I have a habit of keeping stuff that I have changed. I'm going to have a look tonight and hopefully I still have it and give it a go.
Printable View
So I couldn't find the switch last night. But this morning I checked it and it is in spec. I will still get another, hopefully in town, but I'm sure that's not the problem.
I completely isolated the cold start injector and it still started without the fuel pump. Add the pump and it floods.
There must be a dodgy injector somewhere. Vacuum is somehow drawing fuel into the fuel rail and when the injectors fire, there is sufficient fuel for it to run, albeit rich. To test this I removed the cap in the cold start supply line and after 15 seconds it stalled out and took a bit of cranking before it started again with the cap back in.
Short of replacing each injector until it goes away, I can't think of a way to isolate each injector to see which is faulty.
Any ideas of how else fuel is getting in?
There's also the fuel pressure regulator. Pull off its vacuum hose and look for leaking fuel. Internal failure of a pressure regulator is a common enough problem.
I have had this issue exactly on a previous RRC - fuel crossing the regulator diaphragm and being sucked back along the vacuum line. - unmetered fuel. That was on a 1992 3.9 - Aftermarket regulator with inbuilt fuel gauge and Bosch captiva injectors worked well on this model.
See I have already fitted a rising rate regulator. That was my first thought when the problem first occured. If I had a vacuum gauge I'd put it in the fuel line and see how much is slipping through.
Not sure a stuck open injector could draw enough fuel through the pump, with the pump not running, with just engine vacuum to make it run.
What’s the pressure in the fuel rail with everything connected when you try and start it?
From memory, pump on the early flapper system shouldn’t start till you crank the motor, and the flap in the afm moves. Checked the flap moves freely and isn’t stuck?? Blocked fuel return line??
How fast does the fuel pressure gauge drop after switch off? Is it accurate? Can you clamp off the individual injector hoses? Not sure if they're long enough. Also try running a temporary return line to the filler cap of the petrol tank or a jerry can, I've seen restricted return lines before today.
I can't accurately see the gauge when cranking to start. By the time I get to it, it's at about 25PSi and dropping. That pressure is in spec but it would be closer to 40PSi as that is where I set it.
The flap does move and it does activate the pump as it should, that's how I set the regulator. As for the gauge accuracy, not really sure.
The pressure (when activated via the flapper) will hold the 40PSi. You can hear it pushing past the diaphragm at this point. Once released, the pressure will drop of steadily until 20-25PSi then hold for approx 15 seconds, then drop right off. This is out of spec. It is supposed to hold a set pressure for a period (can't remember exactly right now).
Peter51 sent me a PM offering some other thoughts and unfortunately i don't have the oscilloscope hardware or a good enough multimeter to measure the injector duty cycle, BUT I did have a bit of a play. I disconnected all of the injectors. I measured 2.7 ohms at all but one injector (couldn't get to the other one), then tried to start it. It didn't want to run. With the pump relay in or out.
That was a positive as I now have a direction to go. The fuel must be getting drawn past the injectors, yes past the pump and all the way up to the manifold, there is no other way it can get there. This must be because the injectors are open all the time and not cycling correctly. It's the only way.
I tested the plugs for voltage and they all had 12v dropping off the about 8v. I'm under the understanding that they are supposed to be fired at either 3 or 5v, but only when triggered, so I shouldn't see a reading until they are fired.
Tomorrow I am going to carry out the AVO tests from the manual again to see if that can give me a better insite.
I'm hoping I don't have a buggered ECU.
99% sure i have an ECU here if its the problem