996 1999 C4 e-gas issue, not sorted hot start, cutting out at idle and rough idle

Car is a 1999 C4 e-gas. When cold It starts fine, however when hot catches idle briefly, then dips and stalls unless I hold a small amount of throttle. It feels like it’s trying to stabilise but can’t quite maintain idle. I carried out another smoke test and this time found a definite leak at the rear of the intake plenum around the breather connection from the AOS (behind the throttle body / T-piece area). There is also a noticeable amount of oil present in that area. What’s strange is there are no typical AOS failure symptoms: No white smoke No obvious oil consumption Oil cap removes easily at idle However, the leak is clearly significant enough to cause a major idle imbalance and stalling. Access from the top is extremely limited (throttle body and T-piece already removed), so the next step is to go in from underneath and replace the AOS and breather hoses rather than try to chase a single split line. Planned parts to replace: Air Oil Separator (AOS) – 996 107 026 52 Upper breather hose (AOS to intake plenum) – 996 107 145 55 Lower breather hose (AOS to crankcase) – 996 107 147 55 Rear/auxiliary breather hose – 996 106 230 53 Additional small breather line – 996 106 224 51 (if required) Intake manifold gaskets – 996 110 147 03 (if manifold needs to come off) Planning to replace everything in one go rather than risk disturbing brittle hoses and having to go back in again. Has anyone had a similar AOS/breather leak without the usual smoke or oil consumption symptoms, but still severe enough to cause stalling? Will update once it’s apart and confirmed — finally feels like the right area is pinned

Hiya, that does sound like a sensible diagnosis and a solid place to start. An AOS or breather leak does not always have to show as the classic smoke oil consumption failure. If it is leaking around the rear plenum / breather connection, it can still act as a decent vacuum leak and upset idle control, especially when hot. Replacing the AOS and the associated breather hoses together is the correct path rather than disturb old plastic and rubber in that area twice. If the smoke test has clearly shown a leak there, I’d follow that evidence rather than chase another idle theory for now.

When you have it apart, I’d also check the T-piece/plenum area, the throttle body sleeve/seal and the rubber intake boots. They’re all in the same area, and if one old breather hose has split, the other rubber parts nearby are worth checking while access is there. T-piece / intake plenum The central Y/T-shaped intake section behind the throttle body. On a standard car this is the plastic plenum area; aftermarket examples are often alloy IPD-style pieces. Throttle body seal / sleeve The rubber sleeve or sealing ring between the throttle body and the intake/plenum. This is a common place to disturb or mis-seat when the throttle body has been off. Intake boots The rubber couplers/sleeves that join the intake/plenum pipework to the manifold sections. These can split, harden, or sit slightly out of position, giving a vacuum leak.

Thanks great breakdown.

I think you're chasing the wrong thing, or at least something else is playing a part in the issue. With it all assembled, I'd get a can of easy start and go spraying it around all the pipe work. If there's a leak anywhere, the easy start vapour gets sucked in and the engine revs quickly shoot up for a second or so. That's the easy way to find it. Also - are both variocam solenoids operating correctly - if one isn't, your idle won't be too good, if both aren't, you'll have it struggling to idle and cutting out. This needs checking with a decent Porsche diagnostic tool that can operate the solenoids manually. Have you tried running the car in the fixed map by disconnecting the MAF? This puts the ecu in to a fixed program that overrides what different the data given to it from your MAF, etc. Just a few starters as there's plenty of other stuff it could be too.

Quick update on this — it’s looking more and more like the leak is coming from the AOS upper breather pipe. When I had it apart earlier, there was quite a bit of oil sitting in the intake manifold/plenum, which ties in with what I’m seeing now. During the smoke test there’s definitely smoke escaping in that area, but it’s hard to be 100% certain with the throttle body housing still in place. Given how much oil was in there, I’m leaning towards the AOS being the root cause anyway. Feels like it would be a bit pointless not to replace it while I’m this far in, even though it means pulling more bits off and potentially dropping the engine slightly for access. Does that sound about right, or am I missing something obvious?

Yes, that sounds about right. If you’ve got smoke escaping around the AOS upper breather area and oil sitting in the plenum, that’s enough of a clue. It doesn’t need to be the full smoke-screen AOS failure to still cause a vacuum leak and idle issue. I’d quickly rule out the throttle body sleeve, T-piece/plenum joints and intake boots, just to make sure the smoke isn’t tracking from there. But if the leak is at the AOS pipework and it’s all original, I’d replace the AOS and breather hoses while you’re in there. No point disturbing that area twice.

Looking to do this on Friday with a friend who is a mechanic. Watched a few vids, doesn’t seem too terrible. However my local indie has said it’s a nightmare and you will have to drop engine out completely and wants circa £2k to do the job I think dropping the engine by a couple of inches should be enough with access from the top and bottom to get get full access to the desired parts plus removal of a lot of parts. What’s people’s thoughts?

I'm the messenger of bad news i'm afraid, (sorry) there's basically no chance of fitting a new AOS correctly with the engine in the car on a .1 996 , it's a struggle to get it in properly and fully sealed even with the engine out and being able to see whats going on, it's impossible to see if the seal has gone in square which they never do because the correct angle to get it in can't be achieved with it in the car. It pushes the o ring off where it fits into the block, Generally resulting in either an oil leak or an air leak. It's a nightmare. Just getting the old one off is horrendous that's even if the bolts that hold it on still have heads that havn't rusted off. I would drop the engine and box together and take the opportunity to go over everything properly, it will be easier,quicker and your friend may still speak to you.

Quotes source post: 2801301

That is fair assessment by @Hooner . It can be done engine-in, but the issue is not just removing the old AOS, it is getting the new one seated squarely and making sure the O-ring has not rolled or been pushed out. If that happens, you are straight back to an air leak or oil leak. I’d start by making as much access as possible from top and bottom, with a partial drop if needed. If it fights you, the bolts are poor, or you cannot offer the new AOS up cleanly, I’d stop and drop the engine/box rather than force it in blind. Possible engine-in, yes. Sensible at all costs, no.

Parts delivered today, attempting the impossible tomorrow with a friend who is a mechanic.

Haven't read your full thread, but posting to give you some confidence. I have done two of these on 996s and one on a 986. All with the engine in the car. I did my current 996 last year- tips I will give you: Remove the Near side inlet manifold and lower th engine down off the mounts (supported on a jack). With the rear of the car on axle stands, this gave me all the access I needed.

Be very careful you don't drop something in the inlets if you do this, best put some bright coloured cloth /paper towel in them but remember to take it out. As Hooner said, you have to make sure that seal goes in straight . Good Luck

Quotes source post: 2801785

Agree- see the green microfibre cloths in my pic

Good luck. Keep us updated