Our '94 was leaking fuel into the engine compartment last year (2/15) so I tore it apart, bought 8 "new" OEM fuel injector O-rings and installed them. This cured the leakage into the engine compartment, and we still do not smell raw fuel under the hood. However, two thousand miles later (111893 miles) we have noticed pig rich exhaust fumes on cold starts, and yesterday the engine would not catch at all.

I ran the "hold the accelerator pedal to the floor to shut off the fuel injection and run the starter" routine, which was finally successful, with clouds of exhaust smoke (with oil) eventually tapering off and smooth running to follow. But now I'm thinking maybe those "new" (probably 22-year old, although sealed in plastic bags) BUNA-N rubber O-rings have once again "relaxed" and the lower O-rings are allowing gasoline past them into the rotor chambers.

Thing is, I don't know if it's the O-rings or the fuel injectors (stock originals). Either case, of course, results in tearing the engine compartment up again.

I do have a set of Viton O-rings, but they are a tad smaller in uncompressed OD than the OEM parts. Although Viton is great in retaining its compressive force over time and in the presence of petroleum products, I don't know if the initial oversize relative to the fuel distributor bores would be sufficient.

Any advice would be appreciated, thanks!

 

Sounds like a leaking injector and not from o rings I would have the injectors sent out and cleaned and flow tested

 

Yeah, maybe replace them with some cleaned and flow tested low mile one. If you pull the rails off and prime the system, you can tell where the leaks is coming from.

 

Measured ODs on Leaking O-rings Feb. 2015

Primary Injectors
Upper O-rings: 21.47 mm and 21.31 mm
Fuel Distributor Bore: 22.05 mm
Lower O-rings: 13.53 mm and 13.14 mm
Fuel Distributor Bore: (not measured)

Secondary Injectors
Upper O-rings: 21.97 mm and 21.97 mm
Fuel Distributor Bore: 22.05 mm
Lower O-rings: 13.35 mm and 13.35 mm
Fuel Distributor Bore: (not measured)

Measured ODs on "new OEM O-rings"
"Free" ODs
Upper O-rings P/N NF01-13-ZG4: 22.4 mm (all 4)
Lower O-rings P/N NF01-13-ZG3: 14.43 mm
Installed ODs
Upper O-rings: 22.75 mm (all 4)
Lower O-rings: 14.64 mm (all 4)

Changes from "free" OD to "Installed" OD for the new OEM parts were:
Upper O-rings: 22.75 - 22.4 = 0.35 mm
Lower O-rings: 14.64-14.43 = 0.21 mm

which gives an interference fit for the new OEM parts of:
Upper O-rings: 22.75-22.05 = 0.7 mm (diameter)
Lower O-rings: (unknown since lower distributor bores were unmeasured)

Viton O-rings (unused)
"Free" ODs
Upper O-rings: 22.0 mm (all 4)
Lower O-rings: 13.9 mm (all 4)

Since no Viton O-rings were installed, I have no data on installed ODs.

If Viton O-rings had been installed, and the upper O-rings' ODs expanded the same as the OEM O-rings:
Upper O-rings (installed): 22.0 + 0.35 = 22.35 mm OD
Interference: 22.35 - 22.05 = 0.30 mm (diameter)
Lower O-rings: 13.9 + 0.21 = 14.11 mm OD (Probably not large enough to guarantee a seal)

Any comments? Thanks for reading!

 

Thanks to both for your comments. If I choose to test by priming the system with rails lifted, how do I keep gasoline from spraying everywhere? Do you know a good physical test setup that prevents that?

 

When I tested my rails for leaks on the weekend I just put a handful of paper towel around them and primed. You could use a small container and set the primary rail in it and wrap the secondaries in towels.

Just a thought
Mike

 

When I did mine a couple years back, I set them in a rectangler plastic container. One of my secondary had a pinched O-ring from PO, so yea gas was filling up the container pretty fast.

 

MSilk and ZE Power MX6, could you tell whether the leaks were coming from the injectors or the O-rings by watching the rails during the tests? (Looks to me like that might be hard to do.)

Thanks for your replies.

 

I was lucky and had no leaks, I had just got the injectors cleaned and all new o-rings so i was testing to make sure there were no leaks. What about getting someone in the car to turn the ignition on while you watch for a leak, wear goggles of course. That way if there is a leak you can catch it right away without pouring fuel all over the place.

The other thing you could try before pulling everything apart is putting a fuel pressure gauge inline and seeing if there is a fast pressure drop once you cycle the key (watch to see if it goes up to spec, and how long it stays there) per the FSM. If there is a fast drop just proceed with caution once you test the rails for leaks.

Good Luck

 

i did this with mine, and its not easy to get the rails in a convenient spot, its gangly. but once you do, all it needs is the key on an then jump the fuel pump terminal in the diagnostic box.

putting the rails in some clear container is totally fine too. with the f/p terminal jumped, the fuel pump will just run, and you can see if anything drips.

 

As j9fd3s mentioned, you can prime the system from the diagnostic box while watching the injectors at the same time. You should be able to tell if the gas is coming out from the side or in the middle, but if gas come out of the rail then I would just send them out for service.

 

Right, I'm familiar with the Diagnostic Connector. Jump F/P to GND, then have someone turn the ignition key to the "Run" position, and fuel pump will turn on. In any case, car will be down for at least a couple of weeks; if the injectors are leaking there is that turnaround for having them cleaned and flow-tested; if it's the O-rings, will have to order those... it's a bummer that the O-rings I put in last year will have to be replaced again. Not sure I will trust new O-rings if that is the only problem.

 

I usually just leave the key in, then control the pump through the diagnostic box. I would just send them in for service in general, 112k is due for some cleaning, plus they install new O-ring as part of the service.

 

OK, I ran the "Fuel Pressure Hold Inspection" test shown on page F-96 of my '94 Workshop Manual.

It says that after turning on the fuel pump for 10 seconds, turn it off and monitor the fuel pressure for 5 minutes. At the end of that period the pressure should be above 21 psi. And it says if the test fails, the problem is either the fuel pump hold pressure, the fuel pressure regulator, or fuel injector leakage.

So fuel pressure shot up to 42 psi instantly when the ignition key was turned, and stayed there. Turning off the key began dropping pressure down to about 30 psi over 3 minutes, and then the pressure stayed at 30 psi. At 10 minutes, pressure was 29 psi. So I'm guessing there is no leak at the fuel rail from either injectors or O-rings.

However, the manual on page F-97 says the pressure should be 36-38 psi. Is 42 psi too high? I have not yet started the engine since testing per the F-96 procedure.

Can I believe the results of this test show no injector problem? Maybe a grain of sand holding an injector open temporarily caused the flooding?

 

Does the car only do this during cold starts?
Have you upgraded the fuel pump? (I am thinking for the higher pressure?)
When did you change plugs last?

Some small piece of grit may have got stuck but it would have to be extremely small. If anything it could be some fuel varnish causing an injector to stick especially if the car sits for extended periods of time and the fuel starts to go bad.

With the car passing the fuel leak down test I don't think what ever was leaking is still leaking. Might be another issue, you could try cleaning the IAC, and checking TPS voltages (I am thinking that maybe the car is not getting enough air for a cold start so it floods)

I hope this helps and I am not steering you in the wrong direction.
Mike

 

Yes, only on cold starts. No, fuel pump is stock. Plugs were last changed at 90,199 miles, 2/05/2006. (They are platinum NKGs, stock recommended heat ranges.)

Car is driven 2K miles per year or so, strictly street use. The IAC does OK controlling the idle speed both during fast idle and when warm. I've set it for 700+ rpm warm. Have checked TPS voltages (a long time ago) and they were all in spec. The air filter was changed about 5000 miles ago. The air intake system is OEM.

 

I would pull the plugs and take a look at them, 22,000 miles seems like a lot on one set of plugs. I change mine every three years and put the same mileage on my car (I pulled them last year and they were fine so popped em back in)
If you pull your plugs check the spark plug leads too, just to make sure nothing has chaffed though ect.

I am starting to run out of ideas lol if it is slow at work today I will do some more head scratching

Mike

 

Mike, no worries. Keep in mind that the car is now starting and running just fine both cold and hot. Still is very rich at cold startup (going by the exhaust smell) but no hesitation to fire up quickly. I will probably change plugs at around 120K miles (Mazda recommendation, which I have always followed).