It was somewhat difficult to block the return lines. But, as best as I could tell, blocking one at a time made no difference. The rail pressure remained at about 24-25 PSIG (1.6-1.7 bar) at idle. However, with both return lines blocked, the rail pressure increased to 32-35 psig (2.2-2.4 bar).

I tried another experiment, and got some interesting results. I disconnected the vacuum lines going to the regulators. Using the mityvac, I applied vacuum to the regulators one at a time. I got the following results (both regulators performed the same):

zero vacuum, 35 psig rail pressure
10 inHg vac, 32 psig rail pressure
15 inHg vac, 29 psig rail pressure
20 inHg vac, 26 psig rail pressure

I got these results on each regulator both with the engine running, and with the engine off and fuel pump relay jumpered.

Also, using the mityvac, I measured the vacuum on the line that goes to the regulators. I got:

idle: 20 inHg
2K rpm: 21 inHg
3K rpm: 20 inHg
4K rpm: 15 inHg

The vacuum on this line behaved a lot like the manifold vacuum that I tested earlier. With slow increases in throttle, it stays fairly steady. Then, somewhere in the high 3K's rpm, vacuum started falling off. And with a rapid throttle increase, it rapidly decreases to something like 5 inHg, and then rises back up the values above if throttle movement stops.

I drove the car around a little with the vacuum lines disconnected from the regulators (so with a rail pressure pretty steady at about 35 psig). The car seemed to have more pickup. However, it did still "stumble" between 3K and 4K going up a hill. BUT....for the first time since I have been working on this car, full thottle in neutral yielded almost 6K rpm engine speed. Up until now, full throttle was usually in the low 4K's. When increasing the throttle to full open in neutral, the engine speed increased steadily to 4K, then hesitated for a second or so, and then increased to almost 6K.

And the last thing I checked was what the dwell angle was doing at the higher engine speeds that were previously unattainable. And I saw the same behavior that I saw yesterday....dwell angle falling into the low 20's at the high engine speeds.

So....hmmmm.

 

Tom, data aside, how old is the fuel filter?

 

If you had ignition points, I could tell you what was wrong with the car - the pivot on the points would be sticky, or the spring on the points would be weak. The dwell should be set electronically, and I don't think that it should drop.

I would expect higher fuel pressure with the return blocked - about 60 psig.

 

Tom, good test -- it sounds like your fuel pump can deliver the volume but not the pressure. The pump probably corroded from sitting and is not sealing well enough to generate enough pressure. If you can check the pressure ahead of the filter it's worth doing. A filter is much cheaper than a pump so it's better to be sure before parting with the $$.

I'd be taking the distributor apart about now, looking for any obvious mechanical defect, corrosion on the pickup, etc.

 

Yes, I remember the good old days when distributors had points!

Sounds like next things to focus on then are the pump and distributor. Interestingly, this is not the pump that sat on the car for five years. (I hope I mentioned that somewhere early in this long thread.) I replaced the first one, which seemed much louder than it should have been. (It had been hot wired to bypass the fuel pump relay, so came on as soon as the key was turned to run. I guess some PM could never find the recessed fuel pump relay socket that drove me nuts for a while.) But, like so many replacement parts I get, the pump was used. So, perhaps it was not good. Not sure I have a way to test it ahead of the filter, since the pressure gauge I bought is specific for attaching to the rail. But I'll look at it. If I do that, or even find a way to test it on the bench, what kind of pressure should I see (with the ouput flow blocked, I assumed, to get the max head pressure)?

I am ordering a new fuel filter now. I didn't realize it was so cheap (like 20 bucks).

I did take the distributor apart somewhat back at the beginning of all this when looking for a problem with the vacuum advance. I had it down to where you could see the place the armiture from the vacuum advance attaches. And I could not see any corrosion or damage. I'll look at it to see how much deeper I can go without damaging it. Perhaps I will test measure the dwell on our 944 (88) and see what it does. Different system, of course, but it would be interesting to see if it drops as engine speed increases.

 

Just tested the dwell angle on the 944 for reference. It behaved almost exactly as the 928's. It's dwell at idle was a bit higher, at about 40 degress. But as the engine rev'ed, it dropped. And at high RPM, it was in the low 20's just like on the 928.

So...not conclusive, of course, since it is a different system. But I think I am going to put the dwell angle question on the back burner and focus just on the fuel delivery. I have two old pumps laying around, so am going to try to develop a bench test for pump pressure.

 

You were describing a sudden drop in dwell angle, which didn't sound right.

If you have a nut connecting the line to the fuel pump as in the pic below, the gauge might just screw on in place of the nut. You will still have to block the return lines. Thinking about it a bit more though, if the filter was a bit plugged, you would still get the same ultimate pressure at the rail, a partially plugged filter would only create a pressure drop while there was flow. So, not much point in trying to connect the gauge elsewhere.

 

I'm sorry if I am not describing the dwell behavior well. It did seem to drop off (to low 20's) faster at 4K and above, and fluctuations in dwell angle seemed related to the stumbling at just above 4K (when full throttle was just over 4K). So maybe there is still an issue there. But at least seeing what the 944 does showed me that dwell angle decreasing with engine speed is not an unexpected behavior in general. I thought it was supposed to increase with engine speed, based on the numbers in the ignition troubleshooting procedure, and that therefore I had an obvious problem.

I managed to bench test my fuel pump. Using hoses and clamps, I was able to connect the pressure gauge to the outlet. And when I powered it on, the static pressure went off the scale high....more the 100 psig. In fact, it went so high that one of the hoses I was using (which was supposedly a high pressure fuel injection hose) burst. So, I think my pump is good.

I can't really explain why the static pressure was only about 34 PSIG yesterday when I plugged the return lines, since today's bench test was so much higher. Other than perhaps that I really didn't plug the lines completely (I was attempting to clamp the return hoses coming off of the regulators, and there just wasn't much room to work and it was difficult).

Since I had to drain the tank to take the pump off, I took this opportunity to pull the tank off so I can fix a leak on that hose that runs from the top of the tank to the filler tube. (Been putting this off.) So I need to finish that before I can run it again and resume troubleshooting. Also I have ordered a new fuel filter, and that should come by the end of the week.

I have a question though about a part in line between the tank and the pump. Diagrams/schematics I have seen do not show this part. It looks like a small filter or something. There is no part number on it. It looks like it was put in by a PM, based on the jagged edges of the hoses attached to it and the non-Porsche hose clamps. Anyone recognize this? Is it suppose to be there? Or is the pump supposed to connect directly to the outlet on the tank?

Thanks

 

It looks like someone has added an inline filter before the pump. I've always wondered why the filter is after the pump but assumed it made sense, just not to me. I mean, who the hell am I , to question a German engineer?

 

There is supposed to be a mesh strainer with a rubber seal screwed to the tank, then a molded hose that connects directly to the fuel pump.

That could easily contribute to fuel starvation if its somewhat clogged, which is a pretty good bet, since somebody installed it there presumeably to take care of a problem, probably repeated clogged pump due to tank debris.

I've had three cars over the years that suffered severe power loss because of clogged fuel filters. Two GMs and the 84 5spd. The 84 spd and my Jeep Wrangler had what appeared to be a layer of mud in their fuel tanks.

Tom, you might have found the problem. Could cause both lower rail pressure and lower flow rate.

 

The pump that went over 100psi sounds like it's for a mechanically injected car(CIS). Your system isn't designed to handle that kind of pressure. What are the markings on the pump?

If you have a high-pressure pump and you can't get over ~35psi at the rail by plugging both returns at the regulators at the same time, there is definitely a blockage somewhere. I'd suspect the suction side, maybe the too-small filter that was added. The diagram in the WSM indicates a hose clamp on the return lines at the regulators. Try getting ~2" lengths of hose and plugging them with bolts & hose clamps, then attaching them to the regulators with hose clamps. Or, if it's easier to reach the tee that connects the two regulators to the return line, plug it there.

The extra filter doesn't belong there. The in-tank strainer filters out chunks big enough to damage the pump, and the filter traps particles big enough to interfere with the injectors. You'll need a replacement hose. Before you mess with it though, remove the fuel level sender and peek into the gas tank with a flashlight. It's possible that filter was added because the tank was full of crud and/or the strainer is damaged.

BTW the WSM suggests an increasing dwell angle at high RPM. You said you are measuring dwell using your tachometer -- how exactly are you doing that?

 

There is only a Bosch number on the pump: 0 580 464 017. I believe this is the correct pump for model years 80-82. Equivalent to Porsche P/N 92860810401. (The higher pressure one would be 0 580 464 045?) This is correct, right?

Yup, the in tank fuel strainer is a mess. The screens are torn and loose. I bet that's what happened....a PM knew his strainer was bad, had difficulty getting it off to replace it, so added that small prefilter. Which is now bad. There is some crud in the very bottom of the tank, but not too bad. I think I can clean it out.

OK, so...in-tank strainer and rubber seal, pre-formed hose, and new fuel filter are all on order to arrive Thursday.

I'm still puzzled by the really high pressure generated by my pump in a bench test. But assuming that is indeed the right P/N for an 82, I don't think I'll buy a new one until I have put it all back together and can test fuel pressure at the rail with the new screen, filter, and sans pre-filter. And maybe I'll try plugging the return lines as suggested (clamp two plugged hoses to each regulator's output) and measuring rail pressure. That's a great idea on how to do it.

And if this turns out to be the problem (clog upstream of pump), I will also be puzzled by the good flow rate I got. But I'll wait to see what pressures I get to worry about it.

Regarding how I have been measure dwell angle...my tachomter (not the one in dash...a hand held workshop unit) has a setting on it for measuring dwell angle. Connect the leads across the coil's + and - terminals.

Thanks!

 

I'm also watching this thread because timing is next on my list.

 

"Regarding how I have been measure dwell angle...my tachomter (not the one in dash...a hand held workshop unit) has a setting on it for measuring dwell angle. Connect the leads across the coil's + and - terminals."

I have never seen a dwell meter work like that. The usual hook-up is one lead to the negative terminal on the coil, the other lead to chassis (engine) earth.

Decreasing dwell is not normal.

 

My money is on the fuel delivery !!!!
Get rid of that aftermarket filter and get this the correct large one 928-110-253-06.
Replace the strainer, unhook your lines and get a pressure can used for air conditioning flushing. Fill it with Tulane and blow back the lines it will tell you if a blockage is present. I went through this with my 5-year standing 82.
Don't forget you are cranking that pump wide open on the bench. You need to check at the fuel rail to get an accurate pressure.