AOS on a DSM

Bryan 93 · Jan 20, 2016

Anyone ever try using an AOS (air oil separator) on a DSM? They are pretty common to the Subaru community. I am pretty surprised most just go with a catch can that vent with a small filter on a DSM. I always thought the vacuum was useful in helping to seal the bottom end. Any thoughts?

cheapracer · Jan 20, 2016

I run a catch can, but it vents back to the intake. It's just a basic separator. I have steel wool in there to act as a baffle

bastarddsm · Jan 20, 2016

Oil vapor causes detonation. You'll never ever draw enough vacuum on the crank case to help sealing under boost, unless you use a big vaccum pump.

Bryan 93 · Jan 20, 2016

cheapracer said:
I run a catch can, but it vents back to the intake. It's just a basic separator. I have steel wool in there to act as a baffle

It looks like my searching for AOS was the wrong term to use. Seems the DSM community just uses sealed catch cans. I do like the idea of an AOS rerouting the oil back into the system, using coolant to heat the system to ensure no condensation. Whether that vacuum source be the intake manifold, turbo inlet, or a vacuum pump.

bastarddsm · Jan 20, 2016

If your getting that much oil in your catch can fix your engine.... The oil stink sucks but there's not much you can do about it without hurting performance.

Bryan 93 · Jan 20, 2016

bastarddsm said:
Oil vapor causes detonation. You'll never ever draw enough vacuum on the crank case to help sealing under boost, unless you use a big vaccum pump.

And from what I have read the job of the AOS is to actually remove the oil vapor from the air before allowing it to enter back into the system.

The air/oil mix that passes through the separator must make abrupt changes in direction. The inertia of the oil particles exceeds the air/oil mix flow's suspension forces and the oil impacts the wall of the separator. It collects/coalesces with other particles and is allowed to drip back into the oil fill tube maintaining itself.

Not my words, from Grimmspeed.

Bryan 93 · Jan 20, 2016

bastarddsm said:
Oil vapor causes detonation. You'll never ever draw enough vacuum on the crank case to help sealing under boost, unless you use a big vaccum pump.

I did not know oil vapor causes detonation, thanks for that. Do you any idea how much it lowers the effective octane rating of the fuel being used? I did some searching since your post and see much arguing over this, no real data.

EDIT: Did my own research on this one. Not sure I really buy that the detonation theory is all that significant. In order to lower the octane rating of 94 octane gas down to 93 octane, assuming oil has an octane rating of 1, for every 10 gallons of fuel burned it would need to burn around 0.1 gallons of oil. That would be 1 gallon of oil for every 100 gallons of gas. Definitely never going to burn that much! Unless maybe my theory is flawed?

Michael@FP said:
If you're not pulling a vacuum on your catch can you are doing it wrong.

The stock PCV system other than the size of the lines is by far the best way to setup a crankcase ventilation system. If your motor is tight and you have very little crankcase issue more than likely a catch can does nothing for you other than removing the oil from the intake system.

If you have a some crankcase issue enough to where you see some light smoke on decel you're producing more than the stock PCV system can handle and it's pushing enough oil into the intake to make it smoke. Putting a catch can that doesn't pull a vacuum and is vented will make the smoke go away but won't really alleviate your crankcase pressure. It just gives the oil some place to go while the pressure persists, all you've effectively done is increase your crankcase pressure.

Now if you're on a freshly built motor, a motor with loose tolerances, or just a worn out motor you might have excessive crankcase pressure. This can cause all sorts of fun issues, smoking at idle/cruise, smoking under load, smoking on decel, and even pressurizing the bearing housing of the turbocharger and forcing oil out of it. In this instance to truly alleviate your crankcase issue you need to IMPROVE the stock system which means it needs to retain at least the vacuum under load. So the best way to do this is to increase the size of the lines and either run them directly to vacuum sources or run them to a baffled catch can and then run the vacuum sources directly off that.

We have done the testing and so had Garrett and Holset as little as .1PSI in a bearing housing can cause leakage. Let that sink in a for a minute just that little bit of pressure can cause this to happen. A catch can vented to atmosphere with stock sized lines is useless and if you measured pressure in your crankcase vs. the stock system you would see a rise in pressure with removing the vacuum and adding a catch can alone. So if you want to do it properly and get the best of both words, run a baffled catch can with it pulling a vacuum, no oily mess in your intake or engine bay and you still get the benefit of having your crankcase scavenged of excessive pressure.

-Michael

I see posts like this that really seem to indicate a vacuum source is needed. Competing theories that both seem to make sense. The AOS seems like a good solution for maintaining vacuum.

bastarddsm · Jan 20, 2016

You can do all the research you want, but the facts are that oil vapor causes detonation. The other facts are that a google search of car forums isn't research.

A really really good catch can might get some of the liquid oil out of the blow-by gasses. The problem is the oil vapor, the oil that is evaporated, and is now a gas, and has similar density to air. How are you going to centrifuge that out? Then how are you going to get the low octane HC's that got by the rings out of the blowby gasses?

If you recirculate the pvc back into the intake, you are pumping low octane detonation prone fuel into the engine. It's as simple as that. It might not hurt you, but it sure isn't going to help.

I'd like to see data on a stock pvc system on a stock car, or any setup running back into the intake pipe actually pulling a vacuum on the crankcase at WOT.

donniekak · Jan 20, 2016

If you want to draw vacuum on the can under boost, the best way is an exhaust Venturi.

My car doesn't burn oil, push it out of the vents, or blow the dipstick at 42psi of boost. Excessive blow by gasses need to be dealt with at the source, poor ring seal, or worn valve seals.

Crankcase vacuum helps ring seal an extremely small amount. What seals the rings is the 1,000's of psi of cylinder pressure behind them pushing them into the cylinder wall.

Bryan 93 · Jan 20, 2016

bastarddsm said:
You can do all the research you want, but the facts are that oil vapor causes detonation. The other facts are that a google search of car forums isn't research.

A really really good catch can might get some of the liquid oil out of the blow-by gasses. The problem is the oil vapor, the oil that is evaporated, and is now a gas, and has similar density to air. How are you going to centrifuge that out? Then how are you going to get the low octane HC's that got by the rings out of the blowby gasses?

If you recirculate the pvc back into the intake, you are pumping low octane detonation prone fuel into the engine. It's as simple as that. It might not hurt you, but it sure isn't going to help.

I'd like to see data on a stock pvc system on a stock car, or any setup running back into the intake pipe actually pulling a vacuum on the crankcase at WOT.

Yeah I wouldn't mind seeing data as well, one of the reasons I brought it up here. So far no luck, hopefully someone can chime in with some actual numbers.

YUPOK · Jan 20, 2016

I have to agree. Oil vapor will light before gas causing the detonation. There are other threads debating this but the bottom line is, there is always some level of toxin or moisture in the vent, why would you run that right back to your intake?

I'm currently working on an atmospheric vacuum system that should suck the crankcase without decreasing power, complicated pumps and getting the cleanest air into the system. Similar to the exhaust venturi system mentioned above but more compact. We are still working on it, need a few more nights in the shop.

bastarddsm · Jan 20, 2016

Bryan 93 said:
Yeah I wouldn't mind seeing data as well, one of the reasons I brought it up here. So far no luck, hopefully someone can chime in with some actual numbers.

Put a 2 bar map into your crankcase. That will give you a lot more information that searching the interwebz.

Bryan 93 · Jan 20, 2016

donniekak said:
If you want to draw vacuum on the can under boost, the best way is an exhaust Venturi.

My car doesn't burn oil, push it out of the vents, or blow the dipstick at 42psi of boost. Excessive blow by gasses need to be dealt with at the source, poor ring seal, or worn valve seals.

Crankcase vacuum helps ring seal an extremely small amount. What seals the rings is the 1,000's of psi of cylinder pressure behind them pushing them into the cylinder wall.

I have neither of those issues as well. It looks like the Evo community uses an AOS for their street driven cars. I might hit them up if no one here has history with them.

YUPOK said:
why would you run that right back to your intake?

To maintain a good vacuum, thus limiting internal crankcase pressures, and not have to run a separate pump. If the net effects are nearly zero on the possibility of preignition, which the calculations say it is since the amount of oil vapor entering the intake is extremely small, I wouldn't mind doing it.

Bryan 93 · Jan 20, 2016

bastarddsm said:
Put a 2 bar map into your crankcase. That will give you a lot more information that searching the interwebz.

If only I had a running car and it wasn't 10 degrees F outside I would do just that.

donniekak · Jan 20, 2016

How many people in here have ever cleaned out an intercooler on a car that have breather hoses on the turbo intake? I have. They get really nasty, and lose a lot of cooling efficiency. The oil from a blown turbo is easier to clean out than the long term sludge that builds up from breather fumes.

This stuff isn't rocket science, solutions are out there. If you want vacuum drawn on the crankcase through a catch can, an exhaust Venturi is a no down side solution.

For what it's worth, if you have a useful amount of vacuum in the turbo inlet pipe, you have an issue. A large enough filter, and intake pipe will produce very little vacuum. Turbochargers are a lot more sensitive to restrictions before them, than after them.

ThunderChild · Jan 27, 2016

donniekak said:
How many people in here have ever cleaned out an intercooler on a car that have breather hoses on the turbo intake? I have. They get really nasty, and lose a lot of cooling efficiency. The oil from a blown turbo is easier to clean out than the long term sludge that builds up from breather fumes.

This stuff isn't rocket science, solutions are out there. If you want vacuum drawn on the crankcase through a catch can, an exhaust Venturi is a no down side solution.

For what it's worth, if you have a useful amount of vacuum in the turbo inlet pipe, you have an issue. A large enough filter, and intake pipe will produce very little vacuum. Turbochargers are a lot more sensitive to restrictions before them, than after them.

I disagree a bit. An exhaust venturi can and does work to a point, but it's also not great for the roads.

There's bound to be a large difference in capabilities, but I would much rather have a vacuum only pull from the inlet pipe on a filtered can, vs muffler or resonator collapsing and causing a pressurizat ion from the exh back through the crank case.

I know it's a far cry, but a well functioning dual sealed catch can with internal filter and baffles works very well.

donniekak · Jan 27, 2016

How is an exhaust Venturi bad for the roads? If it's putting oil down on the road, imagine what it would be doing to your turbo/inter cooler.

If you are at the point where you actually need a catch can you should also have a straight 3-4" exhaust with no converter, no resonators, and a straight through muffler. Not that an exhaust clog could cause an issue on a properly setup system, they also employ a check valve.

Again, if you have useful vacuum in the intake pipe, you have to small of a filter or intake pipe.

A vented catch can works for many people as well.

Kapok6 · Jan 27, 2016

Kevin at Kiggly touched on oil in his intercooler causing detonation on his FB page maybe a year or so ago.

I also found this to be an interesting read/video.

http://www.jackstransmissions.com/pages/jacks-pcv

bastarddsm · Jan 27, 2016

Yeah, I don't think I would route PVC lines right into the intake like that....

Mines on the cheap, but 2x -10s off the vc stuffed into a stainless waterbottle. It's cheap and works good. Get the kiggly HLA and there will be almost no oil in the bottle. Only drawback is that it smells a bit after an oil change. I just don't understand these guys spending $500 on a pvc system. If you were concerned with the smell you could always leave the intake draw on it, and put check valves on the -10's so that the intake would draw a vaccum under cruize and keep the smell down.

ThunderChild · Jan 27, 2016

Maybe I'm missing something. out of all the pcv setups I've done with two sealed cans, I've never seen oil back in the intake or through the intercooler. Granted I haven't had any "race" setups or 600+ hp, but ill stand behind the setup until it stops working.

I have a freind running a mk2 supra with a built 7m swap right around the 650 whp mark, and he's running a twin sealed can setup I made for him, and hasn't had anything in the intercooler etc in 5 years.

If the lines were run like stock with no check valve I could see alot of vapors being pulled through the pre-turbo intake track, and have seen it many times. Once a can was installed, it caught everything visible, approx. 1/10th of a qt after 3k miles. Now there might be a slight drop in pull if a can is installed, but personally I'd rather have that than a venturi tube dumping oil into my muffler or onto the road. But perhaps in a race car it would be easier, possibly even more beneficial.

I am on board with the detonation issue. I know you could never get ALL of the oil particulate put of the system with cans, but for the majority of ppl they will work fine. At the same point, you can never filter all of the incoming air perfectly either. We're realists, not idealists.

bling5tatus · Feb 11, 2016

I, admittedly, didn't read all of the above comments. If you drip blow by oil (what you are able to separate from the gas coming from your VC vents) back into your pan, you're putting oil that is contaminated back in with your "clean" oil. This is a horrible idea and I'm not sure why OEMs do this. Maybe they just jack the maintenance intervals up in the owners manual, I dunno.

The Top DSM Community on the Web

AOS on a DSM

Bryan 93

cheapracer

bastarddsm

Bryan 93

bastarddsm

Bryan 93

Bryan 93

bastarddsm

donniekak

Bryan 93

YUPOK

bastarddsm

Bryan 93

Bryan 93

donniekak

ThunderChild

donniekak

Kapok6

bastarddsm

ThunderChild

bling5tatus

Similar threads

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds