Sysgen

Here's the issue, randomly my car will go into this mode where I can't press the accelerator, the car want's to die, very similar as if it is not getting air. If I press the clutch I can rev it no problem, but when I release the clutch, press the accelerator and it won't go.

What I noticed is that if I go WOT it instantly wakes up but of course right away I need to release the accelerator then it runs like crap again, I repeat this process until it clears up. Sometimes I press the clutch in, turn the ignition off and then on again, and sometimes that works. These periods can last a good 2-3 minutes, then when it clears up I'm good for a while.

It can do it either with engine cold or hot.

I've cleaned my fuel filter, ordered a new one but not yet received.
Air filter is clean.
I've check the TPS according to Clarks-Garage
I've tested and cleaned my AFM also according to Clarks. <- It looked like new !
I have a fuel pressure gauge in the engine bay and it sits around 40PSI, from memory.
Don't have a spare computer to try.
FPR was changed 2 years ago.

What else can I try?

Sysgen

Forgot, '90 S2 with freshly rebuilt engine. Last summer.

BikePilot

Any black smoke out the tailpipe when it stumbles then clears up? Is the cap/rotor/wires/coil in good shape? Anyway you can monitor fuel pressure while it acts up?

Sysgen

I can't see smoke in my mirrors but it does smell really bad.

Small video from yesterday when I had an issue with a lifter (lifter issue resolved). cap/rotor/wires and coil all seem in very good shape.



No way to monitor fuel pressure while driving. I would have to extend the gauge but I don't have what's needed to do that.

Sysgen

I've read somewhere that when at WOT the AFM readings are ignored, wouldn't this indicate that the AFM is bad even though it's testing ok? I don't know if this is true.

Sysgen

Anyone?

gtroth

For me that behavior was fixed with a new AFM.
I chased that for years. Happened when air was cold and car was warm, but was exactly as you describe. My AFM tested fine too. Can you borrow a known good AFM?

Sysgen

Here's a video of me testing the AFM, it's in french because I was showing it to my brother but you'll get the idea. I fail to see what could be defective if the test is successful, there's not much in this thing!

But I'm willing to try, anyone have a spare AFM for sale?



Regarding my statement that the AFM is bypassed when at WOT, can someone confirm this? Is this true? Where is this documented?

Sysgen

Tried with another AFM and symptom occured again. I guess I need to check the wiring from the AFM to the ECU for something loose.

Sysgen

Found my answer of what happens at WOT, Thanks to TComeau on 944spec.org

The early and late AFM's all work the same with regard to WOT (Wide open throttle). This is controlled by the TPS (throttle position sensor or switch).
The TPS has a two state switch. All it knows, and tells the DME, is:

1. I'm at idle.
OR
2. I'm at WOT.

As soon as the throttle starts to move off idle, the TPS sends a ground signal to the DME (pin 2?) and gives over control of the air fuel ratio to the AFM.
When the throttle gets to the WOT point, the TPS is switched the other direction and sends a ground signal on a different path (pin 3?) to the DME. This knocks the AFM ( and O2 sensor) offline and gives control of spark/fuel back to the DME, which understands the command as " Gimme all you got!". Timing is advanced, fuel is maxed out.
So the DME controls your mixture at idle and WOT, the AFM controls everything else in between.

gtroth

No, AFM doesn't control anything and DME controls everything. DME acts on inputs from the sensors (AFM, TPS, etc).

dave1192

At idle, the engine is trying to pull air through the throttle, but is pretty much blocked by the throttle plate, resulting in a pretty high manifold vacuum. When you open the throttle on a running engine, the existing manifold vacuum pulls additional air though the AFM, and the TPS tells the DME the throttle plate is no longer closed, causing the DME to switch to Part Throttle fuel and timing maps, based on AFM, Ref sensors, temp, etc. If the DME fails to get this either the AFM flow change, it has no idea what to do with adding fuel and changing timing. I'd bet your AFM signal is not making it to the DME. I was able to start and run my 86 NA with the TPS (accidentally) disconnected. It wasn't smooth or pretty, but I did back it out of the garage before I found my error.

Sysgen

Are you saying that the whole text is wrong, or No to the last sentence only? Your No seems to imply that it's all wrong.

I agree the last sentence is to be taken with a grain of salt.

So the DME controls your mixture at idle and WOT, the AFM controls everything else in between.

The AFM does not control anything, it sends a signal (barn door position and temperature) to the DME which controls the fuel delivery, timing, etc.

konakat

This is almost certainly the Throttle Position Switch (TPS). It is attached to the bottom of the throttle body. To test, try disconnecting the cable. The car will not idle as well, but it should make your issue go away.

Lecrok

video shows the internals of the tps.