Generallee

Hi all,
Struggling with a fault on 2000 Boxster 2.7 manual. Starts,runs drives excellent with only 67k miles. When hot, the idle goes from 250rpm to 900 rpm every second. It misfires slightly first for a minute of 2. Have cleaned out the throttle body, changed coils, plugs, MAF, throttle body, upstream lambda sensors etc, etc. Sounds like an air leak, but I cannot find it anywhere - been over the engine with Carb cleaner must be 4 times. Chescked purge valve and cannister , all appears fine. Doesn't throw a code, no codes present. Fuel trims are all over the place when it does it. Seems worse when been for a drive, stop at shops, then get back in - then it will do it, as if the heat is soaking through.. Any ideas? Driving me crazy (er)... :-)

Macster

You say "misfires" yet no CEL nor any error codes? Is this then just your labeling some untoward engine behavior as misfires?

No CEL and no error codes. Any pending codes?

Have you check intake air temperature and coolant temperature when the engine is acting up?

One possible explanation is the AOS (air/oil separator). This fails and an erratic idle is a symptom.

My experience though is the idle isn't as erratic as you report. A 250 RPM to 900 RPM idle swing is huge. You sure about the range? Worst idle swing I've observed from my 2002 Boxster is from around a low of 500 RPMs to around a high of 750 RPMs. This wasn't even during an AOS failure though, it was when the passenger side VarioCam solenoid/actuator was acting up.

This appeared out of the blue -- albeit at some mileage north of 260K miles. No CEL but the idle was up and down. I did find a pending error code which pointed at the VarioCam solenoid/actuator though.

I turned off the engine and after just a few moments off upon restart the symptoms were gone and remained gone as I turned around and drove 30 miles back home.

A shade tree test of the AOS is with the engine running at idle see if you can unscrew then remove the oil filler tube cap. When the AOS goes bad full engine vacuum at idle is supplied to the crankcase and the pressure difference can prevent one from removing the cap.

If you are able to remove the cap this doesn't let the AOS off the hook though.

Another possible test of the AOS is to with the engine running and ideally not running well clamp the hose beween the intake and the AOS shut. (Automotive supply shops sell special clamps to clamp hoses. The Porsche techs have a handful to use to clamp radiator hoses when they do a water pump replacement to avoid having to catch coolant in the radiators and then add this back to system afterwards.)

The idea is to see if you cut off the vacuum to the AOS if the behavior goes away.

Fuel supply/pressure has to be suspect too. While often the fuel pump just up and guits sometimes a bad one can linger between healthy and dead.

It is a bit of work to check fuel supply and pressure and especially pressure while the engine is running and manifesting the symptoms. When I did my own automotive servicing/repairs I was very reluctant to mess with the fuel system. Fortunately the only time I had to was when I replaced an old style engine block mounted fuel pump that was driven from an eccentric on the cam nose to an electric fuel pump mounted at the fuel tank.

Sorry not much to offer in the way of concrete recommendations. While the behavior fits a couple of problems I've encountered over the years the fit is not 100%.

But you have some things to consider. Unless something strikes you as more likely the problem I'd think eliminating the AOS as a possible explanation would be high on your list. 67K miles ain't many miles as AOS's go (I get between 80K and 120K miles from an AOS -- my Boxster is on its 3rd repalcement AOC) but your car is 17 years old and age can play a role in AOS failure.

mikefocke

Crank Position Sensor (CPS) have been known to fail at high temps causing no fault code. Any hot restart problems ever seen?

Generallee

Hi, thanks for the replies. Yes, I have changed the AOS too recently. There are no codes, bearing in mind this has been going on for months and I have been trying all sorts of things. Yes, there are minor misfires first before it goes into this erratic idle state, it certainly isn't running smoothly on all 6 cylinders. If I touch the throttle pedal, it runs fine, drives at all times apart from the idle problem. Nothing I have changed has made a difference. If I switch all the lights on, air con etc it smooths the idle out quite a bit so the 'brain' is sensing load on the motor. My scanner says the inlet air temp sensor is reading correctly,perhaps a couple of degrees higher than ambient, but it is in the engine bay. Always starts fine. Makes no difference when the MAF is unplugged. It all points to an air leak.. only thing I can think is that it is a crack in the inlet manifold but would have thought spraying carb cleaner would find this. Have even used freezer spray to cool components to see if they change.

Macster

Spraying carb cleaner is an old trick to help ID intake leaks. When it works it is almost indistinguishable from magic.

When it works.

The leak might not be at the manifold.

Recently my 2002 Boxster has been tripping a CEL with error code P1128. Long story short I've tried some things but with no luck.

Talked to a Porsche tech the other day about what I've tried and what i had in mind going forward. Not on my list but the tech said what can happen is the intake manifold develops a leak -- each runner I guess is sealed to the head via an o-ring -- or a vacuum hose develops a leak. A reseal of the intake or replacement of the bad hose/hose fitting is the cure.

He said if I want to bring the car in he'll do a "smoke test" to ID where the leak is. If from the manifold the manifold would have to be resealed -- I guess removed and new o-rings fitted. If the leak from a hose or hose fitting of courese the bad hose and fitting would be replaced.

Since I've done about all I can do I will take the car in. The car is booked in for this Thursday. With over 315K miles on the car and the o-rings I guess I've gotten my money's worth out of them.

If you suspect an intake air leak and that has to be pretty high on the list of possible causes maybe it is time for your car's engine to get a smoke test?

Generallee

OK, have finally managed to get the car scanned by a Porsche Tech, found code P1341 which relates to the Bank 1 variator solenoid not working - tried to manually switch using the diagnostics but it didn't work, so going to look at the wiring first (as sometimes it seems fine) then perhaps attempt to change the solenoid (which appear really expensive!!). Anyone got a good parts supplier (UK). I know I will need to fabricate a camshaft holding tool, that much I know, other than that, anything else to watch out for?

Macster

Went through this with my 2002 Boxster. (I touched upon it briefly above when discussing idle speed variation. It was also why I asked you about any pending error codes. A P1341 was a pending code with my Boxster.)

Had the solenoid and actuator replaced by a Porsche tech. The engine remained in the car.

What you have to watch out for is a lot.

First thing to mention is often -- and this proved to be the case with my car -- both the solenoid and actuator are bad. The problem is there is no way to distinguish between the two. The solenoid is a bit easier to change but to test the new solenoid and the old actuator the car must be road tested and for this you have to button everything back up.

If the actuator proves to be bad you have to go through the process of getting to the solenoid then dig deeper to get to and remove the actuator.

The tech that did the work on my car told me it was SOP to replace both the solenoid and actuator because of this.

But for some reason he did not. He didn't even ask me for if he had I would have deferred to his experience and given him the ok to replace both.

After I got the car back the CEL came back and the engine was acting up again. I took the car back and the dealer ate the cost of the labor to replace the actuator. The service manager told me he knew had I been asked I would have ok'd both items be replaced.

If you are doing all the labor and your time is cheap you can elect to replace just the solenoid then button up the camshaft cover replace the exhaust and road test the car.

If the engine behaves that's a win. If not then you have to go back in the engine and replace the actuator.

Besides the above what I know -- 2nd hand -- about this is the camshaft cover bolts should not be reused. They are micro-sealed and if re-used can seep oil. Proper tightening sequence should be followed and a reliable torque wrench used to give these bolts the right torque.

The proper sealant must be used and just as important the right amount must be used. If you want an idea of what the proper amount looks like search out a video or two of new Porsche engines being built. At some stages the tech places the engine on a special stand and a computer driven sealant applicator runs a precise bead of sealant on the surface to be sealed. The bead is not very large in diameter either.

The tech uses threaded rods as guides so the sump plate or the camshaft cover goes on straight and the bead of sealant is not squeezed unevenly one way or the other but squeezed with pressure straight down and flattens out and flows out evenly in both directions. (The machine I mentioned above places the bead equal distances from the inside and outside of the sealing area.)

The camshaft cover contains half the camshaft bearings. The camshaft and head are machined as one unit with no gap between the cover and the head. Might toss in here the proper torque on the camshaft bolts is critical to bring the camshaft cover and head to their exact same mating/fit they had when machined. Too little torque or too much could have the bearings malformed with disastrous results.

Anyhow I mention this no gap to stress that the sealant doesn't form a film like a gasket but gets squeezed down to the point the camshaft cover and the head material come into contact. The sealant used must be able to flow and spread out and not form a layer that prevents the two surfaces from coming direct metal to metal contact.

The sealing is accomplished by the sealant filling the tiny depresssion the camshaft cover and head surface machining operations left so the oil can find no path around the high spots that make metal to metal contact.

Proper sealing of the camshaft cover to the head is critical. There is a lot of oil under there and if any bolt leaks or any spot along the outside leaks the engine will be a mess of oil.

There is also the risk of an interior oil leak. There are holes in the head and cover that match up and these holes are used to route high pressure oil to the camshaft bearings.

These holes are ringed with a bead of sealant. If this sealant leaks there is an internal oil leak. This can result in one or more camshaft bearings receiving less than their full allotment of oil. This is not good.

One may not know this is going on until a camshaft cover or head camshaft bearing fails or a camshaft bearing journal fails.

With the above the tendency might be to really slather on the sealant. This is bad. The extra sealant gets squeezed out and any inside the camshaft cover will become loose and end up in the oil. The stuff is pliable enough it may not be chewed up even if it goes through the scavenge pump in each head. If the pieces are too large down in the oil sump they get sucked up against the oil pump intake tube screen and remain there while the engine is running and these pieces can reduce oil flow to the pump and thus to the engine.

Or excess sealant gets squeezed into one of the cam bearing oil feed holes and the flow of oil to the bearing is restricted.

Oh, when the tech had everything apart -- I have pics -- I asked him about the VarioCam actuator rail guide plastic wear and did this bank's rail guide wear suggest the other side's actuator rails (at least) needed replacing. He said the wear was not bad at all -- the engine had over 250K miles on it at the time -- and with no symptoms of any issues from the other bank he said just wait. That was at least 50K miles ago and so far the other bank is fine. The engine still runs great -- I'm amazed at how well it runs -- even with over 315K miles on it.