pete95zhn

Ok, if all other things remain the same, what might the reason for lean non-boost A/FR's after exhaust change?? ie where's the leak?

Facts:
-from OEM to aftermarket exhaust
-new crush rings
-alignment OK
-same lambda sensors
-according to Durametric both banks show about lambda 1.2-1.3 ( left 0.06 leaner than right ) when the engine is not on boost, before it was steady 1.0
-boosted lambda values are OK, around 0.8
-no fault codes
-low rpm/ low throttle driving jumpy and sometimes it seems to shoot in exhaust when coasting

I really don't want to change back to std exhaust just for testing...

One more thing, not related to this issue...how much should temperature rise be when engine is not loaded, ie intake air T vs OAT?

Macster

Last thing touched has to be suspect and that's the exhaust. An exhaust leak can account for at least some of the behavior.

I assume you did not touch the intake and that it was ok to start with?

Did you check the straightness of the new exhaust system manifold, the portion that mates up to the heat, that bolts to the head? I used a Snap-On precision straight edge or a flat surface like a milling machine table or a flat ground work/assembly table or even a granite surface plate to check that a manifold was flat and would seat evenly against the head before bolting the thing to the engine.

Sometimes these are bent, warped, and will not seat well even if the manifold bolts are tightened properly. And do not overtighten these bolts. You do not want to strip a manifold bolt in trying to clamp a warped maniifold to the engine.

Intake air temp vs. outside/ambient temp varies, I'm sure, but I have little real time data from my Turbo.

I do know from the few times I've looked at this running down the road the intake temp can be pretty close to ambient. The ambient air temp sensor is located at a particularly cool location while the engine air intake is also located at a pretty cool location.

The intake air temp is located I believe in the MAF and of course the MAF is subject to engine heat that soaks through the intake. But at steady state running the Turbo's intake air temp stays pretty cool.

OTOH, my Boxster runs at least 12F to 15F (on a good day) hotter at the intake than ambient but that is because the Boxster's engine air intake is located on the side and far enough to the rear of the car that it picks up the hotter air that comes from under the car courtesy of the radiators.

Now both cars will see the intake air temp rise when moving in stop/go traffic and of course after being restarted shortly after being shut off hot. This arises mainly from the engine compartment heat heating the intake manifold which of course heats the incoming air in the intake manifold.

What kind of differences are you seeing?

Sincerely,

Macster.

Kevin

Pete;

Have you replaced your MAF yet?

Did you drop a O2 sensor?

We need to start replacing the O2 sensors and MAF's..

Macster the temp sensor is actually mounted on the top of your "Y" pipe. The sensor is actually a dual sensor coupled with the MAP sensor.

pete95zhn

I will start by switching back to OEM exhaust although I don't fancy the idea. It'll take a coupleof hours, I just need to get it cooled down. If A/FR's resume back to normal, the fault is in flanges.
No, I have not dropped O2 sensors, they're quite new ( 3k mls ).
MAF is a brand new Hitachi. Blow-through, programmed accordingly.

According to my latest observations it's actually running rich all the time, ie. there's most likely a leak messing sensors...it goes pig rich ( like below 10:1 ) on boost...had an oppoturnity to hold pedal to metal slightly longer.

pete95zhn

Found two warped flanges, about 0.4mm each. I'll get them machined tomorrow.

pete95zhn

Flanges straightened, new pre-cat lambdasensors in order...

pete95zhn

Just to make sure it's not related to parts or installation, I have so far
-checked all vacuum connections and intake bellows twice
-changed exhaust sealing rings and checked flanges ( but not made a pressure test )
-changed the MAF
-changed exhaust manifolds
-changed new pre-cat lambda sensors
-changed the fuel filter
-checked fuel pump and fuel lines

NO CELs whatsoever.

No affect to lambda values, still 1.2 - 1.3 at idle. Tomorrow I will install fuel pressure gauge.
IMO there's left only three causes to the problem:
-failed FPR ( will know that tomorrow )
-something in the tune
-leak between the MAF ( blow-through ) and heads

I have Durametric logs with the MAF in OEM position and blow-through, housing size the same in both cases. When the MAF was at OEM position ( in February ) there were no issues. Although the tune was already made for blow-through. Since then lambdas have gone worse gradually. It culminated with bad driveability when I installed bigger Y-pipe and TB and repositioned the MAF.

From Durametric logs it's clearly visible that lambdas start to deteriorate few minutes after engine start, independent of engine's temperature at start, but in all logs injector time and MAF voltage are more or less the same ( even in February's logs ). What puzzles me is that the Motronic does not try to enrichen the mixture by lenghtening injector timing!? It's just completely happy with the situation...

I have done everything I can and start to be slightly desperate...summer is so short and I'd like to drive the car. Any ideas?

PS. I do have asked the questions from the tuner too, but just tonight.

Kevin

Post a screenshot of your idle logs. Look at when the airpump shuts off. I'm betting that you have a intake leak. Have you pressurized your intake yet? That's the first place to check.

pete95zhn

OK, will do intake pressure test tomorrow. BTW, no CELs or fault codes whatsoever. Air pump is deleted, hose blocked.



1.Time
2.RPM
3.Intake air temperature
4.Ignition angle
5.Injection time
6.Oxygen sensing, bank 1 Lambda Value
7.Oxygen sensing, bank 2 Lambda Value
8.Hot film MAF
9.Boost pressure of sensor
10.Setpoint boost pressure
11.Boost pressure control P/D factor
12.Corr. factor for BPC with knock control

pete95zhn

No leaks @ 2 bar between the TB and heads, some hissing via bank 2 exhaust. I'm doing this with ghetto equipment...have to confirm that sound when I get an extra pair of hands to help.

Fuel pressure 5.2 bar if pump is forced on via Durametric, at idle about 4.5 bar. About because only pressure gauge small enough I could find in this hurry is not fluid dampened. OEM Bosch 5 bar FPR.

What next?

Mikelly

Scratch that... I read the build specs on renntrack... you need the 5bar FPR...

Mike

pete95zhn

Some speculation and thinking out loud follows... While it seems that there's no mechanical problems with the car anymore ( few small vacuum leaks fixed and cracked headers changed ) and yet the mixture is as it was, something else is wrong. Had a walk with the dog and thought what I've done and how it has behaved -car, not the dog- I remembered that first time I had big issues when I tried bigger TB and appropriate T-piece. It was with OEM Y-pipe and MAF ( already Hitachi ) at original place. I removed them, until the new Y-pipe is finished, suspecting a leak . This makes me think that there's a fault in the program, it does not somehow take in count bigger TB.

I have to get familiar with MAF operation too. If the voltage is related to velocity of air passing through, bigger TB might cause issues. Same volume will go through it at slower velocity and increased pressure. But, as I said, I do not yet really know how it works.

pete95zhn

I've been double and triple checking everything and I'm 98% sure that there's no leak anywhere. After pressure test I made few more loggings and for the last one I added TB parameters and slightly revved the engine.

It seems that Injection time does not react to increased engine speed or throttle plate angle at all. Still start and warmup enrichments work ( injection times are set accordingly ) and at that area lambda values are correct.

So is my Motronic toast or is there a wiring difference between OEM 996 turbo TB and Cayenne turbo's TB which I currently use. ( RMK: No fault codes!! )

A: Time
B: RPM
C: Intake air temperature
D: Ignition angle
E: Injection time
F: Oxygen sensing, bank 1 Lambda Value
G: Oxygen sensing, bank 2 Lambda Value
H: Spec. throttle plate angle
I: Actual value, throttle
J: Hot film MAF

Macster

No clear evidence the engine controller is toast.

Since you are running a MAF from a different model Porsche I'd say that's got to be suspect #1.

Would it make sense to disconnect the MAF and see if things improve?

Or try a MAF intended for the engine?

Sincerely,

Macster

pete95zhn

Back to square one..I took off the exhaust again and found definitive leak marks from flanges, same place on both of them, the upper quarter. It might be so that the flange warps when it gets hot and returns to closer to straight when it cools down. That would explain the slowly deteriorating A/FR when starting from cold.
They're built like exhaust manifold flanges, two 5mm flanges together, short tackwelds outside, seamless inside. Maybe the construction allows warping, OEM ones seem to be 10mm cast ones. I try to find out some solution tomorrow.

BTW, MAF disconnection had no effect, already did that. And the box is programmed for Hitachi.