Hesitation on Acceleration and Idle Stalling

[bhoff]

This is going to be a long post because I will describe the symptoms I have and what I have tried so far. This issue has been going on for over a year and I am at a loss as to what to try next.

First starting the car cold, it runs good for about 10-15 seconds. It then starts to miss and stall. Sometimes there is some backfire with the misses. If I keep the rpm's above 2,000 it will usually run ok. When I accelerate in 1st, it will stall out for a fraction of a second and then jerk back to life. It is if the engine just shuts down and then violently comes back. It seems to happen below 2,000 rpm just starting the acceleration. Sometimes it also happens in 2nd if I accelerate slightly hard rounding a corner. Once the engine is up to full operating temp, all of these issues generally go away and the car will idle fine with none of the hesitation during acceleration.

What I have done:
1) Replaced fuel pump and filter
2) Replaced head temp sensor
3) Swapped out the DME for another unit for several weeks
4) Replaced the DME relay twice
5) Checked the sensor plugs for corrosion in the engine compartment, cleaned, and sprayed with contact cleaner
6) Cleaned idle control valve and checked ohm readings
7) Checked air flow sensor ohm readings. I also opened the air meter and checked the resistor wire for any wear. Seems to be ok
7) Checked idle position switch for correct voltage at the plug
8) Checked coil and plug wires. Plugs and wires ok. Replaced distributor cap and rotor.
9) Torqued the intake manifold bolts.
10) Today I replaced both sensors on the flywheel. I really thought this was going to solve the problem.

I have not replaced the oxygen sensor yet. As I understand it, replacing will not have anything to do with solving my issues. I have had the car in two very reputable shops in Seattle and no one has been able to figure it out yet. It was in about a month ago and at first the car was running very poorly. By the time the breakout box was put on, the car was running fine with no issues. I have run out of ideas to check next and am wondering if anyone has either had these issues or has any ideas on what I should check next. Thanks for any help. 15 months if this is getting very frustrating.

[stlrj]

Originally Posted by bhoff

7) Checked air flow sensor ohm readings. I also opened the air meter and checked the resistor wire for any wear. Seems to be ok

You can not use an ohmmeter to test an AFM's potentiometer. It will jump about meaninglessly. You must test it as it is being used. You have to force a voltage UV and measure the output voltage UP. This can be done with the AFM in the car and the ignition on.

Bench testing an AFM. Sophisticated test equipment consisting of a 9 Volt battery, battery connector, and two 0.1 inch female blade connectors (hard to find). Cost, under $2. The output voltage UP must rise smoothly as the door is opened. UP must never decrease as the door is opening not even for a moment. This is called non-monotonicity and can drive your DME into oscillations.

Air Flow Meter (AFM) - from "The 944 Motronic DME" by FR Wilk

[rusnak]

It's a pretty simple system. If it were my car, I'd probably have bought, rented, or borrowed an air fuel meter to see what the system is doing in both open loop and closed loop. You might also want to obtain a fuel pressure gauge. If you don't have the Bentley manual, then you might want to get one.

[bhoff]

Thanks for the replies. I do have the Bentley Manual and that is what I have been using to see if the ohms and volt are correct for the different sensors. I am not sure what an air fuel meter is so have not used one. The issues with the car shutting down for a fraction of a second and then immediately jerking to life does not seem to be fuel related because it happens so fast. It is more like the electrical shuts down and then comes right back. The rough idling could be fuel related. Maybe I have two separate issues I am dealing with here. I do like the idea of testing the air flow meter with a voltage. I read the Bentley Manual wrong and used an ohm reading rather than a voltage change. I will try to check tomorrow and see if it is linear. Thanks for help in giving me a direction to look.

[steely]

I'm not pretending to have a great understanding of the entire system, so I don't have a good set of logical steps to offer beyond some of what you tried already - but I would try to eliminate the O2 sensor as variable.

When I first bought my 87 a couple of yrs ago, the only problem I was told about was a cold starting and inconsistent idle (until warm) issue - granted it sounds like a different problem than yours, but maybe not entirely unlike it either. I was assured by the very reputable mechanic that he tried everything including replacing the ICV (which you did), but to no avail - and I'd have to live with it. I later found the O2 completely disconnected and was told that I didn't need it with the cat removed, and it wouldn't help anyway. For giggles I connected it and bam - no more problem - your mileage may vary. I don't know the theory behind the wrench's reasoning, or why it started working.

Also, while I did not do the plug check of the idle switch, I did find it never toggled the micro switch, so I adjusted that and got some improvement.

Further, and I don't know if this has any bearing, but maybe check your vacuum lines - the part of my vac line called the pipe had a crack in the rubber reducer. No significant change but a fix none-the-less.
hth

[Brando]

I didn't see it on the list, but how is your ignition switch?
Does you car have an alarm? If so, how are the connections between the alarm and fuel pump/ignition circuits?

[T77911S]

check the wiring from temp sensors to the DME and check the pins on the connector and on the DME.
i see you checked for voltage at the throttle position switch (TPS), but did you check the switch to see if it is opening and closing. i cant remember if it is open or closed at idle.

i don t see why O2 snensors would not cause you problems.
get a reading of your exhaust O2. i would put an LM1 on it cold and watch it as it warms up until hot, even with some readings while driving.


you might post the year.

[rusnak]

I agree with Steely and T77.

I used an LM-1 and a cat bypass with a second O2 sensor bung to test my system. I found that the mixture was way rich because the sensor wire in the O2 sensor was broken.

Two shops missed this before I was forced to learn my system and fix it myself.

The problem with using a voltmeter to check the O2 sensor is that you don't know how accurate your sensor is. As a sensor ages, it sends weaker voltage, which causes the dme to read this a leaner than it really is, and therefore you get richer and richer mixture settings as a result.

And the problem with using your DME bung for the testor sensor is that again, you don't know how the dme sensor is working. It becomes apparent that you need two bungs, two sensors (one dme, one wideband test sensor) in order to get to the final result fuel wise.

All of this may be academic if it turns out to be a bad ignition switch, or if the dme harness is somehow bad.

[bhoff]

Thanks for the responses. I do have a bad ignition switch. It has had a small issue from the day I bought the car 5 years ago. I need to move it slightly to get the spedo and radio to work. I do not think this is the issue as my problems go away when the car warms up so it is something affected by either engine or electrical heat. I also need to check the TPS switch for opening and closing.

The car also did have an alarm that was removed by a shop last summer that had experience with that particular alarm. I did not even know it was there the entire time I have had the car. I had the same issues prior to the alarm being removed. I assume it was removed correctly although have not checked myself. It was a real mass of wires taken out and I do not know what components it was hooked up to. The shop recommended removal because they have had issues with this alarm before.

What I can say is that the issues have progressed over the past year and continue to get worse. I am going to start with the AFM today as well as check the TPS switch. I did a visual inspection of the DME plug and wiring and cleaned with contact cleaner. A breakout box was installed on the DME last month so I have to assume that the wiring was ok. The breakout did not show any issues at all because at this point, the engine was very warm and running well. I also pulled the DME out and did an interior visual inspection with magnifing glass at all solder connections on the boards.

I was told by the last shop that the O2 sensor should be replaced because there is a slight variation in the idle (when warm and running ok) but this would not cause the problems I am having. I totally trust his judgement on this as he is the PCA engine forum expert. Thanks again for the opinions. Will post the results of the air flow sensor.

[willtel]

What type of engine and fuel system does your car have? I didn't see that mentioned.

My 930 had similar symptoms and the idle mixture was simply too rich for it to run smoothly at low RPM. When idling at a light it would load up and cause it fall on it's face if you dipped into it before about 3000 rpm. Can you drive around the hesitation at 2000rpm by being very easy on the accelerator?

[techman1]

Sure sounds like the electrical portion of your switch needs replacing.

[bhoff]

My car is an 1987 3.2 engine. I the car will run above 2,000 rpm but is rough, some backfire and dies below 2,000 or when left to normal idle. I also have the cut out on acceleration. This cutout has been happening for longer than the idle problems and they may be unrelated.

I just checked out the AFM and it checked out perfectly, thanks Stlrj for setting me straight on this. The ambient temp sensor also checked out and changed as I warmed the AFM case with my hand. TPS switch also checked out perfectly on both ends and is getting correct voltage. I attempted to move the wiring around behind the starter switch with no changes. Umpluged the O2 sensor and the car ran slightly higher RPM for about 20 seconds and then dropped rpm and started to stumble. It may be an electrical problem with the switch but this should not be affected by temperature. I have run out of things to check. Beginning to think this car is jinxed. After more than a year of this I keep thinking the next thing I check will be the solution but nothing works. I am going to go work on my lawn mower now. Maybe I can get something to work today.

[rusnak]

Why would you do all this work, and refuse to replace the switch and the O2 sensor? I don't care which shop you took the car to, and what they said, this makes no sense to me. Just order the parts. You know you need them, so replace them.

[steely]

I made reference to the idle switch in my previous post, but wasn't clear maybe / it is part of the TPS. Normal function was to close during idle, and mine wasn't closing. You can measure this, but you can also hear and feel it click closed with the engine off if it's working properly.

And btw, the reputable mechanic I mentioned owns his own shop, is award winning in his own right, and is a semi retired engineer for a very large federal organization - I trust him too (he PPI'ed the car) - but a noob fixed it. jus' sayin.

It's been 5 years - I would first fix the known questionable items and then hunt for suspected but unproven thermal issues with the ECU. But if you do continue the hunt, don't wait until it's warmed up.

[bhoff]

I tried to post yesterday with an update but was told it had to be approved by the moderator. Will see if this post gets through.

Thanks stlrj for the information on checking the AFM. Tested it out and it checks perfectly with a linear change in voltage throughout the range. The ambient temp also checked out ok. Also checked the TPS switch for idle, full throttle and correct voltage. It checked out ok. I also unplugged the O2 sensor and it ran at slightly higer RPM for about 20 seconds and then started to stall out like normal. I will change the O2 sensor soon but I do not believer this is the cause of my issues. I have pretty much run out of things to check at this point and will most likely start back through again. I believe the issue with the start switch is unrelated. I attempted to move the wiring around on the switch to see if it would affect anything. I will replace this switch when I can but at this point, I need to look at things that will get me back on the road. I have thrown a lot of money on part replacement that did not need to be done to this point and am very frustrated after more than a year of this and three different shops. I am self employed and my business is down over 80% so need to watch what I spend on parts and shops. I now have the time to check things but no money to replace more parts right now. The next part I purchase will have to get me back running. I really miss driving this car with the top down and the sun out.

[rusnak]

bhoff, if you happen to be in Calif, you can probably find a fellow Porsche owner nearby who can help you in person. If you're near me, we'll have it fixed. Post your location, and you need to get your hands on an LM-1, a cat bypass with two (2) O2 bungs, and a fuel pressure gauge. I can send you a copy of Bosch Fuel Injection and Engine Management for you to borrow if you want. Get a good flashlight too, and make sure the vacuum lines to the fuel pressure regulator are not leaking. Look for cracks in your hoses, including the crankcase vent hoses. These can be wrapped temporarily until you can replace them.

Reading through this again, it sounds like maybe (emphasize maybe) you have a fuel mixture problem that the dme is able to correct once the system goes into closed loop.

[bhoff]

Hey Rusnak, thanks for the offer. I am in Olympia WA. What is an LM-1, not heard of it. Also wondering what the cat bipass and O2 bungs are for. I have not worked with these and have seen nothing in the Bentley about them. Would love to know what direction to go next. I am not familiar with the vacuum lines and will inspect those next. I am also not familiar with what you are referring to by closed loop. Is this where the O2 sensor is working correctly to adjust fuel mixture by monitoring oxygen levels in the exhaust? I do plan on replacing the O2 sensor asap. I just noticed the tail pipe seems to have more carbon than I noted before.

[stlrj]

Sounds like an ignition problem. How did you check the plug wires? Visual inspection? Were you able to apply 15,000 volts to each wire to see if they shorted or a voltmeter to test for continuity?

Also, you might remove each cable and unscrew all the connectors to check for corrosion.

Cheers,

Joe

[rusnak]

Actually, after thinking about it some more, you probably can just skip the cat bypass since your problem occurs when the car is cold.

An LM-1 by Innovate is an Air/ Fuel meter that uses a Bosch wideband O2 sensor to measure and display the air fuel ratio expressed in % Lambda, which is 14.7 parts air to 1 part fuel in terms of volume (not mass).

Your system is designed to keep your air fuel ratio at around .97 to 1.02 lambda except when the throttle is wide open or the engine is under load, or at startup, coasting off throttle, etc etc.

The O2 sensor needs to be sufficiently hot in order to be able to work properly. Because of this, the dme will not use the O2 sensor until it has gotten hot. Until then, and at startup, the dme will time the injector pulses from a pre-determined onboard map, rather than through sensing the air fuel mixture (open loop). The loop between the sensor and the dme is closed when the two begin to work together, thus "closed loop".

The reason to change your O2 sensor is that as sensors age, they send weaker signals, tricking the dme to run your engine, or time the injector pulses shorter, in a rich setting. You won't know this unless you have an air fuel meter.

You can also check "open loop" or pre-sensor running by disconnecting the O2 sensor wire (single black one), and inserting the LM-1 sensor in place of the dme sensor, and reading the air fuel ratio. You want to see around 14.2 or so, or slightly rich. If its much more rich than that, then you want to adjust the bypass screw in the AFM (air flow meter, or barndoor box thinngy next to the air box).

If you can't obtain an LM-1 but can rent a CO meter, then look for a chart that shows CO% and % lambda. I think Scarceller (sp?) has one. Go ahead and use the CO meter if you can find the chart.

My guess is that your fuel pressure regulator may be either bad or not connected via vacuum to the intake, or the connectors are bad, or plugged. Disconnect the fuel pressure regulator tube while the engine is running. Confirm that there is vacuum. Re-connect and look for vacuum leaks. If you can get a fuel pressure gauge, then use it to check your fuel pressure. Fuel pressure affects air fuel ratios, expecially in open loop.

[rusnak]

Dangit, forgot the attachment.