986 won’t idle right

Oye, haven't been here in a while. 
 

A broken car drags me back. My 986 Boxster is giving me some issues. AOS went out at the beginning of the month. Smoke like I've never seen. 
I replaced that and cleaned everything out including the throttle body and MAF. Once the oil had finished burning out it started to run much better but not perfect. I had a rough idle, hesitation/misfire feeling in the midrpm and two codes for lean running. I replaced the precat O2 sensors and that cleared out those problems. 
 

however. Now it won't idle correctly. With the AC on it's at 800rpm which is about right. It seems like it's struggling a little which is odd. But it's stable and mostly smooth. 

When I turn the AC off the RPMs jump up to 1000-1200 and stay there. They fluctuate slightly going from 900 to 1200 or so over the course of 20-30 seconds. If I put the car in neutral while moving it does the same thing until about 20mph when the revs climb to 1500-1800rpm and then settle back to 1000-1200 by the time I'm stopped. 
 

I've given the car over 100 miles of mixed driving hoping the computer would relearn but it doesn't seem to be doing that. I have reprogrammed the throttle body and double checked for vacuum leaks to no avail. 
 

Watching the live data I can see the throttle body open when the revs increase and the MAF sees more flow. Turn on the AC and the throttle closes slightly, the MAF flow drops and the idle settles down. So it looks like the ecu is telling the throttle to open and idle that high but I cannot understand why it's doing that. 
 

I'm getting pretty frustrated and I'm close to taking it to my local Indy German shop to have them look at it. I don't want to spend that money though so I'm here first. 
 

any ideas? I'm all ears. 

Hopefully AAZCD will be here soon.  He's the one that should know.  Sorry but I don't speak Porsche.

Did you change the plugs after the AOS failure?

Do you have any way of verifying the MAF signal is correct? I know you said it seemed to be responding appropriately but in my experience they're incredibly fragile and cleaning them doesn't usually work.

I did not change the plugs. Runs great at high load. Revs smooth. I thought about them but left them alone for that reason. 
 

I did run half a tank of injector cleaner through the car. 
 

Not sure how I would verify the signal. The MAF failed shortly after I bought the car. Long term fuel trims went out of whack over a period of a few days. I replaced it with a OE Bosch unit and it's been happy since. 
 

The MAF comes up a lot in AOS failure threads as if somehow oil would travel backwards up the intake tube to the sensor. Doesn't quite make sense to me. 
 

Oh I did replace the air filter at the same time and found a few leaves in the air box. I cleaned those out and made sure the sensor and mesh in front of it were clear as well. I'm pretty confident the sensor isn't the problem due to its age and the nature of the issue. 

Any codes now, no CEL?  Two thoughts of potential vacuum leaks.

The vacuum leak that I hate the most is a tube that connects the AOS to the PCV valve. That tube is brittle and runs under the intake and oil filler all the way across the top of the engine from the outboard side of the AOS to the front left side of the engine. It's the one that was difficult to pinch the connector to remove from the AOS. Check to make sure that the tubing connected to the AOS is fully seated. It's hard to visually inspect it for cracks, but you can smoke test it.

I'm not sure what model year you have and the vacuum tubing varies a little. The little tubing used for the Secondary Air Injection (SAI), flapper valve, Fuel pressure regulator (Pre 2002), and Tiptronic transmission cooling gets brittle enough that it will crack if your fingers get within an inch of it. May or may not throw a CEL for a while. Take a look at these threads and make sure your tubing is good.

https://986forum.com/forums/boxster-general-discussions/64576-vacuum-line-clarification.html

https://986forum.com/forums/performance-technical-chat/62979-vacuum-confusion-solved.html

https://986forum.com/forums/boxster-general-discussions/66666-vacuum-diagram.html

The hoses do deteriorate. We did a video a while back on 986 maintenence.
Check this out:

https://www.youtube.com/watch?v=Zn35O80EwHI

Look at your short term fuel trims and see if it's adding fuel, indicating a vacuum leak?

No CEL, no pending codes. 
 

for reference, it's a 01 base model with a 5 speed. 
 

I had suspected that tube as well but surprisingly everything under there seemed supple still. I sprayed around with carb cleaner and didn't get any revving. I'll take a another look at that tube though. 

i double checked the connections and all are good. I got audible clicks for all of the quick connects. I haven't had a chance to double check the bellows under the AOS but they were on true when I put it together. 
 

The small vacuum lines running from the reservoir to various locations including the resonance tube flapper are all seated. I checked those carefully. The T fittings are soft with no signs of cracking. Again no response to carb cleaner. 
 

Short term fuel trims do seem to be high. With the AC off they are in the mid teens I think. With the AC on they drop. Weird right?

My current theory/concern is the AOS is bad. It's new OE Porsche but rumor has it they aren't made like they used to be. If the AOS is allowing too much air to get through it could be causing a high idle without a CEL right? 
It's strange that it seems fine with the AC on. 
 

I'll take a look at those posts and see. Thanks everyone. 

Dang. Might have found the problem. I think my vacuum house routing is incorrect. I'm working from memory (the car is home for obvious reasons and I'm at work) but I'm pretty sure both crossover rubber donuts with the vacuum ports are on the rear and only one of them has a vacuum line attached. I remember seeing the bump thing where the hole would be and it being sealed off. But I need to double check that. 
 

and I know that I plugged in the vacuum line with the check valve into the resonance flapper. I swear that's how it came off. 
 

I have some issues to sort out then. I'll report back either way. Thank you!!

This is your correct routing then. (Copied from my old post on the 986 Forum.)

Notes:
-This picture shows what connections have to be made, the actual order of connections to the Y fittings (I) and (G) are not critical.
[The Tiptronic cars will have a 4-way connection at (I) or (G) with a line running to a changeover valve on the transmission.]

It looks like a mess of tubes and parts, but in function it is fairly simple.
This shows 3 vacuum powered components: SAI, Resonance Flapper, and Fuel Pressure Regulator. Brake booster vacuum system is not shown.
SAI: Vacuum source -> electric change-over valve (E) -> change-over valve (D).
Resonance Flapper: Vacuum source -> electric change-over valve (J) -> vacuum unit (K).
Fuel Pressure Regulator: Vacuum source (B) -> Pressure regulator (C) -> injector valves (A).

Rubber sleeves (L) and (B) are the source of vacuum from the intake. (H) is a check valve. (F) is the vacuum reservoir.

Not sure if it matters, but is the "L" boot on the crossover attached to the throttle body?

spandak said:

Not sure if it matters, but is the "L" boot on the crossover attached to the throttle body?

Yes, throttle body crossover. It shouldn't really matter which. Vacuum should be the same in either position.

It was a vacuum leak and it was because of my bone-headedness. The crossover boot with the vacuum port that I thought was sealed off of course wasnt. Should have double checked. 
 

I got the hoses mixed up too. I didn't see the hardline coming off of the solenoid for the resonance flap and the soft line that connects to it came off with the crossover. I ended up connecting the soft line to the check valve thinking that was it. That left the solenoid hose and crossover boot open. Whoops. 
 

Idles good now! And I'm happy to have my car back.  Thanks for the help everyone and particularly Jon. Your the 986 guru for a reason. 
 

now to chase down the front end clunk. But that's for another day. 

Hey Jon... Just found this post while doing research and hoping you'll see this message since the thread above is over a year old.

I own a 2002 base Boxster. I bought it when it was two years old with 20K miles. Just recently I've discovered that the vacuum line is disconnected from the chamber (item F in your pic above). And the hose is plugged with a golf tee(!). Obviously, the prior owner did this. However, and here's the interesting and/or cringy part... I've driven the car this way for nearly 20 years and 50K miles with no problems whatsoever (no CEL, rough performance and emissions problems). I will say I've always thought the exhaust smells a bit rich, but again... no rough idling any other issues. Spark plugs haven't been changed in a while, but I don't recall them being fowled when changed years ago.

SO... why the hell did they do this???! It's especially strange since the car would have still been under warranty at the time I purchased it.

Any ideas?  And I'm wondering if I should reconnect the hose or just let it be.

Many thanks,

-Dan

Bumping for visibility. That's an interesting one. 

If you've got a Mityvac you could try pulling some vacuum on the canister and see if it was disconnected due to a crack that made it a vacuum leak. Or reconnect that line and see if it runs like it has a vacuum leak...

It kinda looks like removing that and plugging it wouldn't change what received vacuum, but the signal would be less smooth, and there wouldn't be a reserve to run an actuator if the instantaneous vacuum wasn't sufficient. Not sure whether secondary air injection or resonance flappers would present as actual running problems, though I'm a little surprised the ECU wouldn't notice things not acting at the anticipated times and throw a CEL.

If that happened under warranty, I assume it was an owner who was more concerned with not having to deal with taking it in than with getting it right?

Hopefully someone with more specific knowledge drops in soon.

My guess is that the vacuum reservoir has a leak. Every year a few people on the 986 Forum post about finding a leak there, usually at the bottom seam. With the golf tee there probably isn't a vacuum leak, but at times when the vacuum reserve is needed it's not there.

The '02 and later 986s don't use a fuel pressure regulator on the rail. Your secondary air injection and the flapper valve on the intake will use that vacuum (and a Tiptronic will use it to flow coolant to the transmission).

I'd say connect it to the reservoir and see if your car runs worse, or just check if the reservoir holds a vacuum.

Hey guys... apologies for not responding sooner. I was hosting out of town guests for the last week and not online much.

Thanks for your responses.  I was wondering too if the reservoir has a leak and so they used the golf tee instead, but it sure is a head scratcher considering the car was still under warranty.

The only other odd thing my car has had since I acquired it is that the oil level indicator on the dash always shows completely full. I've always assumed that's a pressure sensor replacement, and too much hassle/cost for the value when I can (and have) just used the dipstick instead. 

There isn't any way this reservoir issue could be related, could it? I've been trying to research where the lines go from the reservoir, but it isn't easy.

Thanks again.

In reply to dfvera :

The vacuum system has no effect on the oil level reading. These cars use a sensor near the middle if the engine,  slightly under the left intake and it's entirely electrical.

On an '02 with manual transmission, the vacuum hose with the tee plug only operates the secondary air injection and the intake resonance flapper valve.