AOS Explained???

I'm sure the AOS has been beaten to death here as much as the IMS but I'd like an explanation why this happens only once in a while not always. If the AOS is bad wouldnt the car alway smoke ? I've noticed only on real hot days when I have been hitting the gas hard and my oil temps are up that the car will smoke at startup after a brief sit. On days I take it easy(when the wife is with me) it never ever smokes. Is it my AOS or could the oil be thinning and getting by the valve seals or rings? Or, is this normal for an aging AOS? My 996 has 82K and soon will be heading to Jake at Raby for the solution and AOS plus other items so I am not worried but would like an explanantion if someone has one. Thanks and love this Forum!

A puff of smoke - even somewhat sizeable - on startup is incredibly common and is not indicative of a failed AOS. The smoke you get when the AOS goes is entirely different... we're talking James Bond smokescreen, not just a big puff of smoke.

Porsche flat-6s are designed for tight tolerances while hot. As different metals expand at different rates, this unfortunately mean tolerances are much looser cold, and that oil can seep past piston o-rings and into the cylinders. You don't normally notice this when you're driving because the quantity is very small and the catalytic converter catches what's left. Or after long runs because the engine seals well when hot, and by the time it cools and the tolerances open up, the oil has drained back into the sump. But when you turn off the engine before it's warm, small quantities of oil to seep into the cylinders- and then burn off in a small puff of blue smoke when you start the car. Is this what you're seeing? When the AOS goes, it's different, it won't smoke immediately, but generally after a short delay as it takes time for the vacuum pressure to build up and suck oil into the AOS. When it smokes, it SMOKES- like 5CHN3LL says, it's like James Bond trying to blind a helicopter. A failed AOS won't always smoke, though; another (less common) symptom is a godawful screeching that decreases during acceleration.

Yes it is just a puff of smoke at startup and only lasts a few seconds. Its funny though it only happens when I run it hard. could the oil be thinning out after driving it like I stole it? It only happens on my days off when I am out driving it like it was designed to be driven, it never happens during the week when I use it to commute back and forth to work. The oil doesnt have more than 1500miles on it, maybe I should go to 5w-40 rather than 0w-40.

Ensure you don't overfill the oil sump. Also don't shut the car down right after driving it hard. High idle it for a minute to circulate coolant and oil and normalize the engine temp, then shut it down. I believe this is actually in the manual. Mine would puff at startup at autocross when my friend drove it because they were pulling into the paddock after a run and immediately shutting the car down. Once I got them doing the aforementioned idling before shutdown, the problem went away.

While the advice about not overfilling oil is sage, don't interpret that to mean that you should run the car low on oil. Fill it so the dash indicator is completely - or just one bar from completely - full. Oil starvation in our cars is a bad, bad thing.

Quotes source post: 10613099

I attribute this to the AOS just being lousy at its job. Its job may be made harder by the condition of the oil. Lots of miles and thus probably more contaminated with unburned fuel and water the oil is more likely to froth and foam and the engine crankcase sees a heavier load of oil vapor. The AOS is unable to remove this and it gets routed to the intake manifold. The oil vapor can't make that sharp turn at the speed at which is traveling and the intake walls end up with some oil on them. If the engine is running the oil gets consumed with no one the wiser though the engine's oil consumption could be higher than it would be with a better working AOS. If the engine is shut down before this oil gets drawn into the engine and burned the oil collects on top of the intake valves if they are closed or in the chamber if they are open. Upon engine start of course there is the smoke. In your car's case the oil is relatively fresh. You have not overfilled it. My advice would be if you drive the car hard before you shut it off spend some time driving like you do during the week to give time for any oil on the intake from the hard driving to be ingested by the engine while it is running. Besides this gives the engine a chance to shed its heat load, to cool down some. Consider this milder driving to be the cool down lap. And just before you shut off the engine for the day let it idle a while, a minute say though often I let it ldle less than that. This gives any oil on the intake walls one last chance to make it to the engine and be burned before you turn off the engine. There is no need to switch to another oil.

Quotes source post: 10613761

I realize I am raising this thread from the dead, but it is not about the OP's question. I saw this comment, & I have to ask, is the oil level inside correct when the car has been sitting & cold, or after it has gotten hot, stopped, then started again, because there is a difference?

The most complicated PCV valve that the Germans have designed.

Quotes source post: 12137973

The short answer is the oil level is correct if the oil level measured cold is down a bit (1 to n bars) below the max provided that when the oil level is measured hot the level is at the max line. The proper oil change procedure has the engine hot when the oil is drained. And then it calls for a specific amount of oil to be added to the engine. This is then used to confirm the electronic oil level system is working properly. For instance the oil change procedure for my Turbo calls for the hot oil to be drained for 20 minutes (IIIRC) and 7.8l of oil be added to the engine. The tech then runs the engine again until the engine is hot enough to perform and oil level check -- the Turbo engine has to be running at idle and the engine/oil up to temperature before the oil level system will take a reading and I might also mention that since the engine was hot when the oil was drained it doesn't take much idle time to heat up the fresh oil -- and the factory manual says the oil level should be at the max line. Now there is no way to check the oil level cold in the Turbo. But I can and do check the oil level cold (and hot) in my Boxster. When the engine is cold the oil level is down some, from anywhere from 1 bar to in "extreme" cold down to 3 bars. But once the engine is up to full operating temperature and I shut off the engine and then as soon as the oil level count down timer starts at 5 seconds and I get an oil level reading the oil level is at the max. line. The oil expands when hot and contracts when cold. One needs to know, be confident that at an oil change the proper procedure is followed so the oil level is correct and the electronic oil level system is checked and is found to be working correctly. Thus when one checks the oil level cold and finds it down a bit (1 to N bars) he doesn't add oil that then when the engine is up to temperature results in too high an oil level. And once in a while the oil level should be checked when the engine is hot. A good time to do this is after filling the gas tank, provided of course the engine was fully up to temperature. The tank filing takes some time and this is enough time for the hot oil to drain into the sump so when I get back into my Boxster the count down timer starts at 5 seconds -- which is a sign the engine was hot enough -- and the reading will reflect the amount of oil in the engine. And added plus is the drive where one parks his car to refill from the pump nozzle is level so one gets a consistent reading from refill to refill and from station to station. (It is amazing how when I pull into my local station how often I end up at the same pump each time. Even when I travel sometimes thousand of miles I end up refueling at the same stations I refueled at the last time through the area and more often than not at the same pump, too.)

Really strange, when my oil is hot, it reads LOWER than when cold.

AOS problems - contributory factors I have found here: 1. Oil foaming is a contributory factor= wrong spec oil 2. DE/AX?Track -see 3.? 3.Overfilling -well covered by Macster above. 4.Uro brand AOS is suspect? The revisions to the AOS and the changes 'plumbing' of the AOS system in later cars confirm this is a problem part/system ? Here are two of the revisions. Others may comment on any improvements in theses revisions and if they can be retrofitted. 996-107-026-01 996-107-023-04 The later cars drain the AOS into the oil filler tube? Can this improved layout be retrofitted to older 986/996?

Quotes source post: 12137973

Ideally your oil should be between 1/4 down and at the top mark on the stick when warm after sitting on level ground for ~20minutes with ambient temp ~70 deg F. When cold your indicator may show from 2 bars down(or more) to full depending on calibration. On my 2003 C2 Cab that equals 1 full bar down after the car has been sitting in the garage overnight before the first start of the day. On the dipstick the level indicates ~3/4 full. After warm up the top bar is usually lit and the oil level is almost at the top mark on the stick.

Quotes source post: 12138225

The oil level reading lower when the engine/oil is hot (fully up to temperature "hot") is contrary to what have observed over the years. It is important that the oil level when checked hot that the engine and oil are fully up to temperature. Oh, and I can add that once in a while, a great while, after just a short trip -- a trip that is just a couple of blocks long as I sometimes stop at a convenience store for a minute before heading out for a longer trip -- that sometimes when I get back in my car and turn on the key the count down timer for the oil level check doesn't start at say some large time value but surprisingly starts counting at just 5 seconds. But because I know that in the short drive the engine didn't get hot I really don't expect to be able to check the oil level so often, most of the time, while I note the 5 second count down timer starting value I just keep turning the key to the start position and start the engine. But the few times I have had the presence of mind to stop and let the reading happen the level is down which is what I would expect. I am familiar enough with the oil level system to know that the reading is invalid, arises from some bug in the system, but if one is not that familiar he could understandably take the reading as "gospel". Under this scenario then based on the count down timer starting at 5 seconds believe the engine and engine oil was hot enough to warrant just a short count when in fact the count should be not 5 seconds, more like 5, 7 or even more minutes long, which most of the time when I get back in the car after just a few blocks long trip is the case.

The puff of smoke isn't always, and doesn't have to be from the AOS. We will be very watchful with this one when doing the IMSR pre- qualification. That means compression and leak down tests for all the cylinders, as well as a bore scope inspection. We'll also inspect the AOS once we remove it, as the AOS is part of all our preventive packages. The issue probably is the AOS, but we assume nothing, and quantify everything during all procedures.

Quotes source post: 12140559

Any way to bench test the AOS or it's just a visual inspection?

Quotes source post: 12140596

While installed the AOS can be tested using a Manometer. This is something that I go over in my classes, and I give the specifications to the proper values for both a good and bad AOS, at different elevations above sea level. The proper tool to do this will be on the market soon. It, and the process were developed by my co- instructor, Tony Callas. Also, smoke at start up can come from too much fuel dumping into the cylinders at start up, or more often it comes from dirty/ tired injectors that drip fuel when the engine shuts down. Thats the stuff that the other guys don't think about.

Quotes source post: 12140639

Thanks Jake. I was aware of manometer but I'm curious if there's a way to bench test an AOS before installing it. I had one bad out of the box (after it's installed) which I had to swap out in 1k miles.

Quotes source post: 12141361

Yep.. A mighty vac will do it. But it can test good on the bench and still fail in 1k

Quotes source post: 12141410

So I guess you apply full vacuum to the top vent, then seal the bottom side vent and the measure the vacuum at the very bottom AOS outlet to see if you get 4~5" of water vacuum?

Quotes source post: 12141579

I just pull about 20" on the ports and leave it for a few hours per port. The AOS sees around 20" on heavy decel when engine braking.

Quotes source post: 12141890

Thanks Jake and that's interesting. I understand during operation, only the top port sees the 20" Hg vacuum directly from the intake manifold. Then the AOS regulates that vacuum to 4~5" of water and use that pressure to suck the oily crankcase from the lower two ports, one for each bank. The condensed oil just drips out from the very bottom port. Is my understand correct?