Scavenging capacity / blocked oil return line bank 1 / AOS failure

I am driving my 2002 996 C2 now since 35k miles with a modified AOS. The hose leading to the intake behind the throttle body gets bypassed via a catch tank and the catch tank is further connected to the air filter box. This bypass has the advantage that under high rpm when the throttle body closes the AOS does not get subjected to the high vacuum situations which occur in the standard situation when the throttle body closes. Also no more dirt from the crankcase can get into to intake. When the AOS starts failing I will notice it as more oil will get collected in the catch tank. If the AOS collapses suddenly it will prevent my engine from a catastrophical failure. Normally the catch tank stays empty. Only very occasionally I find some oil traces in there. Mostly in combination with a spirited drive. However since last 150 miles I noticed considerable amounts of oil collected in the catch tank. Luckily I rerouted the AOS as otherwise my engine would have swallowed these amount of oil and would haven gone kaboom. AOS failed suddenly (was installed 45k miles ago). After installing a new AOS and a test drive I notice that still lots of oil reaches the catch tank. Only under (faster) right hand turn. So I suspect that on top of my broken AOS there is something wrong with the oil return system on bank 1. As the AOS is also connected to this bank I suspect the oil return capacity being hampered. The right hand turns makes that the oil gets collected in the head on bank 1. If at some point the oil gets not pumped away fast enough it will flood the AOS connection point. Maybe good to know that 2000 miles ago I got the bank 1 head rebuild after a valve seat on an intake valve of cilinder 3 got loose and cracked (see my other post here) . No real collateral damage visible except the broken seat, damaged valve and scuffed piston. I will go and check/replace the scavenge pump. Both bank 1 an bank 2 scavenge pumps were checked with the head being rebuild. Also check if the oil return on bank 1 is not compromised or partly blocked. Any other ideas what else to do or any other possible scenarios which would cause this?

I too run a modified AOS system, I don't trust the factory AOS set-up. I agree you should check the scavenge pump ( and orientation) and drive tang on the cam gear. If the scavenge pump checks out, I suggest you check for excessive blow-by. 1 intall a "T" fitting in the end of the AOS to TB hose 2 cap the straight open end of the T fitting and drill a .221 orfice hole in the cap. 3 on the 90 degree fitting install a manometer 4 make sure there are no leaks in oil cap,oil fill hose ect Run engine and observe crankcase pressure, with a long hose routed out engine lid and into window you can road test and observe pressure under varying loads. On a very healthy 300hp modern engine, crankcase blow-by pressure should be about 1.5 inches of water ( 35 liter/min blow-by gases @ 1000rpm) BEST KEPT SECRET OF ENGINE BUILDERS

Interesting! Thanks for sharing. However ... what would cause the asymmetry that on left hand cornering no oil expelled and the right hand turns excessively puking oil out?

Quotes source post: 15276034

There is symmetry in left and right head vent as for lateral g-force, but for longitudinal g-force the left head will see more G-force due to the cars ability to brake a lot harder than it can accelerate, since the left vent is toward the front of the head the combined lateral and longitudinal g-force will be higher..This higher oil level in the left head with higher air flow from blow-by gases may cause oil to "puke" .. But as we already talked about, check the left scavenge pump also..How much oil did you "puke" into your catch can ?

Last short test drive with only left hand cornering 0. The one with right hand corners 0.2qts This is mainly accelerating through the curve rather then hard braking before the curve. Keeping revs high and climbing. So the test is a steady approach and then just going faster through the curve and holding speed at the point of grip loss. First suspect is the scavenge pump indeed. Will go check things out. What about a clogged oil return channel?

I would suspect design, rather than failure, that causes right handed turns oil issues. A lot of 996 cars lose oil pressure in right handed turns.

Also, the AOS is located on the left side of the engine, and when making a right hand turn, the oil sloshes in the sump towards the left side of the engine, sometimes overwhelming the AOS with oil rather than crankcase fumes.

Is the catch can connected to the throttle body after the throtte plate like in stock form or somewhere before it?

It did note loose oil the 50k miles of my ownership. Even under much more spirited driving. The cornering speeds when the oil splitting starts to occur now are very reasonable/moderate. You could consider them as normal street driving with a +300bhp sportscar. If you would drive my car for your sunday drive and take a right corner entering a highway you would find the same considerable amount of oil in the catch tank. It would be more than enough oil to hydrolock the engine. The AOS is connected via the catch tank and further to the underside of the airbox. So no more high vacuum subjected to the AOS in high rpm when coming off the throttle.still there is some reasonable good underpressure via the outlet in the airbox for ventilating the crank case.

AOS is using the vacuum generated by the intake (after the throttle body) to evacuate the oil vapor of the crankcase. The suction would be maximum when the throttle is closed or even partially open (same principle as the brake booster). Since you connect the AOS (via an oil catch can) to the air filter intake, which is close to atmospheric pressure, I doubt there will be much suction power. That also means when you're at high rpm, generating oil vapor, the vapor can't be effectively evacauted when you lift off the throttle as would be in the stock setup.

Another thing I wonder is the DME expects the non-metered (non metered by the MAF) crankcase air in its stock MAP. Cutting off that crankcase air will probably require the DME to adjust for fuel trims to compensate that. Based on Porschetech3's flow number, the extra air is not negligible albeit not pure air (with reduced O2 content).

from what i've read, the m96 needs vacuum in the crankcase for both leak reduction and to assist the low tension rings to seal. I'd like to delete the AOS and just run a belt driven vacuum pump. not sure this has ever been done.

AOS=air oil separator...separates the crankcase oil fumes laden with oil mist and routes it to the intake manifold for emission purposes. The crankcase is under vacuum when running. Not sure how you can take the AOS out of the equation and replace with a vacuum pump, since the crankcase is naturally under vacuum. The AOS doesn't create the vacuum.

Sorry to slightly disagree with this. At least not these high vacuums which occurs at higher rpms when the throttle body is (nearly) closed.are needed The AOS as on the m96 and m97 engine has a catch-tank function and is therefore connected to the lower part of the oil level. So the oil is brought back from the AOS at the bottom of the pan, an oil return channel from the AOS runs into the crankcase and comes out at the bottom. This is what (like we all know) happens when the AOS fails: Braking on the engine, large vacuum in the inlet, the valve in the AOS should close but is defective and the vacuum has access to the bottom of the crankcase (this is the big problem!), the AOS fills with oil and the oil reaches the inlet. The engine starts to smoke and the chance of hydro-locking the engine is great. The solution to prevent this is to let the AOS breathe on the other side of the throttle body (as it is done with the GT3!). Then there is no under-pressure when braking on the engine (throttle body closed). You can run this by-pass via a catch tank and from there i.e. to the air filterbox. Under driving the air beign led to the inlet will help remove the crankcase fumes. You will also be warned in good time that the catch tank function of the AOS will stop because then there will be oil in external catch-tank, and it will never reach the inlet of the engine to cause damage. You even could skip the catch tank and run it directly to the airbox. About vacuum in the engine and the AOS: There is a (high) vacuum at the time of braking of the engine, the valve in the AOS blocks this vacuum as it closes under these conditions. During normal driving there is simply overpressure in the crankcase and this pressure pushes the AOS valve open and the vapors are exhausted via the inlet. The AOS valve prevents the build-up of a vacuum of the crankcase through the valve, otherwise the crankcase would be drained. Also to consider that under normal driving, even when the throttle body is only opened a bit the effective open surface area is much bigger compared to the small orifice of the AOS outlet to the engine intake. As fluids also look for the direction of the least resistance most of the air will be coming from the airbox. However the under-pressure on the AOS connection will be sufficient to evacuate the air being pushed out via the AOS. On top of that there is the venturi effect which will add a bit to the breathing of the AOS and crankcase. But still the pressure in the crankcase is the main driving factor of the crankcase ventilation.

Maybe quit trying to reinvent the wheel and put in a motorsports AOS?

Why reinvent the wheel as it can be done like this. Crankcase pressure will never go to high as it can freely breathe unobstructed. The airbox will provide under driving for some additional additional evacuation force. Additionally to prevent further over-pressure due to a blocked-off AOS line filled with oil a further breather line was made from the oil cap to the airfilter box (go compare the AOS design of the 997.2 and the 991!) Anyway the problem got solved by replacing the bank 1 scavenging pump. It did not have the optimal pumping capacity anymore. Not sure if it is normal that they 'wear out'

As Porschetech3 already indicated the crank case pressure is there. Stating tma healthy blow-by flow of 35l/min @1000 rpm. The way he describes it I understand they there will be a slight overpressure in the crankcase. Or am I crazy? Do not mistake vacuum and suction power. The suction power of the intake you need to evacuatie the crank fumes that get past the AOS as efficient as possible. As further the original AOS stays in place and in working condition you can consider that the breathing ability will be the same so the valve will open to prevent over-pressure conditions.

Quotes source post: 15279563

Correct. Modern engines utilize low tension rings which require crankcase vacuum to perform optimally by design, to reduce friction and improve fuel economy. However, we also use low tension rings as required with nikasil bores and we don't always have the option of running the crankcase under vacuum.

Quotes source post: 15279902

The early 5 chain engines don't have a direct bolt on Motorsports AOS (which is designed for a 3-chain engine), however I have made adapters over the years to allow you to use one. I just made another batch of them as suddenly there was a demand for them again. I don't think I've ever seen a Motorsports AOS fail either, so this is a sound recommendation.

Hi Silk can you post some pictures showing the catch can and pipe routings/ connection you have made?

Both catch can and oil cap end in the airbox under the airfilter. The airfilter will get more dirty so check and change more often. At idle the engine runs much smoother. As no unmetered gases enter via the AOS outlet into the intake.

I can not see the pictures?

No idea why, I can see the photo. Otherwise access the jpeg file here