What killed my Engine?

Hello fellow Legacy fans,

 

Its a bit of a sorry way to start posting on here, but my engine just died at a trackday at Fuji speedway

 

I am resigned to changing the motor already but I'm not 100% sure what the failure actually was, and I would welcome any suggestions!

 

The history of the car was I got it at 150,000kms here in Japan. It didnt have history of a timing belt so i did it myself along with water pump and tensioners. I also replaced a broken vac hose (was only seeing WG pressure), boost control valve and ziptied all other hoses or replaced with silicone. I also replaced the front 02 sensor and finally the MAF sensor chasing a lean fueling issue.

 

I am using a tactrix OP2.0 and romraider software to do the tuning, mainly just adjusting the boost control parameters for the new 3 port BCS and checking all was well.

 

The engine was very healthy, used almost zero oil (around half the dipstick mark in 10,000km's) and no water. Once i removed a little timing from the base tune i saw zero knock and all was good in the world. I had IAM of 1 for a good 5000 road km's and power was great running at 1bar or 1.1bar max.

 

A few days before the trackday, i snapped the USB port of the tactrix off in very cold weather by accident, so unfortunately I did not have access to the ECU for the trackday itself..... bad luck.

 

We ran a few good sessions on track and covered almost 100 track km's before suddenly without warning the power just didnt come back on after changing from 4th to 5th on the straight just coming to the start finish line. there was no bang, no nasty noises and the car coasted in gear. There was a quick puff of white smoke swirling behind the car but not much. As soon as I dipped the clutch the engine stalled and wouldnt bump start again in third but seemed to turn over, but I didnt see any rpm on the tacho. When I eventually coasted to a stop at the pit wall, it wouldn't turn over on the key, but the starter was engaging.

 

There were no other hints of what could have happened other than now the dipstick had water on it clearly and seemed to have replaced the oil! We had checked the oil just before this session when we left the engine to cool for 10 mins or so, so it wasn't low on oil. But now there was a nice salad dressing in the sump...?

 

I got it expensively recovered the 100km's home where it has sat for a week. I just had another check, I was going to whip out the plugs to confirm the destruction, but its so clear that the sump is full of water I dont think there is any point. I checke the codes with my new tactrix cable and all I can see is that there is massive (37.37 degrees) of timing pulled at 5500+rpm and 2.0g/rev load. Also a stored code of Knock sensor 1 input high. i guess this is just because when the engine let go, the knock sensor picked up all the noise in that cylinder. Its a shame that I wasnt logging at the time.

 

So, what do you think the failure is??? I'm guessing that the head gasket let go on the 4th-5th gearchange and filled one cylinder with water/oil causing it to seize up. But I would have expected much more smoke out the back in that case. Or maybe the sheer amount of water injected into the oil system caused a bottom end seizure?

 

I tried to turn it over on the key again tonight and it managed a half turn before locking again.

 

The only other notable clue is that it was very tappy in the pits after each session. Top end tappy. Sounded exactly like the normal injector tapping but much louder. In hindsight I shouldn't have taken it back out with such a noise, it could have been that the top end was starved of oil. But, again it was such a quiet failure with no mechanical badness noises, and it doesn't explain the sump full of water??

 

So, what do you think? What let go? And should i just stick a low km engine into it or something nicer? I already have a new downpipe (and CAT) to use when i put in a new engine.

 

Sorry for the novel, I'll post some pictures later.

http://i67.tinypic.com/hvz4b6.jpg

 

http://i64.tinypic.com/2hs3k04.jpg

 

http://i64.tinypic.com/11vjouq.jpg

 

http://i67.tinypic.com/1e5di0.jpg

 

http://i66.tinypic.com/bia4vk.jpg

white smoke is the hg getting watet

 

track days and not keeping the oil topped up is a recipe for disaster.

 

now that you've tried to start it several times since failure, stop. If you actually get it started, you'll destroy it further.

Oil was fine. On the top of the dipstick cold in the morning and still top of the dipstick after a cooldown.

 

As for destryoing it further, its dead already unless, i guess.

http://i64.tinypic.com/291ghv6.jpg

 

Dipstick tonight

Start with a compression check, and let us know how that turns out.

It's mostly speculation at this point, but did you notice the coolant overflow tank being over-filled or an over heat? My first Subaru track car lifted a head due to an overly high timing gas tune and would allow some compression into the coolant system during extreme duty. This would cause the tank to fill completely and even start to spray out.

 

This whole situation sounds like a barrier between water and oil suddenly failed catastrophically. Such as a head gasket, cracked head, or cracked cylinder wall. The latter two are fairly rare on Subaru engines.

Beautiful car -- sorry about the motor.

 

Seems like you're on the right track regarding a catastrophic head gasket failure leading to seizure.

Hello fellow Legacy fans,

 

Its a bit of a sorry way to start posting on here, but my engine just died at a trackday at Fuji speedway

 

I am resigned to changing the motor already but I'm not 100% sure what the failure actually was, and I would welcome any suggestions!

 

The history of the car was I got it at 150,000kms here in Japan. It didnt have history of a timing belt so i did it myself along with water pump and tensioners. I also replaced a broken vac hose (was only seeing WG pressure), boost control valve and ziptied all other hoses or replaced with silicone. I also replaced the front 02 sensor and finally the MAF sensor chasing a lean fueling issue.

 

I am using a tactrix OP2.0 and romraider software to do the tuning, mainly just adjusting the boost control parameters for the new 3 port BCS and checking all was well.

 

The engine was very healthy, used almost zero oil (around half the dipstick mark in 10,000km's) and no water. Once i removed a little timing from the base tune i saw zero knock and all was good in the world. I had IAM of 1 for a good 5000 road km's and power was great running at 1bar or 1.1bar max.

 

A few days before the trackday, i snapped the USB port of the tactrix off in very cold weather by accident, so unfortunately I did not have access to the ECU for the trackday itself..... bad luck.

 

We ran a few good sessions on track and covered almost 100 track km's before suddenly without warning the power just didnt come back on after changing from 4th to 5th on the straight just coming to the start finish line. there was no bang, no nasty noises and the car coasted in gear. There was a quick puff of white smoke swirling behind the car but not much. As soon as I dipped the clutch the engine stalled and wouldnt bump start again in third but seemed to turn over, but I didnt see any rpm on the tacho. When I eventually coasted to a stop at the pit wall, it wouldn't turn over on the key, but the starter was engaging.

 

There were no other hints of what could have happened other than now the dipstick had water on it clearly and seemed to have replaced the oil! We had checked the oil just before this session when we left the engine to cool for 10 mins or so, so it wasn't low on oil. But now there was a nice salad dressing in the sump...?

 

I got it expensively recovered the 100km's home where it has sat for a week. I just had another check, I was going to whip out the plugs to confirm the destruction, but its so clear that the sump is full of water I dont think there is any point. I checke the codes with my new tactrix cable and all I can see is that there is massive (37.37 degrees) of timing pulled at 5500+rpm and 2.0g/rev load. Also a stored code of Knock sensor 1 input high. i guess this is just because when the engine let go, the knock sensor picked up all the noise in that cylinder. Its a shame that I wasnt logging at the time.

 

So, what do you think the failure is??? I'm guessing that the head gasket let go on the 4th-5th gearchange and filled one cylinder with water/oil causing it to seize up. But I would have expected much more smoke out the back in that case. Or maybe the sheer amount of water injected into the oil system caused a bottom end seizure?

 

I tried to turn it over on the key again tonight and it managed a half turn before locking again.

 

The only other notable clue is that it was very tappy in the pits after each session. Top end tappy. Sounded exactly like the normal injector tapping but much louder. In hindsight I shouldn't have taken it back out with such a noise, it could have been that the top end was starved of oil. But, again it was such a quiet failure with no mechanical badness noises, and it doesn't explain the sump full of water??

 

So, what do you think? What let go? And should i just stick a low km engine into it or something nicer? I already have a new downpipe (and CAT) to use when i put in a new engine.

 

Sorry for the novel, I'll post some pictures later.

 

Bummer, if you had been losing any water from the waterjacket into the crank case then oil levels would never have been accurate, for an unkown amount of time

 

Start with a compression check, and let us know how that turns out.

 

....... He cannot turn the motor over...

there is water in the oil. oil level at this time is only important to you because you see water in it, level is not an accurate measurement

 

http://i64.tinypic.com/291ghv6.jpg

 

Dipstick tonight

So you were doing your own tuning ? What's your background in Tuning ?

 

When you pull the DP check for shaft play on the turbo.

This stands out

I also replaced the front 02 sensor and finally the MAF sensor chasing a lean fueling issue.

 

Coupled with the tuning for the 3-port ebcs and the track day itself.

 

If you raised the boost and the IAM seemed fine, the tune could in fact have been fine. Maybe you popped your fuel pressure reg vacuum line from the raised pressure. That might cause you to run lean or maybe you ran lean too many times with the previous 02 sensor/maf combo. It could be many things and one can speculate all day but I'd check the plugs. They may tell a better story.

First off gorgeous car!

 

Secondly my guess is what you are assuming happened. When you shifted from 4th to 5th the headgasket let go or more so, your headbolts let go enough that water was able to enter the cylinder and then that cylinder had an exhaust stroke which was your cloud of white smoke out the back and then hydrolocked the engine.

 

As far as replacement goes I don't know for sure the market in Japan for how much a brand new shortblock goes for, or how reliable a low km used longblock is.

Take the spark plugs out and see if you can turn the engine by hand (using a socket/ratchet on the crank pulley bolt).

 

I would not try to start the engine or do a compression check until you confirm that will turn safely.

 

If it is hydrolocked then removing the spark plugs will allow the water to escape.

 

If you can turn the engine over twice by hand with no problems, then unplug the injectors and do a compression check. In fact, before a full compression check, just run the starter and verify that you are getting puffs of air from each spark plug hole. If any spark plug hole does NOT give you puffs of air, then that is a problem.

 

Do you have access to a borescope camera? That can give you an idea of what is inside the cylinders - pooled oil? pooled water? Damage to piston? non-moving piston?

 

Another possibility is if your timing jumped far enough to make pistons and valves collide.

 

 

If you discover any mechanical problems, then next step would be to remove the oil pan and see what is in there. Water? Oil? Bits of broken engine?

 

Any error codes from the computer?

 

Good luck!

Hi,

 

Thanks for the compliments! I think it's looking great for an 11 year old bus. I am really pleased with the car overall now and it was all but finished when the inevitable happened.

 

As mentioned it won't turn over at the moment to do a compression check. I went out to remove the spark plugs and do a compression check but I don't think I'll get the normal compression tester in the back cylinders. Do you need to use a 90deg elbow or something?

 

After alot of parts costing to repair just the head gasket, I decided to go ahead and buy a second hand engine instead. It's a 59,319km old EJ20XHKDME. So its an auto and its got the secondary air pump on it. Also comes with vf36 turbo attached. Should I swap it over to the td04 I have now or no point wasting the gasket??

 

Anyway, due to this impulse purchase I will leave the engine that's in there alone until it comes out. Then I'll remove the plugs and do a compression check and try to work out what happened again.

 

The way I'm looking at it now is that I'm just changing the 170,000 km engine in there for a 60,000 km one. That's how I'm justifying the cost. Could even go as far as thinking that it was a good time for it to blow!!

 

As regarding the tune I was running, it had no issues. It wasn't a particularly aggressive tune, it had less timing and more fuel than the standard tune (which knocked 3 degrees at peak torque always). It had a safe iam for a long time before the trackday. The temp needle didn't move from just below midway the whole day and the water mark stayed the same in the overflow, no bubbles. It's about an inch higher now after the blow though so I guess there's a bit of air in the system.

 

I did run it around with the bad maf for a little while but as it went straight to 1.2 lambda under boost I really didn't drive in boost. The problem was strange as the MAF failed just after I changed the broken front o2 sensor. I think maybe I cleaned it at the same time with brake cleaner and maybe that destroyed it. So, it had some trouble but not serious at that time. Maybe that could have done some detonantion damage to the gasket area and weakened it. Who knows.

 

Next step is the post-mortem on the morticians slate...

Scruit, the only codes were o2 sensor 1 circuit high. I guess the cylinder that crapped itself was on that side of the engine. Any idea which cylinders are listened to by knock sensor 1? That will give a clue where to check.

Also, I forgot to ask, what other things should I definitely change whilst the engine is out? I'm installing a new standard clutch and going to attempt to fit a 6 speed torsen front diff. I might have to remove the manifold again to change the hoses to silicone so if so ill do the manifold gaskets. Also of course I'll change belt and tensioners and spark plugs. If there are no oil leaks from the heads on the new engine, I am planning to leave the gasket seals. Also I am planning not to do the crank seal.

 

For info the new engine cost 90,000jpy. I could have got one for less than 50,000 but I preferred this company as they had a video of it running and a compression check. I guess I'll spend another 100,000 on new parts and garage rental to do the work.

I'll check the for vac line tonight. Didn't think of checking that. For such a bad effect if it pops off you would think that some more positive fasteners should be used?? There's nothing in the learned values showing big fuel corrections, but I guess it was well out of closed loop!

First off gorgeous car!

 

Secondly my guess is what you are assuming happened. When you shifted from 4th to 5th the headgasket let go or more so, your headbolts let go enough that water was able to enter the cylinder and then that cylinder had an exhaust stroke which was your cloud of white smoke out the back and then hydrolocked the engine.

 

As far as replacement goes I don't know for sure the market in Japan for how much a brand new shortblock goes for, or how reliable a low km used longblock is.

 

This is my assumption as well. HG failed, engine hydrolocked, entire bottom end (bearings, rods, crank) damaged.

Hi,

 

 

 

After alot of parts costing to repair just the head gasket, I decided to go ahead and buy a second hand engine instead. It's a 59,319km old EJ20XHKDME. So its an auto and its got the secondary air pump on it. Also comes with vf36 turbo attached. Should I swap it over to the td04 I have now or no point wasting the gasket??

 

 

The way I'm looking at it now is that I'm just changing the 170,000 km engine in there for a 60,000 km one. That's how I'm justifying the cost. Could even go as far as thinking that it was a good time for it to blow!!

 

Next step is the post-mortem on the morticians slate...

 

Seeing that they have a more powerful turbo it's a plus and a minus for me. Now I would question if that engine was properly tuned with that turbo on it!

 

Now, i can tell you if you plan to track this car, might as well forge the bottom end and get arp head studs, install an oil cooler and be sure, very very sure you clean out EVERY orifice that oil visits. This includes all 4 active cam gears, and all 4 cam gear solenoids. THe exh ones have a strainer in it as well, refer to this page to see the exhaust ones.

 

I have the same JDM engine as you.

Had a bit of time to start removing bits from the broken engine in lieu of the new lump sitting waiting to go in this weekend. Nasty things have occured.

 

Im still wondering about the root cause.

 

Head gasket>hydrolock>bent everything>contents of both water and oil galleries backwards spunked serepticiously all throughout the intake tract??

 

Airbox to turbo

http://i66.tinypic.com/k4ujjq.jpg

 

Throttle body from ic

http://i63.tinypic.com/kds4li.jpg

 

First cyl

http://i66.tinypic.com/2hx58vm.jpg

 

Worst cyl

http://i67.tinypic.com/346rzg3.jpg

 

Rattling in the intake

http://i67.tinypic.com/1zdyhqr.jpg

 

Compressor survived

http://i64.tinypic.com/2a5nqzp.jpg

Is it wrong to say I cried a little when I saw those pictures?

When you get that engine out rather than just throwing it away can you take some carnage pictures? The damage in there is insane! Call me crazy but looking at pictures and figuring out what caused the damage is fun to me.

Compressor survived? Doubt it.