S3 cam covers and breathing...
10-17-2017, 12:17 PM
#31
Rennlist Member
Join Date: Feb 2011
Location: Mostly in my workshop located in Sweden.
Posts: 2,226
Received 442 Likes
on
244 Posts
The BMW Alpina 4.6 L V8 engine I have taken apart right now has a modern much more efficient engine ventilation system which can be adapted to the 928 engine. First there is no what so ever ventilation at the valve covers. Second good pressure equalization between crankcase and cylinder heads through the cam chain areas in front of the engine. Three the cyclonic oil separator is sitting behind the cam chain venting through a one-way valve to the intake manifold. The check valve is spring loaded keeping it open except at idle when intake vacuum will close the valve not sucking oil into the engine intake system. The check valve is integrated in the intake manifold (like for the Alpina engine) or a different variant sitting separately outside of the motor connected with hoses. Separated oil will drain through a pipe down to the sump.
Åke
Åke
10-17-2017, 12:41 PM
#32
Rainman
Rennlist Member
Rennlist Member
my proposed system was not intended to deal with blowby/crankcase pressure per se - just to address the oil drainback issue, and to try to have a functional PCV setup via the added AOS.
i would assume that once everything was interconnected sufficiently any excess pressure in the crankcase would be vented to the intake out the AOS lines. excess being higher than atmospheric pressure.
i like Blake's suggestion of repurposing the smog pump, too.
i would assume that once everything was interconnected sufficiently any excess pressure in the crankcase would be vented to the intake out the AOS lines. excess being higher than atmospheric pressure.
i like Blake's suggestion of repurposing the smog pump, too.
I've had a bunch of theories about how to control this issue, too.....which I've actually gone forth and tried on both the dyno and in cars which are daily driven.
When I studied the excess crankcase pressure/cylinder head filling with oil/loss of oil pressure/severe oil ejection out of the valve covers that appeared when we dynoed Andy's engine at 6,000 rpms, I spent six solid months trying to solve the problem. Six months....that's all I did. Smog pumps, vacuum pumps, stacks and stacks of passive pieces, oil returns to existing holes in the pan (which are below the oil level, BTW), oil returns to the pan above the oil level, improved oil flow through the heads, decreases in oil volume to the head, etc., etc.
From that study I came up with a "passive" system for S4 and GTS street engines.....and an "active" system, for street cars that are driven more aggressively.
So, when I ask how these ideas worked for the individual (in real life), there's a bit of a smile on my face.....because I've been there.
Rob Edwards has an interesting video on what kind of volume of oil is inside the head at rpms (we put clear tubing onto the return from my active oil removal system from the heads). You might want to take a look at that video to get an idea of how large the issue really is....it was a real "eye opener".
When I studied the excess crankcase pressure/cylinder head filling with oil/loss of oil pressure/severe oil ejection out of the valve covers that appeared when we dynoed Andy's engine at 6,000 rpms, I spent six solid months trying to solve the problem. Six months....that's all I did. Smog pumps, vacuum pumps, stacks and stacks of passive pieces, oil returns to existing holes in the pan (which are below the oil level, BTW), oil returns to the pan above the oil level, improved oil flow through the heads, decreases in oil volume to the head, etc., etc.
From that study I came up with a "passive" system for S4 and GTS street engines.....and an "active" system, for street cars that are driven more aggressively.
So, when I ask how these ideas worked for the individual (in real life), there's a bit of a smile on my face.....because I've been there.
Rob Edwards has an interesting video on what kind of volume of oil is inside the head at rpms (we put clear tubing onto the return from my active oil removal system from the heads). You might want to take a look at that video to get an idea of how large the issue really is....it was a real "eye opener".
What were the results of this?
10-17-2017, 02:05 PM
#33
Nordschleife Master
It seems to me that we're talking about two related but different topics here.
The first is how to ventilate the crankcase without pushing oil out. As as I'm concerned, I've got a solution in my car that I'm happy with: Vent the crankcase from the front when throttle is open and from the rear when the throttle is closed. Put in sufficient baffling and the cyclonic separators in the venting paths. I know that this works, but that doesn't mean that other things may not work also. See the Gatorade bottles in John's video.
The second is how to make sure that oil drains from the heads. This is harder (but not impossible!) to measure or experiment with, so one in my opinion should be guided by some basic engineering math. Otherwise, experiments are not going to be cost effective.
The oil drainage problems are directly related to crankcase pumping mean effective pressure. If your software reports the number CPMEP that is signed to be negative, that's what I'm talking about. For a given rod ratio and engine type (V8 in our case), CPMEP is approximately proportional to the following formula:
k = pressure loss coefficient of crankcase breather holes and the oil pan bay-to-bay breathing area
rho = crankcase gas density
V = cylinder displacement
A = aggregate breathing area between bays
rpm = engine speed
CPMEP is proportional to k*rho*(rpm*V/A)^2
Holding everything else constant, increasing the rpm from 5000 rpm to 7000 rpm approximately doubles the CPMEP. The hp lost to crankcase pumping goes up by about a factor of 2.7x. Where does this power go? Among other things, pushing oil up the drains to the heads.
If you increase the displacement by 30%, you'll need to increase the breather area by 30% or reduce the crankcase gas density by 41% to keep the CPMEP and hp lost to pumping constant. Porting the breather ports to flow better might also help, but you're not going to get a 41% improvement there.
By far, the easiest way to reduce CPMEP in the 928 engine of given displacement, rod ratio, and redline rpm is to add an oil pan spacer. That's in my opinion the number one thing one should do if one is experiencing oil drain problems in a 928 engine.
If you add a windage tray, the windage tray needs to have breathing area thru it (per bay) that is significantly larger than the bay-to-bay flow area under the windage tray in the pan. You can't just stick whatever windage trays in there without understanding the impacts on gas flows.
The above approximate CPMEP formula doesn't capture the direction of the pumping pulses. Looking at the crankcase geometry, however, one can immediately see that the crankcase pumping pulses are directed towards the head oil drain holes on one of the banks. Shielding those holes from piston pumping pulses will significantly improve head oil draining (a verified fact). That's in my opinion the number two thing one should do if one is experiencing oil drain problems in a 928 engine.
The first is how to ventilate the crankcase without pushing oil out. As as I'm concerned, I've got a solution in my car that I'm happy with: Vent the crankcase from the front when throttle is open and from the rear when the throttle is closed. Put in sufficient baffling and the cyclonic separators in the venting paths. I know that this works, but that doesn't mean that other things may not work also. See the Gatorade bottles in John's video.
The second is how to make sure that oil drains from the heads. This is harder (but not impossible!) to measure or experiment with, so one in my opinion should be guided by some basic engineering math. Otherwise, experiments are not going to be cost effective.
The oil drainage problems are directly related to crankcase pumping mean effective pressure. If your software reports the number CPMEP that is signed to be negative, that's what I'm talking about. For a given rod ratio and engine type (V8 in our case), CPMEP is approximately proportional to the following formula:
k = pressure loss coefficient of crankcase breather holes and the oil pan bay-to-bay breathing area
rho = crankcase gas density
V = cylinder displacement
A = aggregate breathing area between bays
rpm = engine speed
CPMEP is proportional to k*rho*(rpm*V/A)^2
Holding everything else constant, increasing the rpm from 5000 rpm to 7000 rpm approximately doubles the CPMEP. The hp lost to crankcase pumping goes up by about a factor of 2.7x. Where does this power go? Among other things, pushing oil up the drains to the heads.
If you increase the displacement by 30%, you'll need to increase the breather area by 30% or reduce the crankcase gas density by 41% to keep the CPMEP and hp lost to pumping constant. Porting the breather ports to flow better might also help, but you're not going to get a 41% improvement there.
By far, the easiest way to reduce CPMEP in the 928 engine of given displacement, rod ratio, and redline rpm is to add an oil pan spacer. That's in my opinion the number one thing one should do if one is experiencing oil drain problems in a 928 engine.
If you add a windage tray, the windage tray needs to have breathing area thru it (per bay) that is significantly larger than the bay-to-bay flow area under the windage tray in the pan. You can't just stick whatever windage trays in there without understanding the impacts on gas flows.
The above approximate CPMEP formula doesn't capture the direction of the pumping pulses. Looking at the crankcase geometry, however, one can immediately see that the crankcase pumping pulses are directed towards the head oil drain holes on one of the banks. Shielding those holes from piston pumping pulses will significantly improve head oil draining (a verified fact). That's in my opinion the number two thing one should do if one is experiencing oil drain problems in a 928 engine.
Last edited by ptuomov; 10-17-2017 at 04:36 PM.
10-17-2017, 06:21 PM
#34
Rennlist Member
The thing that caught my attention many years ago was when Louie posted his video of what happens at high engine speeds and left hand bends. Louie grafted a video camera onto a modified sacrificial cam cover. The video demonstrated how suddenly the situation went from a seemingly controlled situation to one that had gone ballistic.
Obviously the engine direction swirls the oil cloud in one direction and when cornering hard, for instance at 1G lateral acceleration, the oil wants to sit at 45 degrees instead of flat. The crank action then washes the oil up through the drain holes, some level is lost as the heads fill and hopefully the oil can return before the bearings give up the ghost.
If we work on the principal that every action has an equal and opposite reaction, then on the other bank, the crank action is possibly creating a venturi effect that might possibly suck oil down from the head ports on the other side. Thus having crossover pipes from the cam covers might just help pump oil from the flooded head, over to the other side and back down the drain holes there. Thus I have this feature.
One has to vent a bigger volume of gas than the stock system, can handle and the base of my filler neck is gutted, I have the John Kuhn baffle with the crinkled wire mesh within the body and this knocks most of the oil mist out of the venting gases. I then have a 1 inch vents line from under the filler neck straight into the pro vent separator, then two smaller lines to return the scrubbed gas to the inlet manifold. On top of there is a small mod that Tuomo designed that passes some gas to the inlet manifold under braking. Oil consumption decreased noticeably with these mods but the motor still consumes oil- just that it is now more acceptable.
The problem is noticeably worse with higher rpms. I remain convinced that engine breathing is something that just got wrong. I have no explanation as to why some examples have little oil consumption and other devour the stuff.
Obviously the engine direction swirls the oil cloud in one direction and when cornering hard, for instance at 1G lateral acceleration, the oil wants to sit at 45 degrees instead of flat. The crank action then washes the oil up through the drain holes, some level is lost as the heads fill and hopefully the oil can return before the bearings give up the ghost.
If we work on the principal that every action has an equal and opposite reaction, then on the other bank, the crank action is possibly creating a venturi effect that might possibly suck oil down from the head ports on the other side. Thus having crossover pipes from the cam covers might just help pump oil from the flooded head, over to the other side and back down the drain holes there. Thus I have this feature.
One has to vent a bigger volume of gas than the stock system, can handle and the base of my filler neck is gutted, I have the John Kuhn baffle with the crinkled wire mesh within the body and this knocks most of the oil mist out of the venting gases. I then have a 1 inch vents line from under the filler neck straight into the pro vent separator, then two smaller lines to return the scrubbed gas to the inlet manifold. On top of there is a small mod that Tuomo designed that passes some gas to the inlet manifold under braking. Oil consumption decreased noticeably with these mods but the motor still consumes oil- just that it is now more acceptable.
The problem is noticeably worse with higher rpms. I remain convinced that engine breathing is something that just got wrong. I have no explanation as to why some examples have little oil consumption and other devour the stuff.