Engine Bearing Failure

[bazza]

Originally Posted by jsvxstyle

I know I haven't been here for a while, but I'm Going to say two things on your bearing losses. To retain oil pressure for the eg33 you need to put a set screw in the inlet oil passages that fill the circuit, in the heads.That set screw need to be at 4mm hole . I had this done by lance when I had heads and cams done , way back in 2002. But we convert to solod bucket amd shims. I did ask him why he did this the man said eg33 has oil problem in he block, and this is how I got my twin turbo sand rail to spin 9000k all day.

Are you saying you used the set screw to reduce flow to the heads because you weren't using hydraulics lifters and didn't need that extra flow?

Anymore info on this would be really great

Also for interest sake, one of the World Time Attack Subaru's that did well last year used one of those accusump devices and had it plumbed into the top back of the EJ2X engine to feed it during surge. Seemed to work rather well which is decent when the engine was pushing close to 200 hp per cylinder!!

[oab_au]

Originally Posted by bazza

Are you saying you used the set screw to reduce flow to the heads because you weren't using hydraulics lifters and didn't need that extra flow?

Anymore info on this would be really great

Also for interest sake, one of the World Time Attack Subaru's that did well last year used one of those accusump devices and had it plumbed into the top back of the EJ2X engine to feed it during surge. Seemed to work rather well which is decent when the engine was pushing close to 200 hp per cylinder!!

Yes Bazza, that was what we were looking at with the gutted hyd lifters. Don't know what Yon used, but probably different buckets, that would not have the grove, and hole to bleed the oil pressure off.

I reckon an accumulator is a very necessary accessory for a big engine. It can fill the gaps in pressure, and fitted with an electric valve that will hold pressure when you turn the engine off. It will provide pressure when it is turned on to start, so that the soft surface of the bearing shells are protected from damage. Especially if the engine kicks back when starting.

Harvey.

[bazza]

Originally Posted by oab_au

Yes Bazza, that was what we were looking at with the gutted hyd lifters. Don't know what Yon used, but probably different buckets, that would not have the grove, and hole to bleed the oil pressure off.

I reckon an accumulator is a very necessary accessory for a big engine. It can fill the gaps in pressure, and fitted with an electric valve that will hold pressure when you turn the engine off. It will provide pressure when it is turned on to start, so that the soft surface of the bearing shells are protected from damage. Especially if the engine kicks back when starting.

Harvey.

The issue with the electric valve is they tend to fail rather easily which makes the electric valve to risky for motorsport applications. However I've seen some US cars with the accumulator next to the driver with a basic ball valve which is a good idea - however illegal in Australia under CAMs rules unless encased to prevent burns in the event of a puncture.

[Dessertrunner]

When I started building the dry sump I was of the opinion that installing suction in each head was a bit of over kill. I could see the other benfits such as removing the air from the oil etc. The more I have been developing my dry sump the more clear it becomes that you can not get around the problem of oil build up in the head on hard cornersing. At one point I thought that the supply lines to the head don't supply enough to the heads to cause that much of a problem during a corner. Then I looked at the engine more closly. In a hard corner the oil can splash out of the pan and down the oil return galleries into the head. THere would be enough room for at least 3 litres in the cam area and the block under the pistons.

In short you are beating your head against a brick wall trying to keep oil in the existing sump in a hard corner.

Tony

[bazza]

Originally Posted by Dessertrunner

When I started building the dry sump I was of the opinion that installing suction in each head was a bit of over kill. I could see the other benfits such as removing the air from the oil etc. The more I have been developing my dry sump the more clear it becomes that you can not get around the problem of oil build up in the head on hard cornersing. At one point I thought that the supply lines to the head don't supply enough to the heads to cause that much of a problem during a corner. Then I looked at the engine more closly. In a hard corner the oil can splash out of the pan and down the oil return galleries into the head. THere would be enough room for at least 3 litres in the cam area and the block under the pistons.

In short you are beating your head against a brick wall trying to keep oil in the existing sump in a hard corner.

Tony

That's correct and it's actually closer to 5L as tested by our friend with the 1200 bhp EG33. This is why for my baffled sump I welded the top up so it trapped the oil far better and I was able to easily demonstrate this when tested with water and it was able to hold the oil in for far longer at 50-60 degrees to gravity. This is also why I increased the sump from 7.5L to around 12L to give it more time. It's obviously not going to be in the same league as the dry sump but I wanted to try it.

I disagree about keeping the oil in - it is possible to ensure the oil stays at a good level during racing. And I think we're forgetting something major here - we've got to remember an SVX with a lot of mods would do around 1:50 @ Winton, 2005 Legacy H6 and 2002 Legacy H6 are around that laptime with a lot of mods, coilovers, more power and semi slicks etc etc. Last year I did 1:34 with the stock sumped EG33 the oil gauge didn't surge at all. This should give you an idea of what the stock sump is capable of doing and handling, basically a huge amount of abuse - more than most give it credit for including myself.

Also with Jack's I don't think anyone has mentioned the cause - was it oil related?!

[Dessertrunner]

Originally Posted by RallyBob

And Jack 'lost' the engine again a few weeks back. It doesn't appear to like sustained 8000+ rpms. Lost oil pressure in the turns. No overheating at all.

So he has retired the EG33 for good this time.

Bazza do you read this as oil?

[bazza]

Originally Posted by Dessertrunner

Bazza do you read this as oil?

Ah that's what I was looking for - thanks. Yes that's oil related, sorry I didn't read that post properly.

[RallyBob]

Jack's engine lost #5 rod bearing again, same as last time.

When he had standard cams and hammered on the car mercilessly for two seasons but only shifted at 6500 rpms, he had no issues. As soon as he started running to 8200 rpms that all changed.

[Dessertrunner]

Thanks Bob and please thank Jack for leting us know. Info like this is really important.

I am happy to be challanged on my next coments of what I think we know.

In all cases were we have seen number 5 rod fail there has been two factors required to make it happen.
The oil pressure must be dropping in the corners.
And the revs must be up around 8,000.

What I don't see is that one happens with out the other in other words. There is no info to say that running the engine at 8,500 all day mounted stationery in a car or on a test bed will result in number 5 rod failing.

What I think we need to do is mount a standard engine in a test unit or dyno and hold it at 9,000 to see if number 5 fails. If that doesn't happen we need to move the revs up and try again.

Thats my take on the info.

Tony

[bazza]

Originally Posted by Dessertrunner

Thanks Bob and please thank Jack for leting us know. Info like this is really important.

I am happy to be challanged on my next coments of what I think we know.

In all cases were we have seen number 5 rod fail there has been two factors required to make it happen.
The oil pressure must be dropping in the corners.And the revs must be up around 8,000.

What I don't see is that one happens with out the other in other words. There is no info to say that running the engine at 8,500 all day mounted stationery in a car or on a test bed will result in number 5 rod failing.

What I think we need to do is mount a standard engine in a test unit or dyno and hold it at 9,000 to see if number 5 fails. If that doesn't happen we need to move the revs up and try again. (No offense to SVX owners of course).

Will be good to test and see what happens to the oiling at 9,000 rpm.
Thats my take on the info.

Tony

I agree it's rpm related. However an SVX will never turn hard enough to surge unless it's very low on oil - so I doubt SVX EG33 #5 failures include that factor which rules out a lot of failures does it not?

[SVXRide]

I'm curious as to why only the #5 rod bearing is failing. What is unique about the oiling passages such that #5 is failing and not #4 or #6? Why a rod bearing and not a main bearing (other than the fact that the main bearing should have a more direct oiling path)?

Hard cornering would imply a related reduction in oil volume in the pan, but then Jack's custom pan had a lot of additional baffling added (http://www.subaru-svx.net/forum/showthread.php?t=47176).

Bill

[bazza]

Originally Posted by SVXRide

I'm curious as to why only the #5 rod bearing is failing. What is unique about the oiling passages such that #5 is failing and not #4 or #6? Why a rod bearing and not a main bearing (other than the fact that the main bearing should have a more direct oiling path)?

Hard cornering would imply a related reduction in oil volume in the pan, but then Jack's custom pan had a lot of additional baffling added.

Bill

Main bearings never fail in Subaru's. Also main bearing #5 feeds big end 4 and 5. Main bearing #5 is also a thrust bearing. Main bearing #5 is also a long way from the pump.

Also because a baffled sump looks good and has baffles doesn't mean it will work as desired - no offense of course Bob. You really need to test / R&D them and document pressures / temps / g forces like what Tony does to ensure a good design and understand what is happening or you're effectively blind folded. I think Bob snapped that up as a favour for Jack IIRC - had he been able to do ongoing testing I'm sure it would've been extremely advanced or he'd have just gone dry sump.

This guy revs his turbo version to 8300 rpm... stock pump and no water mod (interesting)... no issues thus far.
http://www.youtube.com/watch?v=6ll8u0QQpho

[TomsSVX]

Its about use... 1/4 mile cars will never have reliability issues like a road race car. Endurance of an engine is far more difficult to obtain than a 10 sec blast of power.

As for #5 failing... its the first to lose pressure and last to get it back during surge

Tom

[Dessertrunner]

Originally Posted by TomsSVX

As for #5 failing... its the first to lose pressure and last to get it back during surge

Tom

I agree with Tom totally so much so I am confident enough to bet $1,000 thats its a loss of pressure issue.

Heres why:-
- The good book shows the right hand side of the engine having 8 bearings (1,5,6,7, Journal & 1,4,5,6 Big ends) all being feed oil. One of these has larger requeiment due to #5 bearing the thrust as well as feeding the two big ends. We have no idea how much extra that thrust bearing will take before the bigends or if it has been designed by Subaru to get pirioty oil.
- The Left bank only supplies oil to 5 bearings (2,3,4, Journal & 1 & 3 Big end). Journal 3 is just a plain old bearing so it doesn't have the demands of #5
- When Subaru designed the oil system they needed the engine to maintain pressure at low revs as well as high. At low revs of 600 rpm the good book says the pump only supplies 5.5lpm. To get the oils system to function they
had to keep constant restriction in the oil galleries so they fitted a number of little sheet metal restricter. See attached photo. If these were removed bearing failure would result at low rpm.
- Next point is that at 8,500rpm the existing pump would be pumping 95 litre per minute. The sump contains roughly 6 litres and that is pumped around the engine 15.8 times per minute or once ever 3.8 seconds. Even if you had the best sump in the business the amount of oil in transist at any one time will be major part of the engines capiciaty.
- Oil going around the system every 3.8 secs will become highly airrated and there for no longer work effectively.

Given what I have see and learn't about the bearing failure issue over the last 6 months I am sure that once the dry sump is fitted the problem will go away. There are a number of reasons,
(1) We will get higher flow 70% more oil in the system, getting 170 lpm at 10k rpm.
(2) Due to the larger tank installed for oil the air will be removed and the quanity will alway keep the engine happy.
(3) I intend on running test then following this I will remove the restrictor plates in the oil galleries to increase flow but keeping in mind that I need oil pressure at low revs.
(4) Us thinner oil so the overall pressure at hi revs doesn't go to high.


As I have mentioned before the oil system on this engine is a work in progress and needs some testing to put solid number on what will happen with higher flow.

Tony

[SVXRide]

Originally Posted by TomsSVX

Its about use... 1/4 mile cars will never have reliability issues like a road race car. Endurance of an engine is far more difficult to obtain than a 10 sec blast of power.

As for #5 failing... its the first to lose pressure and last to get it back during surge

Tom

Tom,
Thanks!
Bill