worf928

I've been helping, remotely, an '87 Auto owner sort out his failing-emissions issues. I finally got to actually look at the car this morning. On of the things I wanted to check was fuel pressure.

With my trusty FP gauge on a drilled and threaded cap nut ready to go, I cracked the fuel rail test port nut. Now, I've done this quite a few times, so, I thought I knew what to expect. But, rather than the quick splooge of fuel followed by a steady but short flow of fuel from the rail, I swear we got pressurized air coming out of the rail. You could hear air escaping from the slightly cracked-open nut along with a pressurized spray of fuel.

So, the question is - how exactly could the fuel rail get pressurized air in it? There were no evident fuel leaks front or rear.

I do have one clue that I will hold-off on for a little bit, so as to not bias any from-whole-cloth diagnoses.

Rennlist immortality awaits the first response with a logical fault analysis.

UKKid35

Ran out of fuel, sucked in air, more fuel in tank, pressurised air in lines?

worf928

Good guess Paul. Might be plausible. But this car didn't run out of fuel and had been driven for three hours on the highway just prior to cracking the rail nut. You would expect any air to get returned to the fuel tank after a while, right? Seems like something had to be continuously feeding air into the system...

worf928

A few more facts: fuel regulator and dampeners hold vacuum.

Garth S

In tank pump has failed or is partially blocking the suction - and the primary pump is doing its best to keep up: in doing so, is sucking some air into its suction due to a porous/failing tank outlet hose ....

This could be compounded if the carbon cannister system is plugged .... and the fuel tank would end up slightly sub-atmospheric ....

the flyin' scotsman

Is it feasible that air was intorduced into the system at one point and like the brake system has to be bled out? If so can you do a volume check of the gas thats being pumped now that you have the fuel system open?

worf928

Interesting. However, this is an '87 so it has no in-tank pump.

But, you are close-enough to the one clue I have:

The vacuum line to the fuel evaporation valve (black valve next to the coolant reservoir) was not connected to the valve.

So, with no in-tank pump on '87s and no vacuum to the fuel evap valve what is the mechanism by which the rails pressurize with air (and fuel?).

worf928

I suppose. But, I've cracked open plenty of 928 fuel systems and never, ever seen this before.

If everything is working properly - and if I understand the general operation correctly - the fuel vapor system draws air from the throttle-body and oil filler neck to add air back into the fuel tank via the canister to replace the volume of fuel used by the injectors so as to keep the tank from getting sucked into a tight ball by the fuel pump as fuel is used up.

The LH sends the electric valve near the crossover (connected to the double-y hose that goes to the t-body and neck) a pulsed signal that is dependent on engine load to meter the return air volume.

Or am I missing something here? Like what the vacuum-operated valve does?

worf928

More detail?

worf928

This is what makes some sense to me: The pump is drawing in air as well as fuel. But, wouldn't you expect the outlet hose to leak fuel when the pump isn't operating?

mark kibort

Thats an independant system. leaks in the diaphram could cause fuel to leak out , but not air to go in.

mk

John Speake

Hello Dave
It's difficult to work out how the tank venting could affect the fuel within the fuel lines themselves.

You could open the return pipe to the fuel tank and see if there is air there.....

If the feed to the pump is partially blocked, could this cause cavitation ?

worf928

One other clue: Above I said we were chasing emissions. The car has super low HC and CO and super-high NOx - with new 02 sensor, new high-flow cats (and no airpump connected to them) No intake leaks that I can find. Engine bay is one of the cleanest I've seen.

The engine is not running really hot, seems pretty normal ~190 degree 'elbow' temp.

Yet, the emissions numbers point to a very lean a/f ratio.

worf928

I don't know if you can get air bubbles from cavitation in the fuel pump. Works in water. Don't know about fuel though. Anyone?

John Speake

Check the volts on the O2 sensor with a voltmeter when running closed loop, check they oscillate from 0.2v to 0.9v or so.