Car ('86 911) slowly dies... Help!

[rallen911]

My car has developed a problem this summer where it loses power and then slowly dies.

I ran it fairly low on fuel (dummy light was on), but then put a half
tank in. It didn't exhibit the problem until I was getting lower on fuel
again (dummy light was not on), so I thought maybe the fuel level indicator
might be inaccurate. I filled it up a few nights later (trip to the station had no problems), but it only needed ~17 gallons (wasn't that low). I had the same problem again the next morning.

Here are the conditions where the dying has happened:
1. Sitting at a stop light on a slight uphill incline, the engine rpm went down and then it died. I was able to start it right away, and continue home with one or two more similar instances (can't remember conditions) where it almost died, but it didn't.

2. Driving up a ~30 - 35 degree slope it lost power and died when I put the
clutch in. Would not restart right away, but after a couple minutes it
started, so I took it home and left it. This was after I had put the half tank of gas in after running it low. I filled it up that night, hoping that it was a problem with the fuel indicator, but it wasn't. (see above)

3. The next morning after filling it up completely, I made it up the same hill without incident, drove down a similar incline with no issue, and as I started up another incline (~25 degrees), it hesitated when I pressed the gas and then lost power and died when I depressed the clutch. As we were still moving, I popped the clutch, and it started again. I continued another 0.5 miles up and down similar hills to get my daughter dropped off.

4. It lost power again as we pulled into the lot of our destination. I depressed the clutch and tried to do another rolling start, and it did. At very low speed in first gear the car did not die until I depressed the clutch. I let it sit while I took my daughter inside, and it started right up when I returned. It made it home along the same route without a problem.

Since this time, I have replaced the fuel filter. I ran it for several minutes (~15) and went up and down the driveway without incident. I have not been confident enough to take it out for a spin. There really isn't ever a good time to go out and get stranded!

Any idea what's going on? Anything I can try fixing myself before I take it
in?

[rick-l]

You could hook up a fuel pressure gauge and jump out when it stalls and read the pressure.

[rallen911]

Yes. I just borrowed one from a friend at work, so that was going to be one of my next steps.

Like I said, though... I can't seem to find a good time to go and get stranded! :-)

[DRACO A5OG]

Hopefuly the fuel pump not damaged but probably got a load of junk stuck in the fuel filter.

Since you got the gauge, check pressure per Bentley then rig up a pump tester per Bentley and test the pump.

I bet it is the fuel filter, what you described is fuel starvation.

Jim

[uwanna]

Classic clogged fuel filter symptoms!

[rallen911]

Here is an update to my slowly dying 3.2 engine.

I was graciously offered the use of a fuel pressure tester with the correct fitting on it by one of the local PCA members, though I would still like to get my own just in case I need it again.

** Note that I haven't had time to go through the test procedure in the Bentley manual yet.

After hooking up the tester, I started the car and the pressure was right in the 35 psi range. The manual says that it should be around 29/30. The value I was getting seemed to be more in line with what it should be when the vacuum line is disconnected (going from memory here, so that could be wrong).

We also noticed that when the throttle was increased, that the pressure would drop down to the 30 range.

I decided that with the tester hooked up, it was a good time to take it for a drive to see if it would die, and then I could check the fuel pressure to see if that was the culprit. I started out slowly just driving through the neighborhood. After nothing happened on the 5th lap, I decided to step it up and went on some more hilly and faster roads, still no symptoms. I ended up driving rather spiritedly ( I missed driving it so much!) for about an hour without any symptoms showing up. I checked the reading on the tester the next morning, and it was quite a ways down like 5 or less. From what I have read it sounds like the system should remain pressurized, so that is concerning too.

It looks like the new fuel filter might have solved the dying problem, but I am still concerned about the pressure being 5 psi on the high side and that the system did not stay pressurized. Looks like it's time to go through the entire Bentley procedure.

I'll keep you posted.

[rallen911]

Does anyone have any feedback on what I've seen, so far?

[rick-l]

The car hasn't died again?

Are you sure of the fuel pressures? When you open the throttle the pressure should go up along with the manifold absolute pressure. I wonder what else besides the regulator would cause that?

It also sounds like there is no manifold vacuum on the regulator

[rick-l]

No one else? I wonder if scarceller does a search for his user name occasionally.

The fuel pressure regulator is supposed to maintain a constant pressure across the injector from the fuel rail to the manifold. With the car off, fuel pump jumpered to run (or the vacuum line disconnected), the manual says this pressure should be 36 psi gauge that is 36 psi above atmospheric pressure (14.7 psi absolute) the pressure in the manifold with the car not running or running with the regulator's connection to the manifold open to atmosphere.

At idle the manifold pressure is probably about -5 psi gauge or 10 psi absolute so the fuel pressure referenced to atmosphere (what you read on the gauge, the number in the manual) should be about 29-30.

Opening the throttle, increasing the manifold absolute pressure and having the pressure go down makes no sense.

Not maintaining pressure in the system after shutdown would not make the car stall.

[mysocal911]

Originally Posted by rick-l

No one else? I wonder if scarceller does a search for his user name occasionally.

The fuel pressure regulator is supposed to maintain a constant pressure across the injector from the fuel rail to the manifold. With the car off, fuel pump jumpered to run (or the vacuum line disconnected), the manual says this pressure should be 36 psi gauge that is 36 psi above atmospheric pressure (14.7 psi absolute) the pressure in the manifold with the car not running or running with the regulator's connection to the manifold open to atmosphere.

At idle the manifold pressure is probably about -5 psi gauge or 10 psi absolute so the fuel pressure referenced to atmosphere (what you read on the gauge, the number in the manual) should be about 29-30.

Opening the throttle, increasing the manifold absolute pressure and having the pressure go down makes no sense.

Not maintaining pressure in the system after shutdown would not make the car stall.

The decrease in pressure indicates that the fuel pump can't maintain the required
pressure given the increased fuel demand at the higher RPM. With the vacuum
line disconnected the pressure should not decrease with a little acceleration off-idle.
The increased fuel pressure with the vacuum line connected, i.e. during acceleration,
provides additional enrichment under load, i.e. manifold vacuum decreases causing
enrichment which basically indicates an additional engine load.

The Bosch D-Jetronic system used manifold vacuum as a load indicator which caused
the injector pulse width to increase as manifold vacuum decreased. That system
relied totally on manifold vacuum for fuel control, as there was no AFM, MAF, or TPS.
Many fuel injection systems still used manifold vacuum via a MAP sensor as the only
engine load sensor thru the '90s.

[ischmitz]

Originally Posted by mysocal911

The decrease in pressure indicates that the fuel pump can't maintain the required
pressure given the increased fuel demand at the higher RPM. With the vacuum
line disconnected the pressure should not decrease with a little acceleration off-idle.
The increased fuel pressure with the vacuum line connected, i.e. during acceleration,
provides additional enrichment under load, i.e. manifold vacuum decreases causing
enrichment which basically indicates an additional engine load.

The Bosch D-Jetronic system used manifold vacuum as a load indicator which caused
the injector pulse width to increase as manifold vacuum decreased. That system
relied totally on manifold vacuum for fuel control, as there was no AFM, MAF, or TPS.
Many fuel injection systems still used manifold vacuum via a MAP sensor as the only
engine load sensor thru the '90s.

Loren, you making something very simple sounding very complicated: The fuel flow through an injector is proportional to the pressure difference between rail and intake manifold at a given injector opening time. It's as simple as that, once you open the throttle the pressure in the intake rises and thus you need to raise the fuel rail pressure at the same rate to maintain the same pressure difference

In addition some fueling algorithms use MAP pressure as a load signal - not the 3.2 DME though....

[scarceller]

Originally Posted by ischmitz

Loren, you making something very simple sounding very complicated: The fuel flow through an injector is proportional to the pressure difference between rail and intake manifold at a given injector opening time. It's as simple as that, once you open the throttle the pressure in the intake rises and thus you need to raise the fuel rail pressure at the same rate to maintain the same pressure difference

In addition some fueling algorithms use MAP pressure as a load signal - not the 3.2 DME though....

Nicely written: the pressure that matters is 'across the injector' that's why the intake vacuum is feed back to the Fuel Regulator.

Per specs:
- Engine not running but pump jumpered on 2.5bar (36psi)
- Fuel pressure at idle with vacuum line connected 2.0 bar (29psi)
- With vacuum line removed from regulator 2.3 to 2.7 bar (33-39psi)
- Pinch return line of regulator slowly with hose clamp, pressure should exceed 4bar.

Once engine is shutoff the pressure should be > 2.0bar and it should retain pressure > 1.0bar (14.5psi) after 20 minutes

You can also test pump delivery rate by routing the fuel from the test port to a container. Run pump for 30 seconds it should deliver > 0.85 liters (0.9qt)

But if you see pressure drop <2.0bar under load you most likely have a delivery rate issue from a bad pump or a clogged system.

[scarceller]

Also the fuel tank has a strainer in it at the bottom of the tank at the location near where the fuel line exists the tank. This strainer can also get clogged but you need to remove it from the bottom of the tank to really inspect it.

From our host's catalog you can see it here:
http://www.pelicanparts.com/catalog/SuperCat/0804/POR_0804_FULFUL_pg4.htm#item9
http://www.pelicanparts.com/cgi-bin/smart/more_info.cgi?command=show_parts_image&Headers_Id=38962&Line_Item_Number=25

And a few threads showing the part
Post Photos of your Old Fuel Tank Strainer
Torque Value for Gas Tank Strainer

[Drisump]

"Also the fuel tank has a strainer in it at the bottom of the tank at the location near where the fuel line exists the tank. This strainer can also get clogged but you need to remove it from the bottom of the tank to really inspect it." I had this problem with a Land cruiser, it would behave in a very similar way to what your symptoms are. I had to clean the intake screen, which had developed a very tough film, with brake cleaner. Good luck