SharkSkin

The tech specs book calls for 2.0 bar at idle, 2.5 ±0.2 bar with the engine off and the fuel pump running(jumper @ relay). That translates to ~30 & 36 psi, so it sounds like you are in the ballpark. If the pressure is not dropping when the car stumbles then IMHO it is not a fuel pump/filter/regulator problem, though injector function has not been ruled out as a cause.

It doesn't sound right that the damper does not hold a vacuum, but the way I read the diagram and docs that is just a damper -- there are two pressure regulators at the rear of the fuel rails that should be the primary influence on fuel pressure. Do you not have these parts? If you do, you should be able to connect the mityvac to the right of the blue cross fitting and influence the fuel pressure as you pump it down. I don't understand why a component that is described as a simple damper should need a vac line at all.

Does the O2 sensor have only a single wire? I thought it had two, but I'm not an expert on the system. BTW, the numbers in my earlier post were missing a decimal point. Range is 0.2v-0.8v, with 0.45v being "correct" mixture.

The fact that the engine runs the same with or without the O2 sensor means it is suspect IMHO, but again I'm not an expert on L-Jet -- hopefully those who are will chime in.

Tom Rathjen

Yes, this car does have the two fuel pressure regulators at the rear of the engine. I verified that they each hold vacuum. I didn't try using the mityvac to see how vacuum on these regulators influences fuel pressure, but I can do that later. I would think that the damper not holding a vac is a problem from the point of view that the vacuum lines to the regulators loose vacuum also because of the loss at the damper. (Since they are all connected at the blue cross.) With the vac line to the damper plugged, and therefore with the regulators seeing the vacuum that they are suppose to, I get the following fuel pressure on the rail: at idle about 25 PSIG, and at full throttle about 32-33 PSIG. So the idle pressure is about 4 PSIG low. I hate to conclude anything yet, though, until I have a good damper installed.

Yes, my O2 sensor is just the single wire type. So my understanding is that I should check for voltage between that wire and ground (and yes I was looking for values less than a volt). And when doing that, my simple analog volt meter shows nothing. So I am also thinking it is suspect, but I would like to be sure my test is valid. Maybe I could do a similar test on my daughter's 944, and see if my meter picks anything up. When I can catch the car here, that is.

SharkSkin

I agree that it's worth looking into. I was hoping that you would see a big drop in fuel pressure at 4K RPM, which would explain that behavior. So, you've found something else that needs attention, but IMHO it's not "THE" problem.

Tom Rathjen

I've ordered a new damper, and will see what the fuel pressure does when I replace the bad one. It should arrive by the end of the week, so I'll plan to install it next Saturday. (Yep, not expecting it to be "the" problem either.)

I took the ohm meter to the rotor, cap, and wires. All the wires tested 3K ohms. The center-to-end on the rotor tested 4.6K ohms, and it looked in very good shape. (A new rotor is only about $12, so if it really is supposed to be 1K that's an easy replacement.) There were no arcing marks or contamination in the cap, and I measured no shorts between the contacts and center.

I also tried testing the O2 sensor on the 944, to "validate" my testor. And I could register no voltage. So I'm thinking that my meter really isn't adequate for this, and I therefore haven't proved the 928's O2 sensor is faulty.

SharkSkin

That's what I get for guessing off the top of my head. The rotor does have a resistor in it, but it's ~5K -- so you're fine there.

Note that while you may not detect a short in the cap, at ~30KV+ it doesn't take much for the spark to find another path. You really need to inspect the cap closely in bright light. Any tracks across the inside surface of the cap should be suspect. They would look like lightning, outlined in extremely fine dust if present.

I've been meaning to do this for a while, along with some other things -- I figured now, when it's needed, is as good a time as any -- I've written up and posted the 16v ignition troubleshooting procedure from the WSM(with some minor enhancements) on this page. Hope that helps.

As for the O2 sensor, I'm suspicious that the car behaves the same with or without it. Maybe shoot "John Speake" (That's his exact user ID) a PM and ask about it. Maybe nothing, maybe something -- he's an expert and can answer off the top of his head.

Tom Rathjen

I'm back from another week of work travel. The new fuel damper arrived, and I just installed it. Now, my fuel pressure readings are:

- idle: 26 PSIG (=1.79 bar)
- all engine speeds up to 4K: still 26 PSIG
- above 4K (with full throttle being about 4.3K), it increases slightly to 30 PSIG
- the fuel pump on (jumpered at relay socket, engine off): 33.5 PSIG (= 2.31 bar)

So....pressure seems just a tad low or barely in spec at the low end. Perhaps I should go ahead and get a new filter, since I was unsure about the condition of mine. I doubt that is "the" problem, though.

I also inspected the distributor cap again very closely with a bright light. I still see no evidence at all of a problem. It looks very good.

Oh, and I guess I should mention that performance is exactly the same (low power and stumbling about 3K) after replacing the damper.

WallyP

Interesting problem on the dampener. The vacuum line appears to be primarily for safety reasons. If the fabric and rubber diaphragm springs a leak, you do not want high-pressure fuel spraying over the engine, so the vacuum line keeps any leakage inside the intake system, where you already have fuel.

If the unit won't hold vacuum, you would expect the diaphragm to be the most likely problem. But - if the diaphragm leaks, the engine will normally be really hard to start when hot, as the excess fuel dumps into the intake, causing a very rich condition.

Sounds as if the housing was leaking, not the diaphragm. If so, it was just a run-of-the-mill vacuum leak.

Tampa 928s

Just skimming the thread if your fuel pressure is low you may have a delivery issue. There is a in tank filter that maybe clogged or you have junk that may have plugged up a new filter if you installed one. Did you measure the voltage to you pump? Either way mine is 32 at idle and 39 at load.

Tom Rathjen

Yes, hard to tell if it was the diphragm or housing that was leaking. It was a small leak, because it would take a couple of seconds to loose vacuum after pumping it down to about 15 psig with the mityvac. In any case, it is fixed now, and as Dave suspected was not "the" problem.

I just ran through essentially all of the tests in the ignition troubleshooting procedure that Dave recently posted. A lot of those test I had done before, but a few were new. I found three questionable results:

1) My plug wires are all 3K ohms, whereas the procedure says they should be 2.5K.
2) Wen testing voltage at coil (terminal 1 to ground) I got about 2.2-2.3 Volts, wheras the procedure says max should be 2 volts
3) Dwell angle seemed strange. At approx 1500 RPM it was within spec at about 27 degrees. And it stayed pretty steady as RPM increased until nearly full throttle when it suddenly dropped to about 23 degrees and the engine noticeably began stumbling. I cound not tell at exactly what RPM this drop occurred since I was using my tach to measure dwell angle. But it must have been somewhere near 4K since that is about all I get at full thottle. The stumbling definitely seemed associated with the drop in dwell angle.

So, hmmm. I don't really think 1) is a big problem. The plug wires all look pretty good. And 3K is about what I measure on my daughter's 944 wires which I bought new not long ago. But...2) and 3) both have do to with the ICU according to the posted procedure. 2) is pretty marginally out of spec, but it is out of spec. And the dwell angle behavior seems especially strange, since the spec suggests that it should increase some at the high RPM. And one other thing that makes me wonder about the ICU is the flakey tach behavior.

But I wonder if this definitely is a problem in the ICU, or something in the distributor Earlier info in this thread mentioned a rev limiter in the distributor, which is something that I haven't tried to look at at all. The distributor sensor and green wire did check OK, though.

SharkSkin

(1) and (2) seem to be close enough. I don't have any technical data on the mechanical revlimiter setup(I'm not even positive that it's present in your MY) but I'm guessing that, where implemented, it is a spring and counterweight design. This type of governer mechanism will activate early if one of the springs is broken or weak. Worth looking at, IMHO.

The overall low fuel pressure means change the filter and check again IMHO, especially since you don't know the history of the filter.

WallyP

A fuel flow test will tell you whether you have a partial fuel blockage or not.

Sounds as if you have found a significant clue to the problem. The dwell change sounds like a pickup problem, although it would be nice to really check the green wire. Substitution is probably the only good test.

Remember that Porsche put out a batch of faulty green wires a year or two ago.

SharkSkin

Good points, Wally. I was going against my own advice about swapping parts without testing when I suggested the fuel filter, and there is a good test for that in the WSM. Given the car's 5-years non-op and unknown history of the filter before that, I would change it just so I know it's good.

I'm still leaning toward a mechanical problem causing the sudden dwell change -- at least it's worth looking at. An impedance problem with the wire or sensor would, IMHO, result in a more gradual change in dwell. Do you have any specs on the distributor, an exploded diagram or pics of one disassembled?

Tom Rathjen

Just did the fuel flow test. I sampled in two places: directly off the supply line that comes into the engine compartment, and off of a return line coming off of one of the pressure regulators (with the return line off of the other regulator plugged). Off the supply line, I got about 1530 cc in 30 seconds. Off the return line, I got 946 cc in 30 seconds. The spec I have says 1150 cc minimum in 30 seconds. Assuming that is off of the supply line, then I have good flow.

Also, regarding my "green wire", I tested it pretty thoroughly earlier. There is good continuity on both conductors all the way from the distributor end to the ICU connector. And I tested the points that the connector attaches to in the distributor and got resistance values in the expected range that "wiggled" as the rotor was turned as I assume is expected. (Did this with distributor out of the car.)

Landseer

Those flow numbers seem reasonable from an acedemic perspective.

What is really going on here?

SharkSkin

OK, so the filter seems to be good. The flow test should be conducted ahead of the regulator. The low pressure may be due to a pump or regulator that is not up to snuff. If you block both return lines, what is the rail pressure? Block each return line one at a time(so you are checking one regulator at a time). What is the rail pressure in each case?