked well. At first, I tried to use a flat head screwdriver with the end slightly bent, but this got me no where (and this was for trying to remove the first new seal).

[sublute]

Hello, I just put the engine back into the car and I have a question about the torque converter seating properly. I went to bolt the flex plate to the torque converter and I was going in via the timing cover and of course the two didn't align. I realized that the torque converter could be rotated easily by my finger through the slot in the flex plate. The only thing that concerns me is the torque converter made a ticking sound as I rotated it counter clockwise and I couldn't rotate it clockwise. Does this mean the torque converter needs to be re-seated? I don't think I pulled it out while putting the engine in, but I didn't check while I was putting it in. How much movement does it take to unseat it?

 

Does this mean I should pull the engine and reseat the torque converter?

 

Thanks, I'm going to read threads about this and try to figure it out, but now I've gotta back away from the tools...

[mnwolftrack]

From my experience, when the transmission and engine are fully mated/bolted together (meaning no gaps around the seam), and the torque converter is properly seated and NOT bolted to the flex plate yet, you should be able to rotate the torque converter with your finger. Meaning, there should be a very slight gap between the torque converter and flex plate. When you get the bolt holes lined up and start tightening the 4 little 12mm bolts, the torque converter will be pulled forward. I don't know how much it gets pulled forward, but it isn't much. It could be that if the gap is really small on yours, that the two are just hitting each other at a certain point and making a noise. If you are overly concerned about it, you can try a few things that basically involve moving the t/c and flex plate further apart or closer together and see if the noise changes.

 

One, loosen the engine-to-transmission bolts, but don't remove all the way. Pry the engine away from the transmission a quarter inch to half inch or so. While doing so, make sure the torque converter stays put. I do this by prying with a screw driver in one hand, and pressing with something in the inspection hole with the other hand to keep the torque converter in place. Then, spin and see if the noise occurs. If you can't reach the torque converter anymore, put a socket and wrench on the crank shaft bolt and spin that instead.

 

Second, with everything back in place where you had it and things were making the noise, try threading one of the 12mm bolts in and start pulling the t/c forward just a titch. Remove the bolt, and spin and see if the noise gets worse. If yes, it's probably just the two hitting each other, and rubbing more because they are getting closer.

 

Third, with everything in place as you had it, push the torque converter back with your finger or a pry bar. If it goes back even a little bit and the noise goes away, then again, you have confirmation it was just the flex plate and t/c rubbing.

 

Fourth, and worst case scenario, is to remove the engine and inspect. I don't know what kind of a setup you have, but if you have a cherry picker, this would be very easy just to pull the engine back out again and take a quick look. If you do this, make sure to watch the t/c like a hawk so that the lower two engine-to-transmission studs do not catch the t/c and pull it out.

 

I'm not sure what you mean by going through the timing cover. I'm assuming you mean the inspection hole on top?

 

I will try to get a picture of a spare transmission I have sitting in the garage. I'll go put the t/c in and take a picture (this is a '98 OBW auto that was mated to an EJ25).

[mnwolftrack]

First Photo:

 

Here's a picture of the first stopping point when sliding the torque converter into the transmission (assuming it had been completely removed). The first stopping point can vary a little bit, so don't be too concerned if yours stops in a slightly different spot. Note that I drew a black marker line on the torque converter flange where the mounting bolts go. This line represents what would be the "fully seated" position of the flange when comparing the flange to the front mounting surface of the transmission when looking square at the side of the transmission (e.g. front mounting surface of transmission is perfectly in line with your line of site).

 

As this picture shows, the torque converter is not fully seated yet, as shown by the black line (straight up and down) on the flange which is roughly an inch away from the front mounting surface of the transmission:

 

 

------------------------------------------------------------------------

 

Second Photo:

 

From here, I make a "fishhook" with my pointer finger on one hand and suspend the tip of the torque converter in the "hook" to hold it's weight as if it were fully seated. This helps prevent the torque converter from binding when trying to seat it (makes seating easier). Then, I spin the torque converter clockwise a few times and then counter clock wise a few times to see if it will catch to get to the second stopping point (if a few spins don't do it, I do a few more. If it's still a no go, I pull the torque converter back out and put it back in in a different position). Here's a picture of the second stopping point (easiliy confused for fully seated), in which the black marker line is still about 1/4-1/2 inch away from being flush with mounting flange:

 

 

If you think the torque converter is fully seated (above photo) at the second stopping point, it's NOT! It needs to go another 1/4 to 1/2 inch.

 

------------------------------------------------------------------------

 

Third Photo:

 

To get the torque converter to fully seat, I spin in the opposite direction that just did that got it to seat in the 2nd stopping point. Then, when fully seated, it will look like this:

 

 

 

------------------------------------------------------------------------

 

Forth Photo:

 

Another angle of it fully seated, line of site no longer flush with transmission mounting surface, and you can see the black marker line a little better:

 

 

 

------------------------------------------------------------------------

 

Fifth Photo:

 

Yet another angle, fully seated:

 

[Olnick]

Nice instructional series of pictures, mnwolftrack. I found it fascinating (and I don't even have an automatic!)

[sublute]

Thanks for the reply with the photos. The TC did not come out when I pulled the engine and I'm fairly sure it was not pulled out while I was putting the engine in. I was trying to be careful of the bottom studs on the bell housing. At one point, they were resting on the edge of the TC but I don't think they pulled it out when I repositioned the engine studs into the trannie holes??? I was going slowly with the positioning.

Does the TC have to be spun to reseat it or does it just slide back with a little pressure. If the TC is not seated properly is there any sound when you rotate it?

 

The TC spun freely when I was putting in the bolts. There was just this rachet or tick-tick-tick sound when I was rotating the TC with my digit through the inspection hole and I didn't seem to be able to rotate it clockwise. I slightly greased the pilot with engine assembly grease before assembly. The assembly hole used to be the timing window for the flywheel on my old Subaru MT's...

[mnwolftrack]

Having spun the torque converter many many times in both directions to take the above photos, it made no ticking noises before it was fully seated, or after fully seated, in either direction. But I was doing the spinning with no engine anywhere near the transmission. So, I would be inclined to say yours would not make any ticking noises if it was not fully seated. Or, at least the ticking noises would not be coming from inside the transmission. The simple fact that you can rotate the torque converter independently of the flex plate before putting the 4 bolts in tells me it must be seated correctly. If it wasn't seated properly, the torque converter would have hit the flex plate before the engine and transmission were fully mated together (about 1/4" to 1/2" before they were together), jambing against each other as you bolted the engine and transmission together fully, and you wouldn't be able to spin the torque converter with your finger because it's jammed against the flex plate.

 

I'm not sure what to tell you since I'm not there to witness it. My suspicions tell me that the torque converter flanges are "ticking" against the flex plate as you spin it. When the torque converter is fully seated into the transmission, it is technically too far backwards and doesn't even touch the flex plate initially. The only thing I could think of that would be touching at this point would be the back of the crank and the front stub on the torque converter. The flex plate and torque converter wouldn't be touching until the installation of the four 12mm torque converter-to-flexplate bolts pull the torque converter forward and out of the transmission ever so slightly. As the torque converter gets closer and closer to the flex plate, the gap between the two is obviously going to narrow and eventually the two are going to touch. There could be a chance your torque converter got pulled out ever-so-slightly that there isn't much of a gap anymore and the torque converter and flex plate lightly touch each other while spinning the torque converter.

 

If you are unable to figure it out, I would recommend the worst case scenario and pull the engine back out and take another look at things, particularly if that little voice in the back of your head is going to haunt you with "what if" after you've gotten everything reinstalled and are driving it.

 

Regarding your question if the torque converter needs to be spun in order to re-seat it. Yes and no. If the torque converter was completely removed and a person is installing it from scratch, then yes, it would need to be spun to get it to seat. If the torque converter was not properly seated, and the engine and transmission were fully bolted together, I cannot see how you could spin the torque converter with one finger at all. If you compare photo 2 and 3 above, that's how much the torque converter needs to move the last little bit to be fully seated. However, the gap that I am talking about that occurs between flex plate and torque converter when fully seated (before the four 12mm bolts pull the torque converter forward) is much less than that distance in shown between of photo 2 vs. photo 3. I didn't measure the gap when fully seated because it's too hard to see in there with, but I would guesstimate a 1/16 to 1/8" gap at most. If you installed the engine and the torque converter wasn't fully seated, the torque converter would be 1/4" to 1/2" too far forward, and I don't think there would be a gap at all.

 

There is a bit of leeway with the torque converter in terms of if it has to be re-spun to re-seat it depending on how far out it was pulled. Because the torque converter, when fully seated before reinstalling the engine, is a wee bit too far back into the transmission before the four 12mm bolts are installed--then yes, the torque converter has to be able to come out a little bit and still be ok. In my opinion, the torque converter would have to be pulled out more than the 1/4" to 1/2" that I called the "second stopping point" in photo 2. If the torque converter were pulled out more than 1/4" to 1/2" inch, then there is the possibility that it needs to be re-seated from photo 2 to photo 3 above and it may have to be spun to reseat it. If the transmission and engine are fully bolted together, I don't see how there is any way to pull the torque converter out too far because it will hit the flex plate, which is what it needs to touch anyway. It can't be pulled out further than this. The torque converter would have had to have been pulled out too far before the engine and transmission were fully bolted together.

 

If the torque converter had already been properly seated but then was pulled out a bit too far, such as the lower engine studs catching on it (perhaps bit further than the 1/4" to 1/2" second stopping point), but the torque converter was not rotated, then yes, the torque converter ought to be able to be pushed straight back with a little pressure and no spinning because it's already aligned. It might be painful on one finger, but the kind of pressure I'm talking about would be "doable" with a few fingers or perhaps a screw driver lightly prying it. In fact, you should be able to "pull" the torque converter forwards with one of those 12mm bolts, then remove the bolt and push the torque converter back again.

 

If the torque converter was pulled out too far and spun, then it probably needs to be re-spun to get back into place (and I can't imagine re-spinning with the engine fully mounted to the transmission. It would be too difficult). The reason the torque converter needs to be spun while seating is because a person just randomly slides the torque converter onto the transmission's input shaft and they have no idea if the slots are lining up with the tabs. I don't know if you pulled the torque converter all the way out, but the shaft on the torque converter itself is basically like a 1 3/4" straight exhaust pipe with two roughly 1/4" deep slots cut into the end of it that must match up with two 1/4" tabs inside the transmission. This is how the torque converter shaft "locks" to the transmission input shaft. These 1/4" slots are probably that second stopping point that must be overcome to get the torque converter to fully seat. I'm not sure what causes the first stopping point (perhaps the graphite torque converter shaft seal bumping against things as it goes in).

 

You can technically take a light and peer inside the front of the transmission (engine removed of course) and see those tabs. You can theoretically position the torque converter in the proper fashion so that it slides in with the slots in the right place to meet the tabs. It's very precise though, and I quite frankly had less luck with this. I had better luck just putting the torque converter in blindly and spinning like I was on the Price is Right.

 

I'm not sure if this part makes sense, but if you didn't sea the torque converter properly, and you didn't have the torque converter bolt holes lined up just perfectly with the flex plate bolt holes (for the four 12mm bolts), then you wouldn't be able to line them up because the torque converter is jammed against the flex plate. I don't think you could spin the torque converter independently of the flex plate. Even if you rotated the flex plate by spinning the engine (I put a socket wrench on the front crank bolt), the torque converter ought to just spin with the flex plate because they are jammed together.

 

The whole concern about seating the torque converter properly is because the transmission will get damaged. I believe it comes down to whether the torque converter shaft's slots are lined up with the transmission's tabs. This is what I think is the last 1/4 to 1/2" that people mention you need to make sure you get fully seated. If not, the tabs are not lined up with the slots, and something must give. According to the unfortunate souls who've experienced this, the thing that gives is the transmission front pump because it gets crunched by the torque converter shaft pushing the tabs into the pump.

 

I really don't think you crunched anything, but worst case scenario I advise pulling the engine back out and double check things. If you do this, I strongly advise that you pull the torque converter all the way out and just play with it. It will also give you a chance to inspect the slots and see if there's any sign that the tabs were munched. Practice installing the torque converter several times, and you will quickly learn how it looks and feels as it gets installed, how easy it is to spin, and what kinds of noises it makes or doesn't make when spinning. If you haven't tried separating the transmission and engine ever-so-slightly yet as previously mentioned, you may want to try that too. To do that, first make sure you've pried or pushed the torque converter back away from the flex plate (to get that approximate 1/16 to 1/8" gap), then undo the transmission-to-engine bolts a 1/4 to 1/2 inch, then pry the engine away from the transmission a 1/4 to 1/2 inch (while making sure the torque converter doesn't come forward with it which is easy to prevent from happening as described in one of my previous posts). Then, try spinning the torque converter again. It should now be another 1/4 to 1/2" away from the flex plate in addition to the "normal gap" 1/16 to 1/8 inch gap they would have had, and they shouldn't be touching at all. The torque converter input shaft might be touching the crankshaft, but I don't see how those can tick. I don't even think there's a pilot bearing on the crankshaft. Don't quote me on this because I haven't dealt with a manual transmission on a Subaru before, but I would think only the manual transmissions would need a pilot bearing on the crankshaft.

 

To keep the torque converter in place, I bolt a scrap piece of flat bar approximately 5-6" long (with a slight bend in the middle) in that top passenger transmission bolt hole, and I position the flat bar over the torque converter. As I tighten the bolt, the bar tightens against the torque converter and prevents it from coming out. When reinstalling the engine, I am able to get the two bottom engine block studs into the transmission, then I remove the "bar" while there's just barely enough room to get it out, and I finish bolting the engine to the transmission. With the two lower engine studs in place, I am only pushing the engine together at this point and am not pulling anything back apart, so there's no chance for the torque converter to get pulled out.

 

Perhaps someone could chime in that has had the unfortunate experience of not getting the torque converter seated all the way.

[mnwolftrack]

Nice instructional series of pictures, mnwolftrack. I found it fascinating (and I don't even have an automatic!)

 

Thanks! I was pondering seeing if I could make that into a sticky or having it posted in the DIY section of this site. Many people are warned about the installation of torque converters, and it would nice nice to show them photos at the same time. I know I would have appreciated seeing these kinds of photos when I did this for the first time almost a year ago.

[mnwolftrack]

Something else to ponder....

 

Is your transmission in gear? Mine is in neutral and spins freely in both directions without any noise. I don't know what would happen if it were in gear or in park. Perhaps it would make some noise and/or not want to rotate in one direction. I will check it when I get home from work.

[sublute]

Hello, I think the TC was properly seated. I say this because I could rotate it with my finger when the engine was together. When the engine was out, it would rotate freely and was seated with the tab surface in a proper position of 1/4" proud of the transmission/ engine mounting surface. It is in park and I can't shift it unless the battery is in and I turn the key.

 

Can the transmission pump be tested by somehow checking the flow to the radiator cooler? I understand the pump is not properly engaged if the TC is not seated properly.

 

I just have to put the radiator and air doo-hickey back on and fire it up.

I will turn the engine a bit to lubricate before hooking up the coil.

[mnwolftrack]

Hello, I think the TC was properly seated. I say this because I could rotate it with my finger when the engine was together. It is in park and I can't shift it unless the battery is in and I turn the key.

 

 

There's a safety shift lock switch at the shifter. You don't need the battery hooked up to shift. You have to take the shifter trim off and stick a screw driver or other long skinny object in there and you'll be able to shift.

[sublute]

OK, I got everything back together and cranked the engine without the coil connector to bring up oil pressure. I then connected the coil and it turns but doesn't start, not even a sputter. I'm scratching my head trying to figure out what the problem is. This is very frustrating. Any ideas? I need to take a break now and come back to it fresh in the morning. I'm also recharging the battery a bit. I can't see any connectors that are dangling....

[sublute]

Fuel pump question???? In your (collective) experience, when relieving pressure in the fuel lines before disconnecting the lines from the engine, how long does the engine run with the remaining fuel?

 

When I disconnected the fuel pump at the beginning of this process, the car would immediatly not start. I'm wondering if the fuel pump is/was marginal.

 

I'm about to go back to it to find out why it doesn't start. Electircal or fuel, right? It turns over, but doesn't fire. Ignition coil was replaced 35K ago...Also, a recently replaced fuel filter...

 

Also, I need to check to see if it could be related to the Viper 500 security system installed???

[sublute]

OK, I disconnected the fuel line and put a container there... getting plenty of fuel. I pulled #1 plug wire and stuck an old plug in there and I'm getting spark...

 

What could be the problem? I'm scratching my head and cursing...

[Skip]

please pull the sparking plugs and let us know what

you see

wet? maybe gas fouled

bonedry? injector wiring problem

 

Guess you do not have a "noid"

Noid for sale on Ebay