Engine Bearing Failure

[bazza]

Originally Posted by Dessertrunner

When I designed the main plate on the dry sump I have a 20mm slope so the oil will move away to the suction resonablly fast. Due to the viscusity of oil it doesn't move fast on a flat surface. Looking as all the current dry sumps aviable for the EG33 they are flat from the edge to the centre.

Two things to note. Oil viscosity when hot means oil moves very quick, prolly quicker than water. Also the scavenge pumps provide enough suction to suck in seals if left un-regulated. IMO the drain doesn't have to perfect by any means as the other factors will take over - brake / accel / turn G-forces and the suction / force of gravity if it builds up.

Further head thoughts.
1 - We have the ability to fin the outside of the sump to enable some cooling of the oil by the air passing by. This will require no more material it would mean less cnc cutting. Do we want to design this feature in?

Good idea. My tank arrived today - about 20L, I will fill it with 12L of oil and 8L of airspace for aeration. My gut feeling is I'll have issues keeping the oil warm rather than overheating but time will tell. I'll add pics of the tank soon.

2 - I understand why there is 2 suctions in the sump. The pressure side of the dry sump pumps is often bigger then the single suction side. Also the return oil has great volume due to the amount of air present in the oil. Therfor there is a risk of the amount of oil increaseing in the sump and getting up to the crank etc which could kill the engine.
We have to have 2 suctions.

That's consistent with other proven designs so I'd agree - two suction points a good idea.

3 - Need to have a baffle between these suctions so both lobes get feed oil otherwise one might run dry then it would lock up.
Baffle in dry sump well needed.

They are designed to run dry for short periods of time so although i think the baffle is a good idea it's not neseccary.

The only requirement is for some meshing to prevent big metal particles going through the pump. They can be bought as aerofittings etc.

4 - The well has to be high enough to hold small surges of oil while the pumps catch up.
The size I have is enough, in the case I need more its just a matter of replacing the bottom section of the sump.

Sounds good.

Question I don't have a answer for yet,

What pressure do we supply?
What volume of oil do we need to pump?
Were or how will we fit the pressure relief valve?
Do we remove the existing pump (I feel the answer is yes)?
How do we drive the pump and from were on the crank?

All suggestion greatly appricated.
Tony

I am going to run between 65 psi and 85 psi. This is above Subaru's required pressure of 45 psi @ 5000 rpm and below Subaru's relief valve pressure. I recall also watching Turbo F1 engine stuff with Cosworth and they said "65 psi" so I'm happy with that figure.

I also think the flow will sort itself out. At 65 psi I would imagine the flow to be the same to build the same restriction because the restriction is still the same. Therefore pretty spot on with Subaru engineering etc.

Mine will be driven off the crank or the water pump idler pulley - which ever is cheaper to install. With the original pump you don't remove it - you simply tee into it and use it to feed the engine block - if you remove it you have a major oil seal issue at the front of the engine and a CNC replacement part will most likely be expensive. So I will be most likely removing the pump internals and then investigating what needs to be blocked. Zen do their dry sump this way and I think Dailey also.

However this brings me back to the sump design. I think we need half of it dead flat so we can mount the pump nice and flat. However this is the type of mount I'd be getting with the pump, however x 2 of these mounts. Then some brackets on the sump to mount with adjust for the belt.

[Dessertrunner]

bazza,
Is that photo of your pump?

[bazza]

Originally Posted by Dessertrunner

Alum plate to make the dry sump will cost $300. So now just need to decide if I am going to build one.
Tony

Option two is also a flat piece of ally laser cut or CNC cut out for the 2D shape.

Then 2 x basic drain point wedges welded on - I cannot see that costing more than $150 - thoughts? Kinda like this but far simpler:



Also filter drain fitting:

[bazza]

Originally Posted by Dessertrunner

bazza,
Is that photo of your pump?

No but it will be very similar. Want me to send it to you first so you can come up with a crazy design :P

[Dessertrunner]

The unit you show in the photo still has a lip and a cover bolts over it to stop the oil hitting the crank. We can try the flat plate But when you look at how close it will be to the crank I think it may cause the oil to hit.

Not sure were you are up to with your CAD guy but I am happy to help in anyway you want just let me know.

By the way if you go from 45psi to 65psi my wild guess is you will pump about 30% more oil. The adjust the pressure in the valve on the existing pump you will need to get a strong spring, washes won't work right.

Tony

[bazza]

Originally Posted by Dessertrunner

The unit you show in the photo still has a lip and a cover bolts over it to stop the oil hitting the crank. We can try the flat plate But when you look at how close it will be to the crank I think it may cause the oil to hit.

Not sure were you are up to with your CAD guy but I am happy to help in anyway you want just let me know.

By the way if you go from 45psi to 65psi my wild guess is you will pump about 30% more oil. The adjust the pressure in the valve on the existing pump you will need to get a strong spring, washes won't work right.

Tony

You're also forgetting the stock windage tray - fair effort to get oil past that unit

I agree with the stock pump and flow. Would be handy to use it to control flow rather than install another part.

My CAD guy is coming over for beers this weekend with his CAD laptop - so we'll come up with a few drawings and I'll share it with you.

P.S. The tank arrived:

[Dessertrunner]

So were are you going to fit that sucker, I though it was a bit crowed uned the bonnet now?

[bazza]

Originally Posted by Dessertrunner

So were are you going to fit that sucker, I though it was a bit crowed uned the bonnet now?

The tank will go in the boot for the best possible weight distribution. I've just gotta move a few things around... nothing an hour or two with the TIG cannot solve

[Dessertrunner]

Just brought this pump for the dry sump a little dear then I would have liked but feel confident its good cond.
bazza has convinced me that the best way is to mount the pump on the sump plate which is a good idea so I need to figure a way to get the suction and belt drive sorted out.



Got to say I use to be **** scared of how to set up a dry sump and also if the expense was worth the gain. Thanks to bazza I have got a better understanding and feel we are on the right track.

For those who wonder were I am heading and why do all this it because I want to build a NA engine that revs to 10,000 (remember a couple of years back I said that) and doesn't blow to pices. I have been getting engines and currently have 6 not in cars, 3 in cars. The plan is to build them up then distroy them on the dyno, do a mood then distroy them again untill I get a engine that works and develops the power I need to make the racing SVX worth running. Also as everyone is aware Matt has had a hell of a time with engines and bearings so we need to solve the problem.

That is why I am moving to the next issue and with bazza's help and Harvey (he has some suggestions as well) we should be able to make a SVX engine fly.

Have a great day all.
Tony

[bazza]

Originally Posted by Dessertrunner

Just brought this pump for the dry sump a little dear then I would have liked but feel confident its good cond.
bazza has convinced me that the best way is to mount the pump on the sump plate which is a good idea so I need to figure a way to get the suction and belt drive sorted out.



Got to say I use to be **** scared of how to set up a dry sump and also if the expense was worth the gain. Thanks to bazza I have got a better understanding and feel we are on the right track.

For those who wonder were I am heading and why do all this it because I want to build a NA engine that revs to 10,000 (remember a couple of years back I said that) and doesn't blow to pices. I have been getting engines and currently have 6 not in cars, 3 in cars. The plan is to build them up then distroy them on the dyno, do a mood then distroy them again untill I get a engine that works and develops the power I need to make the racing SVX worth running. Also as everyone is aware Matt has had a hell of a time with engines and bearings so we need to solve the problem.

That is why I am moving to the next issue and with bazza's help and Harvey (he has some suggestions as well) we should be able to make a SVX engine fly.

Have a great day all.
Tony

Love ya work mate. I think wrap around brackets will work a treat for that pump as per my CAD idea on your sump base. Also is that a weaver or a dailey pump - super purchase indeed.


A suggestion with revving, start by getting in contact with this fella, http://www.youtube.com/watch?v=YvZpGAMgpzQ and his build thread: http://bbs.22b.com/forums/showthread...oob-in-30-Days

Then to do a fair bit of cheating, find the EJ's that can rev to 10,000 rpm and go from there. My old EJ257 used to rev to 9500 rpm... solid buckets, massive 272 cams with 11 mm lift, upgraded springs. Wasn't to much to it. Cannot WAIT to see what you come up with.

[Dessertrunner]

One more point,
I know why a dry sump has to have a well.

If the sump was a flat plate without a well so the suction was cut straight into the bottom of the plate there would be problems I will explain. The amount of oil that can get into that suction would be equilvant to the dia of the suction inlet times the height the oil flows over the edge of the suction. In other words if I had a 30mm suction I would need the oil to be 15mm deep in the try to get close to full suction, considering that the velicioty of that oil going into the suction is very slow the pump would suck mostly air. The more air it pumps the less oils is able to pump and the bigger the problem becomes.
This effect is simlar to what happens with flood control dams. I have heard them on the radio saying the dam has 130% of it full volume.

The well means that the depth of oil in the sump will be far less as the pump will be able to pump more oil due to the fact that the amount getting into the suction is the dia (or edge length) of the well ( which might be 300mmm) times the depth. At the suction the oil depth can be a lot higher with out effecting the crankshaft which intern means the pump will pump less air.

This has most likly bored the hell out of every one but I figure it might be worth sharing this wisdom of thought (even if I am wrong).

Tony

[bazza]

Originally Posted by Dessertrunner

One more point,
I know why a dry sump has to have a well.

If the sump was a flat plate without a well so the suction was cut straight into the bottom of the plate there would be problems I will explain. The amount of oil that can get into that suction would be equilvant to the dia of the suction inlet times the height the oil flows over the edge of the suction. In other words if I had a 30mm suction I would need the oil to be 15mm deep in the try to get close to full suction, considering that the velicioty of that oil going into the suction is very slow the pump would suck mostly air. The more air it pumps the less oils is able to pump and the bigger the problem becomes.
This effect is simlar to what happens with flood control dams. I have heard them on the radio saying the dam has 130% of it full volume.

The well means that the depth of oil in the sump will be far less as the pump will be able to pump more oil due to the fact that the amount getting into the suction is the dia (or edge length) of the well ( which might be 300mmm) times the depth. At the suction the oil depth can be a lot higher with out effecting the crankshaft which intern means the pump will pump less air.

This has most likly bored the hell out of every one but I figure it might be worth sharing this wisdom of thought (even if I am wrong).

Tony

From what I've read, people run -20 to -30 mm Hg suction in the sump, sometimes more. This means it can literally suck oil out of the air - it's lot of suction.

The engine is flowing up to 1L/sec, shared via 4 suction points (2 x heads and 2 x sump). This means if there is oil pooling, it's going to get sucked up quick smart and not hang around long. Also note that the pumps are designed to pump air. I recall reading that whatever the volume of oil coming into the engine, you want the pumps capable of returning DOUBLE that volume into the oil tank as most of this will be air.

Maybe we need to setup a static test... hmmm.

[Dessertrunner]

bazza you made me think if there is this much prssure on the suction line then were if the air coming from to satisify that requirement. in simple terms if there are 4 stages sucking and 1 stage pumping then the system can stand 25% oil 75%air. This is wild bush maths but it helps with undrstanding of the system.

So were are we satisfying the vacum from?
Tony

[bazza]

Originally Posted by Dessertrunner

bazza you made me think if there is this much prssure on the suction line then were if the air coming from to satisify that requirement. in simple terms if there are 4 stages sucking and 1 stage pumping then the system can stand 25% oil 75%air. This is wild bush maths but it helps with undrstanding of the system.

So were are we satisfying the vacum from?
Tony

Notice the little red valve looking thingy, very bottom right. It is installed on the start motor side's ROCKER cover breather. It basically is used to control the air coming into the engine case. The other breathers head back to the oil tank. The oil tank then has a breather tank seperate from it. No idea why Cosworth put it on one side.





Also Dailey engineering have dry sumped that 1000 bhp EG33 I've mentioned. In terms of dry sump engineering and proven race results, Dailey are at the point end, so anything they do is usually well engineered and done extremely well. This is their drainage system, note there is nothing to it:



More Dailey:


More Dailey:

[Dessertrunner]

Looking at that sump and taking on board your comment that maxium G force is under braking the sump in the picture eould not work that well. Putting the suction at the front is a better idea.
Tony