pks

I have a 1982 928 Porsche that stalls when the air intake mass flow level opens and I would highly appreciate any suggestion on what I could try to fix this.

I had to take the engine out to do some repairs. The engine is now back in and the car runs great until the mass flow sensor is activated. When I put the entire air intake on the side with the mass flow sensor connected so the air doesn't have to go through the air intake the car runs fine. However, if I manually push the air intake level down the car stalls immediately even though it is not attached to the engine and no air is going through the air intake so I believe the mass flow sensor gives a wrong signal to make the car stall.
The only electrical I touched when I took the engine out was the 14 pin connector located in the front on the right hand side under the hood. The connector came apart and although I put it back together I'm not completely sure if it is put together correctly since it seems that many of the wire colors do not match up. It was only the upper side of the connector that came apart. I would appreciate any suggestions.

Peter

WallyP

According to the wiring diagrams, the air flow sensor wires do not go thru a disconnect. There are five wires that go straight from the sensor to the control unit:
green/red - term 7 on both ends
green - 8 on both ends
blue/green - 9 on both ends
yellow/red - 6 on both ends
green/black - 27 on both ends

Incendier

Actually, it should stall when you push the flap down when not attached. The flap is designed to be deflected by a greater flow of air; if you're pushing it down, the L-Jet computer is getting the signal that there is more air coming in and is adjusting the fuel accordingly. Without more air actually coming in, you're running dead rich while the flap is depressed. I can personally verify that this causes an immediate stall

However, this doesn't explain why the car doesn't run with the AFM attached. Having just dealt with a bad connection to the AFM, I would check the 5-pin connector as well as check the integrity of the wires that attach to the plug. They take a lot of flex over the year and mine were trashed.

Having said that, I wouldn't immediately indict the AFM based on this info. There's a lot of other things that could cause a stall on throttle open which are not AFM related - if the idle contact remains closed even at off-idle, fuel problems, etc.

FWIW, the connectors at the front 14 pin are as follows:

1: BL, to Alternator B+
2: BL/BK: Oil pressure sender
3: BL/YE: Coolant temp sender for gauge
4: BL/WT: Oil pressure sender
5: BL/GN: Coolant temp sender for gauge
6: BL/BR: Engine oil level sender
7: RE/WT (2 wires) One each to the cold start valve and the auxillary air valve.
8: Open
9: BK: A/C compressor switch
10:BK/YE: Starter
11-13: Open
14: YE (2): Lg = starter to fuse panel. Sm = To cold start valve

Good luck.

pks

Thanks for the responses. That means the problem should be somewhere between the AF connector and the control unit. I checked the 5-pin connector and measured the ohms on the AF sensor and they both seem ok. I haven't checked the wires yet. If any of you would take a guess, what do you think would be the most possible problem, the control unit, a ground, a short or something else. Do you know what parts the AF sensor steers.

Peter

Incendier

Ok, you got me really interested in this one. It sounds very similar to what I was fighting with a while ago. I'll contribute whatever experience has taught.

Could you describe in sequential detail exactly what the symptoms are? When you start the car with the AFM attached (and bolted - can make a vacuum seal difference), does it run and idle smoothly? Does it kill on any throttle application that would open the AFM flap? Will it drive at all?

The AFM sends signal to the brain on flap deflection (from which the brain deduces intake air flow) and air intake temperature (the temp sensor I is the little **** inside the throat) as well as controlling idle air/fuel mixture via a set screw in the AFM base. Basically, as I understand it, this is all the info the brain needs besides the engine temp and ignition (RPM) signal (and 02 sensor for fine adjustment) to determine injector pulse. These inputs are overridden when either the idle contact or the WOT contact are engaged, and the brain then goes to a preset map.

My experience has been that the L-Jet brain is probably the last thing to suspect (not as fragile as our LH-Jet brothers). The harness is highly suspect IF it is a signal problem, but my last AFM ohm'd out perfectly at static settings and still was screwing things up - turns out that the resistance did not sweep smoothly with the flap deflection but would "step" between random resistances throughout.

There are other places to go with this, but I'm interested in your answers first.

2V4V

Peter,

The easiest way to test the airflow meter would be to swap it into a car that is known to be running well and see what happens.

You could also swap in a new/known good used one into your car and see if that works. The only danger with that plan is the (tiny, but real) possibility of a short inherent to your electricals that will kill the replacement unit too. Unlikely, but can happen.

All that having been said, you may be having the problem that James described with his air flow meter. If you are pretty certain this is where your problem lies, you can open the top of the box and clean the wiper/coil assembly with electrical contact cleaner - I would not use any friction material, except perhaps a Q-Tip.

This procedure is not officially sanctioned, but it has been around for about as long as L-Jet, and has worked for me. Be careful not to damage any of the wires on the coil. Don't overtorque the lid when you close it back up, it could warp.

HTH,

Greg

pks

I appreciate your responses. When the AFM is attached and bolted the car does not start. When I turn the engine it sounds like it is about to start but when the car sucks in air and the flap opens it kills the engine.

I haven't had a chance to swap in a different AFM. It is possible that I could have the problem that the AFM "steps" instead of sweeping smoothly with the flap deflection. However, when I have the air intake on the side so the air doesn't have to go through the air intake the car runs fine and as I slowly push down the flap, the car slowly runs rougher and rougher until it stalls.

As James informed me, the car should stall when the flap is pushed down but it seems that the car dies in a smooth pattern as the flap is slowly pushed down. I don't know if this is an indication that the AFM resistance sweep smoothly with the flap deflection but it does not seem like it is working in "steps." However, if you still feel this could be a problem I will try to clean the wiper/coil assembly like Greg described.

Peter

Incendier

Excellent info. Here's what this is telling me, but I'll put in a healthy caveat here that there are many more people on this list that know more than I do.

I have wrestled with this issue recently, and am now the proud owner of a well-running 928. So, here goes...

Does the intake throat smell like gas if you try to start it and then pull the meter? Is a pulled plug wet with fuel after you try to start it?

My car starts and runs with the AFM attached, and runs very poorly when it is removed. Deflecting the flap when the AFM is removed causes the engine to die immediately once it's moved beyond a certain point.

Your car won't start with the AFM attached but will run with it off, and deflection causes a slow death.

What's in common here? Air. My car runs crappy with the AFM removed because I have significantly decreased the restriction in the air intake; it leans out dramatically. If I move the AFM flap slightly, the fuel metering catches up with the air coming in, and it runs great. Conversely, yours runs better with the airflow restriction removed.

This indicates to me that your car is running (or not) very rich as a general condition. By removing the AFM, you are essentially adjusting the mixture by adding a ton more air. The slow death when the flap is depressed is the brain adding even more fuel until you finally exceed the intake air volume, and it dies.

That said, what's wrong? Dunno, but here's some shots - cold start valve and/or temp sensor II. If the valve is leaking or perpetually on, you're getting a lot of fuel that the brain is not accounting for. I would disconnect the plug on the back of the cold start valve and give it a shot; if all is OK, check Temp II and/or wiring. If no joy, you may have to pull the valve and check for leakage when it is attached to the system and the fuel lines are pressurized.

Second shot - AFM is sending perpetual "cold air" info to the L-Jet brain. The temp sensor is suspect, although you did say that it ohm'd out perfectly.

Finally - badly leaking injector(s)? WOT switch stuck in the engaged position (resistance check at the brain terminals - should be infinite resistance at idle)? Or, the dreaded brain failure.

Oh, and BTW - if rich is the case, go ahead and replace the plugs and the O2 sensor. My extended rich running destroyed all of the above. And cross your fingers on your cat. <img src="graemlins/oops.gif" border="0" alt="[oops]" />

Good luck; keep us updated.

pks

Thanks James!

I appreciate all your suggestions. I will try it this weekend and let you know. Thanks again!

Peter

Incendier

*bump*

Peter, what happened? Any joy?

pks

Hi James, the car has a very faint smell of gas and I'm not sure how much above normal the smell is. However, if the car is running very rich I believe that I would clearly smell the gas but this is just an assumption from a novice. The plugs are dry after trying to start the car with the AFM on, however they were fully covered with a fine layer of dry black soot/dust.

I did disconnect the plug on the back of the cold start valve without any success. The valve did ohm out correct however, I have not pulled it to check for leakage yet. The injector(s) seem fine although I just did a random check. Could you explain how I check if the WOT switch is stuck in the engaged position.

I think you are on the right track, we just have be a little lucky since it could be so many probable causes or a or a combination of things. Thanks again James, I'll keep on trying the probable causes we have discussed and keep you up to date. Please let me know if you think of something else.

Peter

Incendier

Yeah, we do need a bit of luck. I hope we're on the right track - this problem took me hours to fix on my car.

If it were running rich enough to not start at all(pretty extreme), I would think you'd smell it. This no-start sure seems electrical...(oh, did you check your engine grounds?)

Been thinking about this - you'd have had to take out the FI wiring harness when you pulled the engine. This may sound ridiculous, but it happened to me - a break in the FI harness will cause a no-start, and (for me) simply grabbing the harness would restore the connection. Possible that moving the AFM is doing the same thing on your car, i.e. it's not the removal of the AFM that's allowing the car to run, but the movement of the harness? Might be worth it to get an accomplice and check.

I don't think the WOT switch would prevent starting, but might as well check anyway.
The easy way (check both the idle and WOT switches) - they give a faint audible click when engaged. Manually move the throttle by the switches and check when you get the click for both of them. There is a spec in the manual for when the WOT switch is to engage (% of travel) - I don't remember what it is. The idle switch should engage when the throttle is closed and release on light throttle application.

The switches adjust by two hard-to-reach screws on the back. When the screws are loosened, the switches can be rotated to engage earlier or later.

The RIGHT way is to first check the switches at the switch (they should show infinite resistance when open and continuity when closed). Simple VOM stuff. Then, just to rule everything out, the hard part - pull the multi-pin at the brain and check the pin versus ground. I extended the VOM leads and tripped the switches with my finger to check it, VOM propped up to view.

I'll check the manual and let you know which pin is for each one.