IW STi Forum banner

81 - 100 of 1821 Posts

·
Registered
Joined
·
3,026 Posts
First of all, thanks for posting those diagrams and starting an interesting discussion. What follows isn't just aimed at you specifically, but at everyone following this thread...

I don't "get" the diagram above. Is there positive pressure in the crankcase the intake manifold is in vacuum?

If there is, then there's really no need for the PCV anyway. I mean we might as well put a cork where the PCV is... The crankcase can just ventilate itself into the turbo inlet due to the crankcase pressure and the (however mild) inlet vacuum, as shown in the 'under boost' diagram.

If there is not any positive pressure in the crankcase in cruise, then I don't see why the intake manifold vacuum pulls much air from the crankcase. I understand the vacuum created by the intake manifold, but the path of least resistance is to just pull air in from the turbo inlet. It seems to me that without positive crankcase pressure, the line that runs from the PCV tee to the crankcase is basically just stagnant, with the PCV pulling lots of fresh air over the top of the tee, and perhaps pulling a minute amount of air from the crankcase just from venturi effect at the tee.

Why does air flow from the intake into the heads? Where does that air go? I don't see why the air in the head breather lines doesn't just stagnate as well. If anything I'd expect it to flow back to the turbo inlet, due to the mild vacuum there and perhaps mild crankcase pressure due to blow-by. Does something beween the crankcase and heads pump air from the heads into the crankcase, or even just act like a one-way valve?
2nd to last paragraph: This is a very good point and something I had not considered. When the pcv line is sucking in air/crankcase gases a lot of it may be coming from the turbo inlet and not the crankcase (at least some if it) - kind of like taking a pull from pipe without having your finger on the carb :lol:. This must be why the line going from the plastic pcv "y" fitting to the turbo inlet is restricted to about the size of the end of a chop stick at the fitting. I would still think that most (or more) of the air it's pulling in would be from the crankcase. I say this because the pcv valve sits directly on top of the crankcase fitting with a 1/2-3/4" ID line going right to the the pcv valve, unobstructed. The line going to the turbo inlet sits above the pcv valve and is restricted to the size of the end of a chop stick. It really makes me wonder why they chose a setup like this; I wish I was more knowledgeable on the subject.
 

·
Registered
Joined
·
1,174 Posts
I wasn't aware of the restriction or size difference on the inlet side of the PCV tee, but it sounds like a good way to at least partially put a finger on the carb. :)

I agree with your assessment of the airflow through the heads. The arrows in that diagram must be backwards. Probably the only difference between vacuum and boost is the direction of airflow at the PCV tee. In boost it's as described in the Boost diagram - crankcase gases turn left at the tee. In vacuum the crankcase gases just take a right at the tee and get drawn in via the PCV valve, and all other flow is the same as the boost diagram.

Crawford's approach seems totally reasonable to me now. Instead of crankcase gases being sucked out by a strong manifold vacuum that's compromised by the hose between the tee and the inlet, they're sucked out by a weak inlet vacuum augmented by a thicker-than-stock hose.

I'm going to do my Ixiz the same way, as the OP did.
 

·
Registered
Joined
·
1,174 Posts
Why do you combine both the clean and dirty side of the PCV system? These are two mutually exclusive parts of the PCV system that should NOT be combined.
I can't make sense of your 'clean side' and 'dirty side' terminology. The closes I can come is by looking at it from the perspective of the "stock system under vacuum" diagram in the first post in this thread...

Do you mean that "clean" air should be drawn into the heads and expelled as "dirty" air coming from the crankcase vent? If so, what provides the pressure differential to move air from the intake into the heads? The flow lines in the diagrams in the first post show that happening, but as I noted a couple posts ago, I don't think that's what actually happens.

What am I missing?

Can you add some flow lines to the 'Crawford' diagram in the first post, to show the flow path that you're concerned about?
 

·
Premium Member
Joined
·
616 Posts
I can't make sense of your 'clean side' and 'dirty side' terminology. The closes I can come is by looking at it from the perspective of the "stock system under vacuum" diagram in the first post in this thread...

Do you mean that "clean" air should be drawn into the heads and expelled as "dirty" air coming from the crankcase vent? If so, what provides the pressure differential to move air from the intake into the heads? The flow lines in the diagrams in the first post show that happening, but as I noted a couple posts ago, I don't think that's what actually happens.

What am I missing?

Can you add some flow lines to the 'Crawford' diagram in the first post, to show the flow path that you're concerned about?
One of the major reasons for the PCV system is to evacuate combustion by products from the crankcase. The dirty side pulls out of the crankcase and into the intake manifold or turbo inlet. Because of the draw on the crankcase from the dirty side, air is pulled into the valve covers via the clean side.

The flow lines in this picture are not exactly correct:



As stated earlier in the thread, the clean side will flow in both directions.
 

·
Registered
Joined
·
1,513 Posts
I wasn't aware of the restriction or size difference on the inlet side of the PCV tee, but it sounds like a good way to at least partially put a finger on the carb. :)

I agree with your assessment of the airflow through the heads. The arrows in that diagram must be backwards. Probably the only difference between vacuum and boost is the direction of airflow at the PCV tee. In boost it's as described in the Boost diagram - crankcase gases turn left at the tee. In vacuum the crankcase gases just take a right at the tee and get drawn in via the PCV valve, and all other flow is the same as the boost diagram.

Crawford's approach seems totally reasonable to me now. Instead of crankcase gases being sucked out by a strong manifold vacuum that's compromised by the hose between the tee and the inlet, they're sucked out by a weak inlet vacuum augmented by a thicker-than-stock hose.

I'm going to do my Ixiz the same way, as the OP did.
Thanks for seeing the light. We spent six / seven months of street and track testing to develop our system to the stage it's at now.

Team Crawford
 

·
Registered
Joined
·
1,174 Posts
As stated earlier in the thread, the clean side will flow in both directions.
I can see why the heads would flow air out toward the turbo inlet. I can't see why air would ever flow from the intake into the heads. In other words, I don't see why you call it the "clean" side.
 

·
Premium Member
Joined
·
616 Posts
I can see why the heads would flow air out toward the turbo inlet. I can't see why air would ever flow from the intake into the heads. In other words, I don't see why you call it the "clean" side.
As the crankcase is evacuated via the dirty side it will create a slight vacuum. Since there is not a one-way check valve on the clean side, it pulls air from this open source and fills the vacuum. This is how the PCV system works to flush the crankcase of combustion byproducts. Slight vacuum on one side with open end on the other, what do you think happens? There is not a flow director on the clean side like on the dirty side.

First of all, thanks for posting those diagrams and starting an interesting discussion. What follows isn't just aimed at you specifically, but at everyone following this thread...

I don't "get" the diagram above. Is there positive pressure in the crankcase the intake manifold is in vacuum?
Positive pressure in the crankcase due to blowby. Under boost, pressure in the intake manifold. No boost, vacuum in the intake manifold.

If there is, then there's really no need for the PCV anyway. I mean we might as well put a cork where the PCV is... The crankcase can just ventilate itself into the turbo inlet due to the crankcase pressure and the (however mild) inlet vacuum, as shown in the 'under boost' diagram.
You don't understand the way a variable orifice PCV works. A PCV valve is a controlled vacuum leak, with a variable orifice valve allowing more flow with less vacuum and less flow with more vacuum. If you were to map crankcase flow, it is low at idle and increases with engine speed. A variable orifice PCV valve works in the same manner by limiting flow (smaller vacuum leak) when flow is lower and allowing full flow (larger vacuum leak) when PCV flow is the highest. If you remove the PCV valve from the circuit, your throwing away the vacuum control function within the system and allowing the same flow at all engine speeds.

If there is not any positive pressure in the crankcase in cruise, then I don't see why the intake manifold vacuum pulls much air from the crankcase. I understand the vacuum created by the intake manifold, but the path of least resistance is to just pull air in from the turbo inlet. It seems to me that without positive crankcase pressure, the line that runs from the PCV tee to the crankcase is basically just stagnant, with the PCV pulling lots of fresh air over the top of the tee, and perhaps pulling a minute amount of air from the crankcase just from venturi effect at the tee.
Under non-boost conditions there is 16 - 17 inches of vacuum in the intake manifold and nowhere near that in the air inlet. How is the turbo inlet the path of least resistance? Under boost conditions the PCV valve closes, but it's important to note that stock PCV valves will leak under boost introducing additional pressure to the PCV system. When the PCV valve is closed due to boost, flow is diverted to the turbo inlet through the dirty side 'Y'. At this point, there is significant flow in the air inlet line to evacuate the crankcase. If this reaches capacity flow will reverse through the clean side.

Why does air flow from the intake into the heads? Where does that air go? I don't see why the air in the head breather lines doesn't just stagnate as well. If anything I'd expect it to flow back to the turbo inlet, due to the mild vacuum there and perhaps mild crankcase pressure due to blow-by. Does something beween the crankcase and heads pump air from the heads into the crankcase, or even just act like a one-way valve?
See above.


If anyone with an AOS is doing UOA before and after, please post the results. Obviously, the manufacturer has not performed complete testing of the product or will not publish the data.

If you look at commercial air/oil separators from Mann+Hummel (Pro Vent 200) and Racor (CCV4500), they are installed inline on the DIRTY side of the PCV system with the clean side left intact. These systems also have built in pressure regulators to modulate crankcase flow and keep it evacuated.
 

·
Registered
Joined
·
2,490 Posts
I didn't have oil in my inlet pipe until I installed my Crawford AOS. I don't really understand how it is supposed to work. Having the crankcase vent line combined with the AOS return line just doesn't make sense to me. If you have pressure moving vapor into the can from the valve cover vents and the crank case vent, how is oil going to drain from the can back into the block being that the crank case vent is having to pull double duty as a vent and return line at the same time. I hope that description makes sense. I've had to vent my AOS to the atmosphere to keep from having my inlet lined with oil. It almost seems like it would make more sense to run the AOS return line into the turbo oil return line.
With all of that said, I am no where near any kind of exert on the PCV system. I try all of the time to make sense of it myself.
 

·
Registered
Joined
·
1,254 Posts
I had the PCV fail on my 08 STI and put a quart or 2 through the intake. The intercooler was full of oil. It took 3 months to fully get it all out. The PCV failure also caused detonation and ring, piston, and rod bearing failure.
 

·
Registered
Joined
·
8,267 Posts
I agree with your assessment of the airflow through the heads. The arrows in that diagram must be backwards. Probably the only difference between vacuum and boost is the direction of airflow at the PCV tee. In boost it's as described in the Boost diagram - crankcase gases turn left at the tee. In vacuum the crankcase gases just take a right at the tee and get drawn in via the PCV valve, and all other flow is the same as the boost diagram.

Crawford's approach seems totally reasonable to me now. Instead of crankcase gases being sucked out by a strong manifold vacuum that's compromised by the hose between the tee and the inlet, they're sucked out by a weak inlet vacuum augmented by a thicker-than-stock hose.
Totally. I have disagreed with that diagram from day one. There are positive pressure oscillations in the valve covers just like in the crank case.

The crank case would actually have to be a vacuum in order for air to flow from the inlet into the valve covers. Not only that, but the vacuum has to be quite a bit greater than the vacuum inside the inlet tube. I just don't see this ever happening on high boost applications.

Now, maybe this is why subaru used the 1/4" line from the PCV directly to the manifold. The only way for air to flow in to the valve covers in a cruising situation is if that 1/4" line can actually create a vacuum greater than the vacuum in the inlet tube. I just can see this happening.

I'm perfectly fine venting both the PCV and the valve covers like I have done on all my previous cars. Especially on a 65K ej257 where I'm sure I have a little wear on the valve guides creating more pressure inside the valve cover.
 

·
Registered
Joined
·
8,267 Posts
If you have pressure moving vapor into the can from the valve cover vents and the crank case vent, how is oil going to drain from the can back into the block being that the crank case vent is having to pull double duty as a vent and return line at the same time.
Yeah, I understand where you're coming from.

The drain from the AO separator isn't exactly 'flowing' oil like you'd think. Any oil that accumulates in the AO separator is immediately led down the drain tube. Picture a very small amount of oil constantly running down the side of the drain tube.
 

·
Registered
Joined
·
2,490 Posts
Yeah, I understand where you're coming from.

The drain from the AO separator isn't exactly 'flowing' oil like you'd think. Any oil that accumulates in the AO separator is immediately led down the drain tube. Picture a very small amount of oil constantly running down the side of the drain tube.
Yeah, but if you have CC pressure pushing back up the drain line, which has to be happening since one of the can supply lines shares the drain line, where is the oil going to go? I mean we are not talking about "oil flow" we are talking about oil vapor. It seems that it will take the path of least resistance. The turbo inlet line...
 

·
Registered
Joined
·
8,267 Posts
Yeah, but if you have CC pressure pushing back up the drain line, which has to be happening since one of the can supply lines shares the drain line, where is the oil going to go?
Oil is WAY heavier than crank case gas. Gravity is plenty of force to make sure it drains into the oil pan instead of being sucked back up into the can. Remember, the stuff that;s draining is condensed oil vapor...it's oil in liquid form. The oil goes down, and the gasses pass right by it in the opposite direction. If the drain was filled with oil, then there would be a big problem.
 

·
Registered
Joined
·
8,267 Posts
I wasn't planning on draining it back anyways.

Try and run the drain to a can of some sort and see if that helps. I hate to sound like a crawford fanboi (cause I don't own a single piece of there gear) but swirl pots are among the top ways to separate vapor from air.
 

·
Premium Member
Joined
·
616 Posts
<snip> The crank case would actually have to be a vacuum in order for air to flow from the inlet into the valve covers. Not only that, but the vacuum has to be quite a bit greater than the vacuum inside the inlet tube. I just don't see this ever happening on high boost applications.

Now, maybe this is why subaru used the 1/4" line from the PCV directly to the manifold. The only way for air to flow in to the valve covers in a cruising situation is if that 1/4" line can actually create a vacuum greater than the vacuum in the inlet tube. I just can see this happening.
For a dedicated race car, high boost for long periods would be common place. For a street car, high boost is the exception unless your one of those people that has the pedal to floor everywhere they go.

This means the intake will be under a vacuum with the PCV open for longer periods than it's closed (under boost). Under these conditions, MANVAC will be much higher than turbo inlet vacuum where air is pulled through the valve covers into the crankcase and out of the PCV valve.

Typically, the clean side won't reverse flow unless dirty side flow is at capacity.

I'm perfectly fine venting both the PCV and the valve covers like I have done on all my previous cars. Especially on a 65K ej257 where I'm sure I have a little wear on the valve guides creating more pressure inside the valve cover.
Eliminating the PCV system and venting to atmosphere is an option, but not a real good one. You end up with oil all over the engine bay from saturated breather filters.
 

·
Premium Member
Joined
·
616 Posts
<snip>but swirl pots are among the top ways to separate vapor from air.
Not for automotive PCV systems. Flow is too low to properly induce the swirl effect to remove carry over oil from the PCV stream. Swirl pots are one the least efficient methods of removing carry over oil from a PCV stream.

Using a multistage catch can that slows PCV flow to less than 1 m/s is the best method. The key to carry over oil removal is slowing the PCV stream enough to let gravity do it's job.
 
81 - 100 of 1821 Posts
Top