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

[Wayne 962]

Originally Posted by john walker's workshop

When the valve covers are off, do a cam timing check. Much fun in a 914.

I woke up this morning thinking about the easiest way to do that. I think I figured it out. Not really old school, but by using technology, I think I can very simply check the cam timing left-to-right. I can't see the front pulley, so putting a degree wheel on there would be difficult / impossible to use. So, I'm going to hook up two small microswitches to the exhaust rockers for cylinders 1 and 4, and then tie them into the 2 channel oscilloscope, which will record the position as a function of time (while running the car with the starter). The difference in time between those two should be equal. If one cam is slightly off, the difference in time will be slightly longer on one cycle, and then slightly shorter on the other cycle.

I have a drawer full of microswitches (had to buy a 10-pack when I needed one to repair an old robot), so those should work well. The only tricky part will be making sure the switches are mounted at absolute max cam lift and are consistently mounted the same way both on the right and the left. That might be just a little tricky to get 100% right. Worth a try though, and beats trying to do this with the degree wheel where I can't see anything (on a 911, it would be much easer).

-Wayne

[Wayne 962]

Originally Posted by blyguy

How about adding a fuel pressure gauge to watch fuel pressure in real time? There’s a tap w male connector M12x1.5 I think on the drivers side.

Yup, good suggestion -> already done. I left my fuel pressure gauge hooked up and inside the engine compartment. It reads 2 bar every time I look at it...

Originally Posted by blyguy

Good piece of info to swap for a known-working DME, your problem is probably downstream then (fuel supply, fuel injector, less likely combustion chamber or spark since you have assessed those) Youre still in the dark if you are lean or rich- and those AFR numbers are definitely head scratching !

I pulled the plugs - they look "okay", but some look better than others. I'll post some photos soon.

Originally Posted by blyguy

‘Upstream’ changes that modify the DME output AT IDLE are not dramatic like you said. Fuel injection timing based on the crank position sensor/TDC seems to be ok if you have good initial idle- might be worth checking those signals and sensors again to be sure. The baro sensor even at its worst will lean things out but not to the point of poor running. What happens when you rev the engine when it is running poorly? Stumbling, stalling or running rich?

When I rev the engine, it revs okay, but it does still have that miss. Best described as "random flubbing". It doesn't stumble terribly. In fact, revving it probably makes it run better, as the missing / flubbing is more pronounced at / near idle, and as you rev it the engine has more momentum...

-Wayne

[GH85Carrera]

I have not read the entire thread carefully. One thing I wonder if you have checked is the coil. Is it one of the black Bosch German made coils? The silver Bosch coils are total junk.

I had some totally weird running issues with my 3.2 and it it had me stumped until I chaged the coil. Finding the old black German made coil is tough. And finding a coil that is the right voltage and drop in replacement has eluded me. I bought an old used black Bosch coil and all is great.

[proporsche]

Hi Glen, i did asked in post 86..did hot hear about it....
Ivan

[Wayne 962]

I wrote in post #89:

Originally Posted by Wayne 962

Just looked at the coil a mere two min ago and thought "I wonder if it's the coil". I have some spares at work (not here at home). But the spark is consistent and level - this doesn't feel like an ignition problem. I've seen ignition problems, they are typically easy to spot - one cylinder runs colder than others.

The car is running with an original black coil. I have a new silver one on order that I will get today. Although some people do complain about the silver coils, I just called our BOSCH distributor and their return / failure rate on these is 3%, which is not too bad. In past history, I know that if someone has a marginal car (let's say bad grounds or something else off with the system), then the silver coils may be more sensitive to other problems (I.E. they put in a silver coil as a spare or replacement to a black one and the silver one doesn't work, but they may have something else that's marginal).

I jacked the car up and emptied the oil from the sump - looks good. The car has only done a few hundred miles since it was completed, so the oil should still be good. On these air-cooled cars, there's not much concern with coolant leaking into the oil and contaminating it (like there is on the Boxster / 996). I also took a closer look at the lower valve covers - they are nearly completely blocked by the aftermarket headers, so I might check the valve clearances at the top and see if any are off before removing a whole bunch of stuff below. I really don't think the issue is with the valve clearances (the compression check came back okay, and the valve train is not noisy). Checking the top clearances will give me some clues before I spent 4-8 hours adjusting the bottom (for perhaps no reason).

So, today/tomorrow, I'll check the upper valve clearances, check the timing (as best I can), fill it up with oil (important), put on the new spark plug wires, and see if that makes a difference. Then I will try swapping out the coil to see if it makes a difference.

It ignition wires might be a challenge, as I think the Magnecor wires were used specifically because of the different mounting location of the coil.

-Wayne

[Steve W]

Maybe this is a dumb question, is there any gas left in your tank?


And if so, when you checked the fuel pressure, have you been monitoring it while you've been testing the car through the warm up phases?

[Wayne 962]

Yes, plenty of fuel in the tank (I believe), and the fuel pressure has been consistently 2 bar.

Come to think of it, when I reinstalled the injectors, the fuel rail was making a slightly gurgling sound - I should run that again for a length of time and see if the noise continues. I figured it was just air trapped in there making its way out because the injectors were recently removed...

-Wayne

[Wayne 962]

Just because I'm slightly bored being at home, and frustrated with the car, I decided to do a quickie left/right cam comparison with the oscilloscope. I mounted a small switch on one of the studs and put it on the tip of the valve adjustment screw. Then I wired this to a 9V battery and sent the corresponding signal to the oscilloscope. I did this for both sides.

The goal is to compare left/right to see if there are any variances. Results? Generally it looks pretty good - there are no major variances - at least ones I think would be causing my problem. I'm not 100% sure though, there are some slight variations, so I think I will dig out my dial gauge and z-block and double check it. Maybe this afternoon after I take a break. On the 914, it's really annoying to turn the crank and get to TDC - you need three hands and a mirror.

Anyways, here's my setup. I initially thought that it would be affected by changes in the valve clearance, but I set that out of the equation by setting the switch to activate only at the very top.





-Wayne

[blyguy]

You are really down the rabbit hole Wayne! :P

I watched your video from page 2, where the engine is idling well cold (Lambda is right around 1, equal to AFR around 14.7) I assuming that voltage reading on the multimeter is the oxygen sensor? Oxygen Sensor is cold then starts to dither up to 0.7 down to 0.3 (also seen in the voltage tracing) when engine starts to go lean and run poorly. Then you have this condition where your LM1 is measuring lean AFR and oxygen sensor is reading up and down. There’s a problem with this scenario- your sensors should be agreeing with each other.

You can see that your voltage reading is increasing as the sensor warms up. That’s basically the sensor ‘detecting’ a rich exhaust mixture. 0 V is lean, 1 V is rich. Seems like your sensor is kaput and reading falsely rich.

Hope that is the oxygen sesnsor voltage I am seeing on the yellow meter. The car seems to run nice and can maintain Lambda 1 at cold idle- must not be that broken!



I noticed when you are in open loop (o2 sensor disconnected) everything looks good. Sensor reads 0.5V and you have Lambda 1.0/AFR 14.7. You sensor seems to be reading correctly there (even though it’s not reporting to the DME) Is it the correct sensor? It should be reading voltage below 0.5V when your LM1 is reading lean yet is doing the opposite and reading 0.7V. I don’t think would be possible but is it wired in backwards? I don’t think that would cause this, the sensor detects oxygen gradient between ambient and exhaust. Rich running causes more voltage because of the difference in oxygen level. It’s the intrinsic way sensors work so it can’t be a ‘reversed’ sensor. Have you tried swapping out the sensor?

[mysocal911]

Originally Posted by Wayne 962

Compression check results:

1- 170 psi
2- 168 psi
3- 175 psi
4- 175 psi
5- 172 psi
6- 175 psi

These results look fairly good to me.

I think tomorrow I will head to work and grab the leakdown tester kit that is there. Then, at the same time doing the leakdown test, I'll do a valve adjustment and change the oil (get back to the basics).

Also:

- New plug wires (this still might be the problem)
- Post photos of the plugs (Some look better than others, I'll probably just use new ones)
- Check distributor alignment with TDC
- Check each injector harness

-Wayne

Fuel injection systems that use an AFM are more susceptible to intake valve leaks reducing the engine intake vacuum in the
plenum causing the AFM to lean the mixture. This becomes problematic as the engine warms-up and Lambda approaches 1.0.
Given that, leak down tests need to be performed on a warm engine. Having a longer intake boot (2X) may accentuate this problem.
Using a scope to monitor the AFM signal (DME pin 7), it may be possible to see variations in the AFM output voltage (spikes)
causing the average output voltage to decrease resulting a lean condition.

[Wayne 962]

Originally Posted by blyguy

There’s a problem with this scenario- your sensors should be agreeing with each other.

Yes, that is very confusing, and making me doubt nearly everything. When running the car off of idle, the air-fuel mixture is pegged at 14.7 (or was in the past, everything is a mess right now).

I took a look at the upper valve clearances (five min). They all seem okay. Since the compression test looked okay, and the upper clearances were good, I'm going to assume that the lower ones are okay for now. It would probably be close to 4-5 hours to check them). Later on tonight, after dinner, if I have energy / motivation, I will install the z-block onto the car, and check the cam timing. I know there's a book somewhere with a fairly decent procedure on how to do that!

Then, on Steve W.'s recommendation, I will remove the distributor cap and double-check to make sure that the distributor has been installed correctly. I don't think this is a problem, as it wouldn't show up at idle (more at higher rpm), but since I'm there, I'll check it.

Then, I will replace the plug wires and the coil.

We'll see if that makes any difference...

-Wayne

[Wayne 962]

Okay, I just spent quite a few hours checking the cam timing. It wasn't as "centered" as I would like, but it appears to be correct. It's very, very, very difficult to get the engine to TDC when it's installed in a 914. You have to use multiple mirrors and look at it from different angles (see photos below). You also can't easily spin the engine - but I have a quickie solution for that. Jack up the car, and then chock one of the rear wheels in place with your floor jack. Then put the transmission in fifth gear and spin the wheel to spin the engine. Make sure that you don't spin it backwards. I've used this trick all the time when adjusting the valves on a four-cylinder 914. You can actually get it so that you just take your foot and push the tire to turn the wheel and turn the engine - even easier than using a wrench.

But access is really difficult. I also didn't want to take off the upper valve gasket because these break apart when you take them off, and you can spend another 3 hours just scraping off the remains! And in a 914, the access is terrible. Did I mention really terrible?

Cam timing specs:

0.046" = 1.16mm (cylinders 1-3)
0.054" = 1.37mm (cylinders 4-6)

Spec is 1.1 - 1.4mm with ideally the timing being set at about 1.25mm. Again, it's very difficult to find the absolute TDC for overlap on the 914, and if I'm off by a degree or so each time, it can create some variances. So, these numbers look "good enough" for me to eliminate the cam timing as a potential problem / issue. If they were really, really off, then I would measure again, and use some video cameras to get in there, and also remove the valve cover gasket, etc, and recheck the valve clearances one more time, etc. Then if that still showed a problem, it would be engine out time. But, as I said, these measurements are in spec and they look "fairly okay" to me.

I also checked to make sure the distributor was installed properly (it was). No problems there.

Next it will be buttoning all this back up, replacing the plug wires and the coil.


Dial gauge at .046":



Dial gauge at .054":


Distributor aligned correctly at the line / notch:


My handy dandy Engine Rebuild book for reference (this is copy #1! 17+ years old now!)


Mirror with flashlight, holding iPhone to see TDC:


Slightly better angle for the last photo:

[john walker's workshop]

I sponsored a 914 race car for a while. Glued a mirror to the firewall (no sound pad) in just the right spot to do ignition timing.

[mysocal911]

Originally Posted by Wayne 962

Okay, I just spent quite a few hours checking the cam timing. It wasn't as "centered" as I would like, but it appears to be correct.

I also checked to make sure the distributor was installed properly (it was). No problems there.

Next it will be buttoning all this back up, replacing the plug wires and the coil.

Well, that's great! Sounds like you'll have solved your running problem once the new plug wires and coil are installed, right?

[Wayne 962]

Originally Posted by mysocal911

Well, that's great! Sounds like you'll have solved your running problem once the new plug wires and coil are installed, right?

Wow, I certainly hope so.

Wayne

[Steve W]

Are there any hoses that lead from the intake system to elsewhere, such as to a power brake booster, oil tank, etc that could be letting in unmetered air?


Have you checked that both the fuel pressure regulator, and the fuel damper are not internally ruptured and allowing fuel to be sucked in through the vacuum lines? I've found this before on a 964 that prior to was a complete mystery. Are their vacuum lines connected to the correct ports? All the vacuum lines on the back of the throttle body correctly routed? Do you have the recirculating air valve 930.207.222.0X ? Is the thermovalve still on the throttle body and if so what and where are their ports connected to?

[Wayne 962]

Originally Posted by Steve W

Are there any hoses that lead from the intake system to elsewhere, such as to a power brake booster, oil tank, etc that could be letting in unmetered air?


Have you checked that both the fuel pressure regulator, and the fuel damper are not internally ruptured and allowing fuel to be sucked in through the vacuum lines? I've found this before on a 964 that prior to was a complete mystery. Are their vacuum lines connected to the correct ports? All the vacuum lines on the back of the throttle body correctly routed? Do you have the recirculating air valve 930.207.222.0X ? Is the thermovalve still on the throttle body and if so what and where are their ports connected to?

All great questions!

- There are no mystery hoses that appear to be going anywhere. When I performed the smoke test, I actually had some issues keeping the rubber cone stuck in the intake because the back pressure was building up. Still, it was odd that I didn't see any smoke coming out of the tailpipe. But I guess that wouldn't happen unless one of the cylinders was at TDC overlap.

- The fuel pressure dampener and fuel pressure regulator are properly hooked up to vacuum, and when they are plugged into the intake manifold, they pull vacuum and change the fuel pressure accordingly. I'm assuming if the rubber diaphragm had failed on these that we would be sucking fuel into the intake (the common failure for these). There's no fuel smell coming out of the vacuum pipe and there is no fuel in the vacuum line/pipe either that I can tell.

- The vacuum lines on the back of the throttle body only go to the pressure regulator and the pressure dampener. The thermotime switch is plugged up as the charcoal canister and other emissions-related stuff has been removed. I thought that the thermotime switch might be leaking, so I did smoke the car when it was warm / hot. There were no leaks that I could find.

Been lazy today, binge watching Netflix with Mom and kids on Mother's Day. Heading out to the garage shortly to install the new plug wires, fill with oil, and test the fuel injector wire harness.

-Wayne

[john walker's workshop]

Try an MSD 8222 coil instead of that failure prone bosch silver bullet. Their return rate is probably low because most ended up in the round file.

[Wayne 962]

So, I realize now that the Magnecor wires were probably used because the stock harness doesn't fit too well with the side air flow channel near the distributor. I was thinking this a few days ago. I'll make it work, but there is nothing easy on this car.

But, I did pull off the plug wires and one of the end boots came off revealing a mess underneath. I'm now optimistic that this is the problem (despite me checking the ignition system several times).

Of course, this being a 914-6, the boot from the wire fell off and got lodged behind the distributor where you can't see it! There is *nothing* easy about working on this car!

Take a look at these photos, has anyone ever seen this before? It almost looks like the distributor cap boot or the outside sheath for each wire has melted / fused into the wire itself. Just to see if it was only on one wire, I took another boot off - same thing:





-Wayne

[john walker's workshop]

I really see no reason for fancya$$ plug wires when good stock ones work just fine.