LI Steve

OK, so I should have listened to THE VERY FIRST RESPONSE I GOT (Thanks V2Rocket)!

I removed the AFM and checked it using the 9V battery method. It checked out perfect, but the tracks did look more worn at full open than anywhere else.

I decided to bend the arms to change where they move along the track anyway, and I re-tested the AFM - the results were the same.

So I did not think that I fixed anything (in fact, nothing changed from test to test). I put the AFM back in and drove the car - the problem is gone! Tested about 20 full throttle runs - no problem at all. So, I guess that even though the AFM was testing fine, something was up with it. I'm glad I decided to bend the arm even though it tested out fine.

I REALLY WANT TO THANK EVERYONE WHO RESPONDED!

I could never have fixed this without your help. I hope I have the opportunity to help you guys out in the future (or at least buy you a beer).

Regards,

M758

That is good info Weston. I had to re-track my AFM many years ago, but my problems were not like this. Go to know a bad AFM can do different things.

Steve... glad you got things fixed. Worn wiper tracks do odd things. My worn tracks result in a cold part throttle stubble. It went away when warm. Clearly different from your problem, but it appears a worn tracks can create lost of problems.

V2Rocket

yessss got one rightttt

StoogeMoe

I don't think that is correct information about the AFM being in use at WOT. It doesn't make sense. The barn door will be full open so there's no point in using the signal it provides.

I think the problem here is that the TPS isn't sensing the WOT, therefore it IS using the AFM at WOT.

M758

No.. the AFM is use at all times. The throttle position and AFM are not the same things.


Throttle position is what the driver uses to control engine power. The AFM is reading how much air is actually going in the motor. WOT at low RPM is very different from WOT at high rpm as the faster the motor turns the more air get sucked in. The TPS is used to tell the computer when the driver is closed throttle or WOT. This changes the fuel maps, but still needs input from the speed sensor and AFM to measure how much air is actually going in the engine.

Spidey944

Correct, unlike modern TPS' ours is mainly just an ON/OFF switch, you still need to know the volume of air coming itn, THus, the AFM is used all the time.

Makis

That was my initial thought too. But is possible that the ECU still checks AFM to make sure it reads fully open before using the fixed WOT fuel/ignition maps. If the AFM does not show full open the ECU may cut fueling to avoid any possible over-fueling of the engine. But this still does not explain why the problem is isolated to 5400. AFM should show fully open at other rpms at WOT.

One extreme scenario will be that at 5400 the engine resonates is such a way that the spring loaded track in the AFM causes a no read as the slide jumps of the track due to resonance. By altering the loading in the slide the problem suddenly got resolved.

The TPS on my S2 has tree terminals. At full close(idle) middle terminal and one of the terminals show short. Move throttle slightly open I can hear a click and the connection becomes open circuit. At full open (WOT) the middle terminal and the other terminal show short.

Makis

As said the AFM opens fully at WOT at most rpm, so it cannot show how much air goes in. The ECU knows as the flow for given engine design can be calculated at WOT. It then uses fixed ignition/fueling maps to control performance.

DarylJ

If you WOT it at, let's say, idle in 5th gear....I can guarantee you the barn door will not be open all the way. Obviously that's an extreme, but I think it proves the point....you would instantly flood the motor if WOT on the TPS disabled the AFM and just considered it all the way open. I'm not sure there can be such a thing as a fixed ignition/fuel map - you are missing a variable in there to make the car run (air flow to calculate a proper mixture).

Catfood

I dont think V2 is old enough to drink!! However I am 20 and therefore much more responsible than him, you should buy me the beer and all will be well.

Makis

Once the AFM riches maximum (flap open fully) the AFM cannot measure flow any more. So the variable value he reports is now wrong. The ECU must now estimate that variable. For a given engine capacity and air flow restrictions of the system (inlet, exhaust, cams, timing) the ECU can calculate the variable value at any RPM. To do this in practice porsche evaluates the maps in the laboratory and then creates fix ignition maps for the ECU to use when the AFM stops showing the correct variable. This is also the reason that some of the modes we are doing to the engine like better flow exhaust or inlet do not produce the power we expect at WOT and is recommended that we remap the ECU to make use of this modes. If the AFM was able to work out the variable correctly we would not need to get our ECU remaped for best power.

LI Steve

I guess I forgot which forum I was posting on (average age is probably a bit younger here than old guys like me)! How about some chocolate milk?

Anyway, the car has been running great since I moved the wiper arm on the AFM. Again, I am not really sure why this worked, because the voltage through it was the same before and after I made the change, but I guess I shouldn't complain...

DarylJ

How is it wrong? You go through a lengthy explanation after this, but miss a simple point: if the barn door is all the way open, it's all the way open. So you know the volume of air that can pass through it - and that won't change until it starts to close again. Basic physics. How is the variable it reports wrong? Air flow at WOAF (wide open afm flap - yeah, I just made that up) is a constant, known value.

V2Rocket

deal.

tastes better too :P

Makis

Just try to explain again, we may be talking the same thing though.

The variable is a voltage say 1V when the flap is maximum open. The flap can be open fully for two sets of flows f1 and f2 which can be different, but the voltage variable will still stay 1V. The ECU in this case must estimate f1 and f2.