Need some input -> Problem with 3.2 transplant going lean and misfiring after two min

[Wayne 962]

Originally Posted by Steve W

BTW if you ever fix this thing, the next thing you should address is the clusterfuch of elbows om your intake. From dyno testing every 90 degree bend kills power by 4 hp to the wheels, and the one from your air filter to the AFM is horrible, at least a 10hp killer!

Yes, I wouldn't have done it that way. The good news / saving grace, is that this is in a 914, and is pretty much the only car I have that can do burn-outs. I.E. it has a lot of horsepower to spare for the car that it's in. Still, every HP is sacred!

-Wayne

[Steve W]

Probably easiest and an elegant way is to take a piece of mandrel bent aluminum tubing and fab an aluminum adapter between the AFM and intake boot to a minimum angle to straighten the air path, maybe ending up with a 30-40° elbow, and replace the air filter adapter with one from MSDS, internally ported as below to maximize the velocity stack from the filter to the meter. This is what most of Porsche club racing on the west coast is running after extensive testing I did years ago:

[jpnovak]

That AFM trace is trashed. Reposition so the wiper is actually on the resistor trace.

On the filter adapter, 3D print one...

[Wayne 962]

I just got off the phone with Mike at Autobahn in San Diego, and he's going to be sending me a really nice Air Flow Meter from a 964 ($350, Phone 619-444-3229). They have sold about 76 of these, and have about 60 in stock. Seems like the best approach, as the rebuilt ones are suspect, from what my research says.

As for my old one, I took two more photos. Steve W. - I do agree with you that this looks odd, as the trace is "worn" off of the plate on the right-hand side. I can think of no reasonable explanation for this, as the unit is currently "homed" in the proper position. Looks like perhaps someone mucked with this in the past, and then someone then went and "fixed" it and centered it back again.

Here are two photos showing what I'm talking about:




closer up:




Heading out to the garage to start on my list for today:



-Wayne

[ischmitz]

wow - this thread moves fast:

- The noise you see is ignition-interference most. Don't worry about it
- I don't think the track is your current problem, the scope verifies that. I might turn into an issue later
- The spring tension might be way off because someone was clearly in there before and that could be most of your problem
- BTW, Sal here on the board has largely disassembled and modified the machine code for the 3.2 and really developed a new chip rather than just changing the fuel and ign maps like all other "tuners" here. He is a real expert on this matter when it comes to the Motronic software.



Ingo

[Wayne 962]

Originally Posted by Steve W

Porsche superceded the AFMs on the 3.2s to a 964 part number. A few years ago, Porsche offered replacement AFMs on an exchange basis for something like $700. No idea who redoes them for Porsche.

I had a conversation this morning with my contact over at one of Pelican's main suppliers, SSF. He confirmed what I thought was true:

- Porsche does have a supply of these available (about $900 at Pelican).

- These are BOSCH remanufactured units (you can tell from the sticker on the side).

- Just because it says BOSCH remanufactured on the side doesn't mean it's been remade or redone by Hans or Franz in Bavaria. From what I've been told, BOSCH contracts nearly 100% of their rebuilt parts to outside contractors, many of them here in the states. So, what you might be getting would be a Programa or Brett or who the heck knows US-based company that rebuilt this and BOSCH then stuck a sticker on it, and then sold it to Porsche, who then stuck it inside a Porsche box. Unfortunately, this is how it has been for many years, I would suspect that it's similar with this particular unit:

https://www.pelicanparts.com/More_Info/930606123X.htm?pn=930-606-123-X-OEM&utm_source=googleppc&utm_term=930606123X&utm_campaign=Shopping+Feed&gclid=Cj0KCQjw-_j1BRDkARIsAJcfmTHAPLHWNgI4aHF1X_-P4p59WL6WFuQf_phdpv_1od00cae3LFBao5QaAtJ-EALw_wcB

https://cdn1.pelicanparts.com/catalog/images/SuperStock/930-606-123-X-OEM.jpg

-Wayne

[Wayne 962]

Originally Posted by ischmitz

- The noise you see is ignition-interference most. Don't worry about it
- I don't think the track is your current problem, the scope verifies that. I might turn into an issue later
- The spring tension might be way off because someone was clearly in there before and that could be most of your problem

I agree. The scope (4 times) has said that the track is not a problem, so I think I'm going to use some very strong tape and just tape up the cover for now.

As for the spring tension, that may be off too, although I really don't see any evidence of the spring holder being moved. Like in this shot from a video, you can see a mark where the years of the spring holder being in place has deformed and marked the plastic slightly. I don't see any of that in this AFM.

-Wayne

[gtc]

Looks like you've... traced the problem to it's source.



I'll show myself out.

[mysocal911]

Originally Posted by Wayne 962

As for the spring tension, that may be off too, although I really don't see any evidence of the spring holder being moved.

OK, but since the source of the poor running engine has not been determined yet, to conclude that the spring setting is correct, is naive.
Furthermore, with the engine running, the wiper arm has not been slowly moved, without changing the spring tension wheel,
in either direction to determine if there exists a fuel mixture problem causing the poor running condition. If little or no change occurs,
then one can basically eliminate a fuel problem the result of; a marginal fuel pressure, a bad temp sensor, poor injectors, an intake air leak,
or an incorrect setting of the AFM spring tension.

It's that simple! Obviously easier and less costly than using the continual shotgun approach of buying and replacing unnecessary parts, right?

[Wayne 962]

Originally Posted by mysocal911

OK, but since the source of the poor running engine has not been determined yet, to conclude that the spring setting is correct, is naive.
Furthermore, with the engine running, the wiper arm has not been slowly moved, without changing the spring tension wheel,
in either direction to determine if there exists a fuel mixture problem causing the poor running condition. If little or no change occurs,
then one can basically eliminate a fuel problem the result of; a marginal fuel pressure, a bad temp sensor, poor injectors, an intake air leak,
or an incorrect setting of the AFM spring tension.

It's that simple! Obviously easier and less costly than using the continual shotgun approach of buying and replacing unnecessary parts, right?

Bingo. "Dave" is correct on this, it's premature to say the problem is with the air flow meter. The meter resistance strip is definitely worn, so it's a worthwhile endevour to get a replacement one that hopefully is in better condition. Everything that I've replaced has been visibly worn or tested to be defective (with the exception of the coil, which was just a guess - coils are tough to test). I'm uploading the latest video right now.

-Wayne

[Wayne 962]

Friday update:

- Installed the cylinder head temp sensor
- Reinstalled the narrow-band oxygen sensor
- Buttoned up and reinstalled the air flow meter
- Installed the new wide-open-throttle switch, and repaired harness. Tested with “wide open throttle tester” (a big, long poster tube that presses down the gas pedal).
- Warmed up the engine and adjusted the idle screw and then the mixture, then the idle again, and the mixture again.

Result: NO CHANGE.

Video:

0:00 - Engine bay - engine running poorly when warm (o2 sensor unplugged)
0:14 - Tailpipe – you can hear it missing
0:26 - AFR on exhaust shows spot-on 14.7 (where it should be)
0:37 - Back to tailpipe
0:56 - Tach -> about 800 or so rpm
1:06 - Rev car to about 2000-3000 or so, revs fine – can’t hear it on the video but I believe it’s still missing. Certainly sounds good on the video.
1:39 - Rev car again to 2000-3000 or so, mixture shows 14.3-14.4
2:12 - o2 sensor – plug it in (put phone down to do that)
2:35 - Back to AFR monitor – the engine starts to run lean (15-16 AFR)
3:00 - Rev engine with camera on AFR meter – 15.0 / 14.9 etc. engine at 3000+ rpm, which is leaner than before the o2 sensor was plugged in
3:32 - Back to idle -> goes lean



My thoughts:

- Just to recap, I have replaced only the stuff that was worn out, looking worn out, or tested faulty:
- Spark plugs, wires, cap, rotor, (coil, although I put the old black one back in)
- Cylinder head temp sensor (tested faulty)
- Wide-open-throttle switch (I broke this removing it to test)
- Took injectors out to have them cleaned and tested (they were dripping according to the report we got back)
- Cleaned the fuel injection harness connectors which seemed to fix the higher resistance of some of the injector plug harnesses
- Replaced O2 sensor (original was 12+ years old)

- The car has run consistently poorly this whole time – this is not an intermittent problem. It’s very, very consistent. It seems to warm up and run fine and then after about two minutes, it runs poorly.

- As mentioned in the early pages of this thread, the car runs poorly at idle, missing (after warming up), with the O2 sensor unplugged. AFM out the tailpipe shows 14.7, but the car is missing.

- If the sensor is then plugged in, the car starts to run lean. We checked the sensor (multiple sensors), and the O2 sensor is indicating that the car is running rich on cylinders 1-2-3 which presumably makes the Motronic system try to lean it out (we’re seeing it go from 14.7 to lean when the O2 sensor is plugged in). PROBLEM: the tailpipe shows 14.7, yet 1-2-3 show rich. This has been a consistent problem during our testing. The presumption is that 1-2-3 are rich, and 4-5-6 are leaner, which is causing the O2 sensor in the tailpipe to read normal. I currently do not have a quiet way of individually testing left / right on the car (I might take it to the shop where we can take off the muffler and test each bank individually).

- Plugging in the O2 sensor makes the Motronic system lean out the engine (it thinks the car is running rich when it doesn’t appear to be).

- This does not appear to be an ignition issue. I've checked the ignition system numerous times with no signs of problems, and an ignition issue would not give this weird left/right lean/right issue?

Again, car seems to much, much better well when stone cold. See stone cold video:



As I’ve mentioned before, I do not think that the air fuel meter is the issue, as the mixture appears to be correct at the tailpipe, but the engine is running terribly. The evidence seems to point to 1-2-3 running rich, and 4-5-6 running lean. Again, I won’t be able to tell unless I head to the shop, remove the muffler (again), and individually test the left and right banks of the engine.

QUESTION: If my supposition is correct (lean on one side, rich on the other), what is a plausible explanation of what could be causing that to happen?

Steve W lent me a fuel rail, injectors, fuel pressure regulator and damper yesterday – I will probably just swap all of these out over the weekend and see what happens. I’m not sure if I will go to the shop and test left/right before or after I do this (probably before).

Thoughts?

-Wayne

[Wayne 962]

Okay, just went for a bike ride and thought about this a bit. If the supposition is true, that one side is running lean and one side is running rich, then what indeed would cause that to happen?

- Problem with the wire harness? I checked out the wire harness and cleaned the contacts and both sides seem to be okay now. Since pin 14 and 15 are joined together in the ECU, the only way this could be a problem is if there were a problem in each individual side harness. After cleaning the contacts on the harness, the resistance seems to have evened out. Steve W. lent me a harness for the injectors, I could try swapping it out (would take about five min). Maybe tomorrow.

- Clog in the fuel lines somewhere? More than one person has suggested this. Higher pressure on one side of the engine than the other. The fuel pressure gauge seems to indicate 2 bar of pressure, which is accurate, but there's only one pressure port on one side. If there is something causing a pressure drop (like perhaps a bad fuel damper), then that could be causing the issue.

- Problem with injectors on one side? I swapped around the injectors and the problem seems to stay the same. Plus I had all of the injectors tested and cleaned. Plus I individually looked at them yesterday for leakage and fuel flow and they all seemed similar. No obvious issues.

- Air blockage somehow to one side of the engine? This is an interesting thought. So, if there were a paper towel or rag stuffed in the engine / intake manifold, it would cause an unequal distribution of air throughout the manifold. I could take off the manifold and look inside. It also might show me the air route to each cylinder and see if any thing is blocking either the air or the fuel spraying into each cylinder.

- Also worth thinking about out loud is the almost 180 degree bend behind the air flow meter. Could this be somehow interfering with the routing of air into each cylinder? I doubt it, because it still has to make it past the throttle plate / bypass. A test for this would be to try to run the engine with the boots off, and the airflow meter disconnected from the air supply. You'd have to prop open the air flow meter and hold it at a certain amount, but then you could measure the mixture on the right bank and then also out of the tailpipe and see if that discrepancy between the two gets any closer. Note: I'm not sure this test would work very well and I don't think this is the problem.

- Air blockage somehow in the exhaust on one side? If there were something stuffed inside the exhaust causing a blockage or something then that would cause air to circulate around the pipe with the blockage and would feed more air to the other pipes. A blockage in the exhaust might show up if I take off the muffler and measure the mixture on each side - I could measure the volume of air coming out of each bank and see if there is any discrepancy left-to-right. I'm not sure how each this would be to do either.

I'll have to think how any of these potential issues fits in with the observations observed so far. Most curious is the "running on three cylinders" and then plugging in all six injectors causes the car to run increasingly poorly. That is a big clue that seems to perhaps point to fuel pressure / damper issues.

Hmmm....

-Wayne

[mysocal911]

Originally Posted by Wayne 962

Most curious is the "running on three cylinders" and then plugging in all six injectors causes the car to run increasingly poorly.

You're not reading/understanding the posts! That was explained in post 194;

Remember, the inductive load of the connected injectors affects the injector pulse width, given the design of the
911 3.2 fuel injection system versus the later 964/993 design. As a result, the effective fuel mixture gets changed.

Having six injectors or three injectors connected is significant! Closely check the injector pulse widths with your scope.

When listening to your videos, it's not easy to detect a miss. Time to take the vehicle to a knowledgeable Porsche shop,
e.g. Aase Motors presently has three 914s with 3.2 motors.

[Wayne 962]

Originally Posted by mysocal911

You're not reading/understanding the posts! That was explained in post 194;

Having six injectors or three injectors connected is significant! Closely check the injector pulse widths with your scope.

When listening to your videos, it's not easy to detect a miss. Time to take the vehicle to a knowledgeable Porsche shop,
e.g. Aase Motors presently has three 914s with 3.2 motors.

Yes, yes, yes. "Dave", I do understand your posts. I am well versed in problem solving skills. I graduated from MIT a semester early undergrad with Tau Beta Pi honors, and also from MIT grad school a semester early with a master's in manufacturing. They taught us the "proper" way to test systems and analyze results - nearly all of which I have perhaps completely flubbed here. That was indeed about 25 years ago and perhaps I'm rusty on my skills. Probably been spending too much time behind a computer at Pelican, and for the past two years probably spending too much time with my family. Agreed - I do not see the volume of cars that a local shop like Aase sees - daily experience does often tremendously outweigh "practical" or "institutional" knowledge. These guys "on the ground" know their ^%$# and there is indeed absolutely no replacement for the experience that a shop like them would have.

BUT (and this is a big BUT), this car is my first car - the first car I ever bought. The first car I ever owned on my own, my 30+ year old 914 which I cut the trunk floor out, the car I drove 700 miles to Phoenix to go see my girlfriend, the car I swapped in the 914-6 powerplant - the car that I have spent more time on than I spent working on any of the books. This tiny 914 is my personal Ahab. I let the car out of my control when I had someone install the 3.2 motor and I've never been quite happy with that decision/result. When it was stolen I went out at night at 2AM for weeks at a time looking for it (found it). So, for me, it's a personal mission, a vandetta per se, to see this through and to "take it over" myself, no matter how hard the mission and how difficult it may be to solve. Indeed, this is what our Pelican customers strive for with their own cars - that "feeling", that sense of accomplishment, that "win" and sense of ownership that is difficult to explain. Sure, anyone can take a car to a shop and pay someone "else" do it. But isn't that like sending off your wife to be serviced by some other… , well, you can fill in the blanks...

There's the technical side of this business - the nuts and bolts, and the volts and the resistors. And then there's the passion, the smell of burnt air-cooled motor oil, the challenge and the reward of figuring it out, of restoring something to new (or better than new!). When one looks at it from a 100% practical sense, it doesn't add up - it doesn't make sense. Why spend so much precious time and money messing around with a bucket of bolts? But when passion and pursuit is figured in, that human element is what makes this special - it's what I did for 20+ years (before the stress started to take chunk out of my stomach lining – just recovering now).

So, I appreciate your input, your knowledge, and your “witty” sarcasm – I have a thick skin and I am not easily bothered. But, I am also stuck at home with my three teenagers, and I don’t have very much else to do but wrench on this car and struggle to figure out what the heck is going on with it. Maybe someday in the near future I will give up and ship it to Kenny/Ray/Grant @ Autowerkstatt or Aase, or someone else. But for now, I really want to fix it myself – I think I owe it to the car at this point for letting it out of my control (multiple times) in the past.

I really do appreciate everyone’s help and assistance here on the boards.

😊

-Wayne

P.S. This post was written after at least a ½ bottle of wine…

[Wayne 962]

Here's another question - the 964 has only a pressure regulator and no "fuel damper" in the fuel system. Why is this?

-Wayne

[manbridge 74]

From the Service Information Technik for the new 964 platform.....

“The use of sequential fuel injection means that the fuel pressure damper is no longer required on the 911 Carrera.”

[Wayne 962]

Originally Posted by manbridge 74

From the Service Information Technik for the new 964 platform.....

“The use of sequential fuel injection means that the fuel pressure damper is no longer required on the 911 Carrera.”

Indeed, that makes sense, since all of the injectors are not firing at once in sequential injection...

-Wayne

[FrankM_]

Can the position of the o2 sensor in the custom exhaust system have anything to do with it ? Contamination by condensate so for down the line (if I remember the picture well) ?


Sent from my iPad using Tapatalk

[ischmitz]

Wayne, you never mentioned whether you tried the suggested test of moving the barn door while the engine is idling. This will quickly tell you if a different mixture makes things better.

[Big Dav]

Originally Posted by ischmitz

Wayne, you never mentioned whether you tried the suggested test of moving the barn door while the engine is idling. This will quickly tell you if a different mixture makes things better.

+1 on this

I'd be keen to see the results of this too as a number of people have previously mentioned. Remove the intake pipe from the AFM so you can access the door inside the AFM and run the motor without the filter for a few minutes. Using a screwdriver or a pencil push the swinging door inside the AFM in a touch and see if that smooths out the running. Also try pulling it back a touch with a small pick and see what difference that makes. In this way you will be able to gauge if the spring tension is incorrectly calibrated.

My 3.2 AFM had been tampered with and altering the spring tension just two teeth made a huge difference to idle and general running and fuel mixtures are now perfect from idle to redline.

Pushing the door in will richen it up and pulling it back will lean it out. If one way helps you then have some clues on whether to tighten or loosen the clock spring inside the AFM.

Once you get the spring tension right you can then adjust the idle mixture with the hex screw on the body of the AFM to get the mixture correct at idle. If the spring tension is right the hex screw should only be one or two turns out from fully in.