SharkSkin

The number on the pump checks out. I've never bench-tested a later pump, but if it's putting out 100psi and you can't get spec pressure at the rail then there is clearly some blockage.

I reviewed the WSM and the spec calls for the given volume off the return line, so your flow was low.

I'll defer to Wally on the dwell meter. I haven't used one in so long I don't even remember how to hook one up -- sounds like a moment of is in order. It sounded like you were using the in-dash tach, which had me wondering.

Sounds like you have a good plan going forward.

Charley B

A dwell meter should have green and black connectors. The green connector goes to the Distributor side of the coil (as opposed to the batt. side) and the black leed to a chassis ground. This connection should also give a good tach. read as well.
Reconnect the meter and try again. Address the vacumn as directed in the WSM for your model.

Tom Rathjen

Yeah, funny how this thread started out as a timing concern. Of course, I initially did have a timing advance problem because the vacuum lines were connected to the wrong TB ports. And I’m still not sure my retard function is doing anything. So we’ll see if I need to come back to that.

So I connected my meter up incorrectly for measuring dwell angle? Hmm, interesting. Of course, I don’t have the manual that came with this tachometer/dwell meter that I bought in high school (i.e., early 80’s….like when this car was new! Ha). So I was going by the last picture in the recently posted WSM for ignition trouble shooting that shows the meter connected across the coil terminals. Which I guess I misinterpreted, or is for a different type of meter. Well, I can’t wait to hook it up as described above, once I get the fuel tank/pump/filter back together (or can get access to my daughter’s 944…which is really mine).

The new filter (the correct, larger one) arrived today. Yea! The tank-to-pump line and in-tank screen should arrive by tomorrow. And I successfully got the old screen out (reeeaaally tough). The extra/improper pre filter has been properly discarded.

Good idea!

Thanks for double checking me on the pump part number. Glad to know I probably don’t need to order (another) replacement!

I’ll report back after I get it back together and tested. Hopefully Friday, if parts all arrive as planned.

Thanks!

Tom Rathjen

Well, took me longer than I had hoped to get all the parts in and get it back together and test it. But done tonight.

New in-tank strainer and new fuel filter installed. The "extra" filter has been eliminated, and repaced with the proper tank-to-pump hose.

I got the following rail fuel pressures initially (i.e., engine cold):

Pump relay jumped (engine off): 41 psig (2.8 bar)
Engine running at idle: 31 PSIG (2.14 bar)

At first, this appeared to be a big improvement and I was hopeful that I had at last solved the problem. But....when I took it out on the road, the "stumble" still occured at high engine RPM under load (like up a hill). That is, power/acceleration was low, and the engine would "hesitate" (sort of misfire or begin to cut out). And in neutral, I could attain RPM between 5 and 6K at full throttle, but as engine speed increased from idle with the throttle wide open it would hesitate/stumble around 4K for a couple seconds before increasing again. All this is pretty much what it was doing before.

Then, I rechecked the rail fuel pressures immediately after the road test (i.e., engine hot). And I got about the same pressures as I had been observing before:

Pump relay jumped (engine off): 36 psig (2.48 bar)
Engine running at idle: 27 PSIG (1.86 bar)

I also checked the rail pressure with the return lines blocked, and this time got about 60 PSIG...maybe a bit higher.

And I checked the flow at the return line (pump relay jumpered, 30 seconds) and got 1183 ml. (Spec says at least 1150 ml). So that looks OK now.

And I rechecked the fuel rail pressure after the engine had cooled several hours (pump relay jumpered), and it was back up to about 40 psig.

And I connected my tach/dwell meter as described in earlier thread replies (between coil negative and ground), and this time got 27 degrees at idle, and 31 degrees at about 3K.

So....all these results seem to be within specs and as expected. (Although I hadn't noticed before that the fuel pressures were 4 to 5 PSIG higher when cold....I guess because all my troubleshooting before was with the engine warm.) Although it is definitely good that I corrected the filter and filter screen situation, it seems that wasn't "the" problem.

To recap progress since starting this thread....I have corrected vacuum line routing associated with timing advance, replaced the vacuum advance, replaced the leaking (vacuum) fuel damper/accumulator, and corrected fuel filter/screen problems. All of these were necessary and had small positive effects. And I have done all the ignition troubleshooting steps and can find no other problem. Yet I am still having performance problems at high RPM. Perhaps things still on the "list" are: distibutor (rev limiter)?, O2 sensor?, cat converter/exhaust? Still thinking about what to hit next. Suggestions?

Thanks!!

SharkSkin

Well, now your fuel delivery system is shipshape! Apparently it was marginal; good enough to fuel the car under the present conditions, but if not for this other problem you would have lean issues at WOT. The 60PSI measurement means you have plenty of surplus pressure, which is needed to ensure the regulators work properly.

So when you rechecked the dwell, was it stable at 3K and beyond, or did it still drop abruptly?

I re-read this thread, and Wally mentioned closing the plug gap to see if that helped. I don't think you were asked, or in any case you didn't mention, what plugs are you running, what is their condition and what is the gap set at?

Also, while the ignition wires check out, I suggest you try a more aggressive test of the insulation on those wires. Get a spray bottle that can output a fine mist, and fill it with tap water(don't use distilled or deionized). With the engine running, thoroughly spray the plug wires and coil wire and see if the car stumbles or if you can see arcs.

Getting good vacuum while slowly increasing RPM (Post #82) tends to rule out plugged exhaust; it can't hurt to re-test but it sounds like you're OK there.

I doubt the O2 sensor would cause a sudden stumble, but there's no harm in validating that it's working properly.

The distributor(revlimiter) is still worth looking into IMHO.

You've also verified cam timing, right?

Honestly, I am running low on ideas; I'd like to hear back on the dwell, plugs and water test before speculating further.

erwalker

I also have an '82 with a similar problem. Runs fine up to around 3000 to 3500 rpm then starts to break up, miss-fire and stumble. I've gone your route and looked at wires, rotor, cap, plugs, fuel filter, pump, timing, vacuum lines, etc to no avail so I have been researching over the winter with plans to tackle the issue in the spring. As of right now, there are two things I want to look at:

1) internals of the distributor.
2) one of the two microswitchs located on the throttle housing. I looked at this in the fall and the switch is working but I did not get a chance to follow the wires to ensure there were no breaks. The switch and circuit I want to look at closer is the Wide Open Throttle switch. This link may be of help:
https://rennlist.com/forums/928-foru...le-switch.html
Still not sure what a malfunctioning switch/circuit will do but because it is supposed to kick in at around 2/3 open throttle, my suspicion is that it may have some affect on performance at the upper end of the rpm's.
Hope this is of help and let us know how you make out.

Tom Rathjen

It does not drop suddenly at the high RPM. As RPM increase to somewhere about mid throttle, dwell increases steadily from about 27 degrees to about 32. Then, when throttle is increased from there to full throttle, it smoothly decreases back down to around 27. I can't be precise about where the hump in the curve is, because I'm using my tach for the dwell measurement. But it is probably somewhere around 3K.

The plugs are brand new. I installed them in October, as I was just getting the car running. Bosch Platinum WR8DP. I ordered them specifically for this car, so I assumed they were pre-gapped correctly. But....I just pulled one and it is gapped at 0.034. The spec I have says 0.024-0.031 in (0.6-0.8 mm). So….they are gapped little wide. Guess I will pull them all and regap them. I will be surprised if this is “the” problem, though.

Haven’t done this yet. Need to do this earlier in the evening.

Yes, I did this previously. And I rechecked it just now, and it is still correct. (Well…to be precise: when I line the crank shaft up to the TDC mark, the drivers-side cam sprocket’s notch is lined up exactly with the reference mark. The passenger side cam sprocket’s notch is slightly less than half a tooth advanced from the ref mark.)

Yes, understand. Two reasons I’m still considering the exhaust. The first is that I’m just not sure what I am looking for during this test. The manifold vacuum does drop some at high RPM, even when throttle is increased very gradually. And drops rapidly at throttle movements other than very gradual. And I just don’t have experience with what is normal, and it is hard to describe in words. The second reason is that I’ve decided the car just doesn’t sound right. It is very loud. Looking at it, I don’t see any obvious damage to any exhaust components. But I don’t think it is all stock. (There are remnants of a label on the tailpipe that would not have been on the original Porsche one.) So…replacing the exhaust is something I very well may want to do anyway, so it might be good to go ahead and do that and then rule out that as a cause of the stumbling.

Very interesting thread on WOT switch. I just looked at mine. (On my 82, the switch is a lot different than the one in that thread for later models.) It seems to work fine, as does the idle switch. I assume its wires lead to the DME. Any idea which connector contacts?

I guess I am saving distributor internals for last. I am afraid of destroying it in the process of disassmbly if I go deeper than I already went when looking at the vacuum mechanism.

Thanks!!!

PS--it is really interesting how it behaves when all warmed up and in neutral. When I slam the pedal down full throttle and hold it there, the engine rev's up to about 4200 RPM, then literally stops accelerating and sputters/stumbles for a couple of seconds. Then slowly (but smoothly) rev's up to about 5500K. This engine is trying to tell me something, if only I could understand what it is saying.

SharkSkin

If it's not dropping off suddenly, it's probably not the problem IMHO. I think you're in the ballpark on this.

I wouldn't worry too much about this -- I just wanted to get a sense for the condition -- sounds like you're fine. It can't hurt to bring the gap in spec, but IMHO it's not far enough out to be causing the stumbling.

I'm thinking this is in the top 3 likely causes...

Close enough. Moving on...

When performing a vacuum test looking for a plugged exhaust, the clincher -- the positive indication of a plugged exhaust -- is when you find a throttle position such that when you hold that throttle position steady, the vacuum continues to drop. If you get a stable vacuum reading at any throttle position, even if it's a couple in/Hg lower than at a lower RPM, then your exhaust is not plugged -- at least, it's not plugged enough to cause the stumbling.

Do the exhaust tips point straight back? If so, you probably have an Ansa exhaust -- not so good, but as long as it's not plugged or rusted out I'd put it on the back burner. For a bit more than a replacement Ansa, you can get a far superior Borla Cat-back. They were unobtanium until a week or two ago -- 928I just received a batch. The setup would bolt right up to your car, you wouldn't have to modify it like I did. Of course, that won't help the stumbling if it's the cat that's plugged, or if nothing is plugged.

Looks like pins 3 and 10. Pin 10 is also connected to one pin on the idle switch. The other pin on the idle switch goes to pin 2.

I understand your hesitation on this. One thing you can do is take it to a shop with a distributor diagnostic machine (ask around till you find a place that can re-curve distributors) and have them just spin it up and see how it behaves.

I'm still thinking this sounds like the distributor. Does black smoke come out of the exhaust when this happens, or no? Another thing you can eliminate is the AAV. See post # 106. The AAV is item #7. Once the engine is warm, that should do nothing -- it should be closed. If it's somehow opening at the point the car stumbles, it will go lean and cause a stumble. Granted that's a real stretch, but it should be easy enough to pull the hose(to the right of the AAV in the diagram) and block both sides.

Based on all of your troubleshooting so far I'm thinking it must be the distributor or coil/plug wires. When you do the water test be sure you spritz the coil wire.

Tom Rathjen

I had time to kill last night, so I regapped all plugs to 0.025 in. This made no noticeable difference in performance. The car still lacked power and stumbled at high RPM under load. (Didn't really expect it to make a difference, but nice to know it is gapped right now.)

I did the water mist test just now. I sprayed all ignition wires (including coil wire) and the distributor cap pretty thoroughly, and I kept spraying while my daughter rev'ed the engine. I did this in the dark. I did not see any sparks or notice any stumbling affected by the water.

I checked the WOT and idle switches at the DME connector. I found them to be connected to pins 3 and 18, and 2 and 18 respectively. And they produced the proper open and closed signals when the throttle was opened and closed.

I did not notice any black smoke. In fact, I really haven't every noticed black smoke.

I haven't looked at the AAV yet. It looks a little tricky to reach and plug. I also haven't redone the exhaust test yet.

FYI, enclosed is a picture of the tail pipes. I believe they do point straight back.

Thanks

SharkSkin

OK, good work. Sorry about the pin 10/pin 18 confusion -- that part of the diagram is very blurry. IMHO it's down to the distributor(pending your exhaust test). If the AAV is hard to reach on your car, don't worry about it. It is so unlikely that it would behave in a way that would cause a stumble, it's hardly worth checking, but you've covered almost everything else.

That's the Ansa exhaust alright. There is certainly nothing "wrong" with it unless it's plugged or rusted out(they do rust at an alarming rate), but for the money you could do better having one fabbed from scratch. The Borla cat-back is all stainless and is worth every penny IMHO, unless you have done enough engine mods to need more flow.

Charley B

One unlikely scenario, but it happened to me. The formed pipe running between the fuel filter and fuel pump can become crimped when tightening the fitting. It allows enough gas through for idle and normal acceleration but in my case, once above 4000rpm I would loose power. Also on left hand turns it would cut out entirely. Just a thought.

Landseer

Methodical testing. Admirable. You should be with NASA.

How about some from the hip brainstorming to round-out the approach?

Some obvious, most likely checked already. (You can always relegate me to ignore if its coloring too far out of lines.)

Cap condition. OEM (appears Bosch in the one photo of it)? No cracks or carbon tracing beneath?

Coil. (forget that its new, it could have been dropped). Possible breakdown of fields within it under high load.

Two coil resistors. 6 and 4 Ohm. Again, could be misbehaving under electrical load or mechanically resonating causing electrical problem while under vibration associated with higher rpm.

Intermittent wire short in the harness leading around to the coil resistors and coil. The harness passes through some sharp transistions on each side of the radiator and has been known to wear-through and short against body.

Intermittent short (or failure to ground) of injector wires, similar reason, another harness of course. Very specifically, have you gone through the struggle of dremel wire brushing the injector wire grounds located under the PS valve cover, which are in my experience very difficult to reach? If you loose that ground at load, it will starve the motor.

Oxygen sensor. It affects mixture. Mine are multiwire and they are from 92 Ford Taurus (benz logo on connector) for $60. No need to buy Porsche part here if money is an issue. Its an old car, if it hasn't been replaced, I'd change it. Pox on the concept of "buy it only once its proven bad".

Intake boots. They don't look new in your picture. Could you have a tear or dry rotted boot allowing unmetered air to enter during high load?

More pictures. Slam us some shots all around this engine compartment and the CE panel in-situ, please. Need more input.

You need some parts to switch in and out. I might be able to drive the 84 automatic up there (before I finally pull the transmission for fixing the parking pawl and TC leaks) and we switch parts back and forth (AFM, coil, etc). I think its mostly the same stuff, we'd have to check though.

Just trying to help you.

Gotta also thank you for this thread because of the fantastic troubleshooting techniques and data. Definitely a 'subscribe" thread. Am sorry you haven't gotten it running to specification yet. You've got a lot of patience, though.

SharkSkin

Good stuff, Chris. Several of those things have been closely examined more than once, but it's always good to get a fresh perspective. BTW, if the coil fired across a 12mm gap per the troubleshooting procedure it's probably good.

The wiring can be checked by bypassing it, wiring +12v directly to the ballast resistor.

Tom, it may be hard to reach that WOT switch, but see if you can figure out if the stumbling is happening at the point where that switch closes.

Tom Rathjen

I am with NASA! Ha. How did you know?

Tom Rathjen

The WOT switch is actually very easy to reach on my car. I do not believe the stumbling coorelates at all to the switch being activiated. Stumbling seems to be related to RPM and load (i.g. hill), not throttle position. (That is, stumbling occurs at a variety of throttle positions, but at the same RPM for a given load.) To test this further, I tried disconnecting the WOT switch and driving around. This made no difference (that is, it still stumbled).

I repeated the exhaust test by connecting the mityvac to a port on the front of the intake manifold (the part in the middle of the spider legs). I still don't see the "clogged exhaust" behavior that I think was described, where the vacuum keeps falling (to zero?) at some throttle position(s). What I see is that it stays fairly steady at a given throttle position, starting at about 20 inHG at idle. When increasing throttle, it momentarily drops but then rises back up and stablizes....but the vacuum it stablizes at becomes gradually less as RPM increases. At the highest RPM positions (above about 4K), the vacuum drops to around 5 inHG and stays there...it does not rise up as the engine speed stablizes, but nor does it keep falling as the throttle is held in nearly-wide-open positions.