1987 911 3.2 Motronic Question(s)

[johnman001]

I've been fighting a running issue with a 3.2 Motronic engine for some time now and have been doing a bunch of testing, but not definitively finding my problem. I even have gone so far as swapping out the DME and AFM for known good units just to have the same symptoms.

I've built a breakout box and have tested everything that I know to test, but today I was fiddling around and I tested something that I thought wasn't actually required.

I tested my altitude correction sensor (the round sensor mounted near the DME relay and the DME Unit itself with a two pin connector). In any event, when testing the sensor, it shows closed circuit. My manual says that the contact is supposed to close at altitudes above 1000 meters above sea level. Being in the midwest, I doubt I will need this device unless I drive to Colorado or in other mountains.

Am I correct in believing that this device should be open circuit or infinite Ohms unless I'm at 1000 meters above sea level?

My other 3.2 cars measure open circuit, so I think I'm correct, but wanted a second opinion.


My other question would be, when does this sensor come into play?

I figured out at the last track event that the car will run ok at part throttle (I'm assuming up until the full load switch (Throttle Position Sensor) contact closes, because if I lifted and ran at part throttle, the car did pull strong to a point, but not like full throttle.

Because of the way the car was running, I was suspecting the cylinder head temp sensor, but perhaps a shorted altitude correction signal would throw off the air fuel mixture as well.

Anyone else run into a shorted Altitude Correction Box? Do folks replace these or just unplug them?

Thanks in advance for your suggestions.

[johnman001]

So I took the Altitude Correction Sensor apart. Inside is a sealed expansion diaphragm, a spring loaded plunger and a snap action micro switch (Omron brand). Upon investigating it looks like the housing was slightly crushed at some point and the micro switch was being held in the closed position. I'm guessing most of the time I will be operating the car below the threshold where this sensor will be needed (1000 meters above sea level), so I will either bend the housing back so the switch isn't made or simply leave this unplugged.

This sensor had been mounted under the passenger seat (914-6 application), and it looks like the seat may have been pressing down on the can. By doing this, I probably caused my own problem over the last 30 years the engine has been in the track car.

If I do put this back in, or I find a replacement for my damaged sensor, I'll probably mount it so that nothing would potentially push down on the housing because now I know how it actually works.

Does anyone know if this Altitude Correction Box leans out the mixture 100% of the time, or only when the O2 sensor is not de-activated (when the Wide Open Throttle) contact is made?

I suppose it is quite possible that at part throttle, the O2 sensor was keeping the mixture close, but at WOT, when the O2 Sensor is taken out of the loop, the car would lean out and lose power. It makes sense if the seat pressed down on the sensor during left turns it could have pressed down hard enough to trip the limit switch.

Not sure if this is actually the root of the problem or not, but it is the only thing that I have measured that was actually out of tolerance.

[ErrorMargin]

Let us know what happens when you swap the altitude sensor for a known good one.

[GH85Carrera]

Here in Oklahoma City, we are at an averal of 1,200 feet. I have driven to the Colorado Rockies, and out to California over the mountains. Kansas is about the same range of altitude except in western areas, so yea not much need for it in Kansas or Indiana.

I assume my altitude sensor is working. The car is way down on power at high altitude, but it runs just fine.

It will be interesting to hear your results.

[johnman001]

It might be a while before I end up with definitive results, but so far, this is the only item that I can find that didn't check out.

Since the device is a limit switch that would otherwise be normally open, I see no harm in running it with this device unplugged.

While the car is a dedicated track car an not exactly street legal, I may take it out for a run, but won't really know until I can make it to the track next summer. It would, however, be good to know if it is really fixed this time or not, so I may chance taking it out or finding an autoX to at least give it a workout and try to see if problem is gone.

If my theory is correct, I wonder how it would respond to just running off of the O2 sensor by disconnecting the WOT (throttle position sensor).... I wonder at what point the fuel maps run out of gas when tricking the car to run at part throttle.

I also wonder if timing is affected (retarded) when the ECU thinks it's at higher elevation.

Too bad I can't find FR Wilke. I'd like to get some of thier information on Motronic.

[Techno Duck]

A lot of FRWilk's old website is still available using the WayBackMachine on WebArchive.Org

https://web.archive.org/web/*/the944.com*

FR Wilk had one of the only DME chips that i remember that came with a dual map, he did this by disconnecting the altitude sensor and installing a jumper (or just a standard switch) to switch.

Here is one result i pulled up quickly. You can navigate the page as normal from here. This saved the website from 2008 until 2015ish when the site went down, you may find 'snapshots' of the page saved with more info.

https://web.archive.org/web/20150224084714/http://www.the944.com/altitude.htm

[E Sully]

Your correct about the altitude sensor. It should be an open circuit till 1,000 meters. When above that it closes the contacts and leans the mixture.
I live in NY where it would never be necessary, and drove an '86 engine with no sensor installed on my car and it worked perfectly.
As far as the cylinder head temperature sensor, I had installed the newer style 2 wire sensor before I put the engine in my car. There were occasional issues with the original single wire unit, that is why Porsche updated it to the 2 wire unit.
When the full throttle switch engages it uses a different map and makes the mixture richer, bypassing the O2 sensor and shutting off the A/C.

[wazzz]

Originally Posted by E Sully

Your correct about the altitude sensor. It should be an open circuit till 1,000 meters. When above that it closes the contacts and leans the mixture.
I live in NY where it would never be necessary, and drove an '86 engine with no sensor installed on my car and it worked perfectly.
As far as the cylinder head temperature sensor, I had installed the newer style 2 wire sensor before I put the engine in my car. There were occasional issues with the original single wire unit, that is why Porsche updated it to the 2 wire unit.
When the full throttle switch engages it uses a different map and makes the mixture richer, bypassing the O2 sensor and shutting off the A/C.

Not exactly, the DME unit has no possibility to shut off the A/C. There is indeed a wire link between A/C compressor clutch power line and pin 29 on DME unit, but this pin is an input, not an output. It is used to inform the processor that the A/C compressor clutch is engaged.

[E Sully]

While I have not seen this in the Porsche manual, it is stated in the Bentley engine management section.
I know modern cars generally disengage the A/C clutch on full throttle to decrease engine drag and increase power to the wheels for better acceleration. The A/C system returns to normal operation once the ECU sees that full throttle is no longer engaged.

[johnman001]

Thanks for those replies. I think I may take the car out this weekend to see if my issue seems to be gone. I suspect it just might be. I have downloaded and archived most of the wayback machine as it pertains to the old info once available from FR Wilk (the944.com), but it is/was operational theory that seems to be lacking. For instance, was fuel delivery reduced at elevation, or was timing adjusted as well?

Also, I know that at full load (full throttle switch made), the O2 sensor input was supposed to be ignored. Not sure if this was also true with the altitude sensor or not, but in my experience, my car ran better at part throttle with the altitude correction sensor switch stuck closed, so I'm guessing that at lower throttle positions, the O2 was adjusting the mixture over-riding the altitude correction sensor.

I've never had my TPS apart, but I had assumed that it was a carbon track wiper instead of a micro switch and a cam. Now I'm curious. If it is a micro switch, those can be repaired. Interesting.

Imagine driving near the limit but only able to press the throttle down half way or so. Wish I had discovered the problem with the altitude correction sensor while I was at the track in September.

[mysocal911]

Originally Posted by E Sully

While I have not seen this in the Porsche manual, it is stated in the Bentley engine management section.
I know modern cars generally disengage the A/C clutch on full throttle to decrease engine drag and increase power to the wheels for better acceleration. The A/C system returns to normal operation once the ECU sees that full throttle is no longer engaged.

Their errors are frequent!

[mysocal911]

Originally Posted by johnman001

Thanks for those replies. I think I may take the car out this weekend to see if my issue seems to be gone. I suspect it just might be. I have downloaded and archived most of the wayback machine as it pertains to the old info once available from FR Wilk (the944.com), but it is/was operational theory that seems to be lacking. For instance, was fuel delivery reduced at elevation, or was timing adjusted as well?

Also, I know that at full load (full throttle switch made), the O2 sensor input was supposed to be ignored. Not sure if this was also true with the altitude sensor or not, but in my experience, my car ran better at part throttle with the altitude correction sensor switch stuck closed, so I'm guessing that at lower throttle positions, the O2 was adjusting the mixture over-riding the altitude correction sensor.

I've never had my TPS apart, but I had assumed that it was a carbon track wiper instead of a micro switch and a cam. Now I'm curious. If it is a micro switch, those can be repaired. Interesting.

Imagine driving near the limit but only able to press the throttle down half way or so. Wish I had discovered the problem with the altitude correction sensor while I was at the track in September.

Likely, masking another problem!

[wazzz]

Originally Posted by E Sully

While I have not seen this in the Porsche manual, it is stated in the Bentley engine management section.
I know modern cars generally disengage the A/C clutch on full throttle to decrease engine drag and increase power to the wheels for better acceleration. The A/C system returns to normal operation once the ECU sees that full throttle is no longer engaged.

Ed, thanks for pointing this out. That's one more error I will write down on my hard copy of the Bentley. So many of them, unfortunately.

[johnman001]

So, as a follow-up, my running issue seems to have fully disappeared by unplugging my shorted altitude sensor.

The car once again seemingly has full power and pulls strong all the way to the rev limiter in at least the first 4 gears.

My issue may not directly relate to most as my engine is transplanted into a 914 with the altitude sensor mounted under the passenger seat and had been that way since 1991. In the last few years with little or no clearance between the bottom of the passenger seat and the top of my old altitude sensor, eventually, my pressure can was crushed just far enough to allow the limit switch contacts to close sporadically until it finally became crushed enough to close the contacts permanently where I could finally find the problem.

My suggestion would be to mount the altitude sensor such that it would never see pressure against the outside of the housing. For completeness, I did purchase a used unit and mounted it so it won't be pressed against while keeping all the wiring and sensors per factory wiring diagrams for the engine.

Someday, someone else may end up working on it and I'd rather make it easier on them if they have access to factory manuals.

I thought I would pass on what I had found just in case someone else may run into similar running issues.

[Cairo94507]

Thanks for circling back to let us all know; great work.

[ErrorMargin]

Originally Posted by johnman001

...Am I correct in believing that this device should be open circuit or infinite Ohms unless I'm at 1000 meters above sea level?...

Did you confirm this? I want to test my altitude sensor since I have my car apart right now.

[johnman001]

Originally Posted by ErrorMargin

Did you confirm this? I want to test my altitude sensor since I have my car apart right now.

This is confirmed. The sensor is a simple device consisting of a standard micro switch and a pressure capsule.

The pressure capsule below 1000m (3,280 ft above sea level), does not press against the switch, so it should be open circuit below 1,000m and the contact closes when the pressure capsule swells due to the lower atmospheric pressure at 1,000m and above.

944's use the same altitude sensor in case you are wondering. The altitude sensor can be mounted in any position.

In my case, my seat pressed against the sensor causing it to close the micro switch and cause my running issues. Prior to permanently deforming, my theory is that the weight of the seat itself could cause the micro switch to close under certain cornering conditions (intermittent fault). It caused me to chase my tail because until I caught the actual condition, I could not find an actual fault.

To make a long story short, don't mount an altitude sensor that works like this one where the sensor could become crushed or damaged.

[johnman001]

Originally Posted by Cairo94507

Thanks for circling back to let us all know; great work.

No problem. I hate reading some of the threads where you never know if the problem was solved or what actually solved the problem. I try to post what was done to solve the problem if I can.