[KarimJGobbar]

Vehicle: Dodge Grand Caravan 1995 FWD MPFI 3.8L
Problem: Hard start when hot, stalling while driving.
History of the problem: Long one. Available upon request.
Observations: The fuel pressure does not hold with key off, it immediately drops to 0. But it holds much better when engine is cold and the return line is pinched either right after fuel rail or near the tank. Fuel pressure is close to normal when engine is running. Fuel pressure regulator passes the test with pulling out the vacuum line - the fuel pressure increases by 10 PSI. No smell of gas in the pulled vacuum line. No smell of gas near the fuel rail or fuel lines between the rail and the tank.
Several key on - key off cycles are required to start the engine when it is hot.
Fuel pump check valve isolation test shows acceptable results, fuel pressure stays at 25-30 PSI for a couple of hours
Deadhead fuel pressure test shows very high value - 100 PSI or above.
No OBD codes.
Replacements:
Fuel pump - 2 years ago;
Fuel pressure regulator - recently, few times;
Fuel injectors - recently;
Fuel filter - 2 years ago.
Hypothesis:
Honestly, at this point of time could think of very few: Geometry of fuel pressure regulator socket is violated. In support to this theory there is a fact that I had to put extra washer on the regulator to prevent an obvious fuel leak through it. By other hand, I was not even able to find any mention of such a thing like fuel pressure regulator socket.
Or there is something in the system (fuel pump?) that immediately destroys the regulator. The fact is that one of the FPRs I changed had distorted lower part.
What else could I measure/test to diagnose the issue? Any ideas and brainstorming will be highly appreciated.
Thank you,
Karim

 

[RalphP]

At this point, I'd bet the drainback valve in the PUMP has failed. I'd start with an inline valve (not knowing what size your fuel line is, I can't link to an exact, but my Dakota uses this one: Amazon.com: 8mm 5/16" Fuel Non Return One Way Check Valve Petrol Diesel Aluminium Alloy: Automotive ) near the regulator if possible; at the tank if not.

Let's see where it's draining from.

RwP

 

[KarimJGobbar]

Thanks, Ralph! Do you suggest to install the valve into the supply line?
If yes, then existing check valve in the pump is already doing its job - according to my fuel pump check valve isolation test I mentioned abobe: I plugged the gauge into the supply line near the fuel filter, primed the pump, turned it off, and pressure stayed.
If you meant return line, then I cannot figure out what role the valve can play there...

 

[RalphP]

In the supply line. So it drains down as soon as you turn the fuel pump off with the regulator attached, but not with it disconnected?

That could be an injector draining down, or the regulator draining.

As many regulators as you've been through, I'd look at the injectors next to see if one's leaking.

RwP

 

[KarimJGobbar]

Yes, exactly. It drains when the regulator and the rail with injectors are in play, and does not - when there is only fuel pump with its internal check valve in play.
Yes, to me also it sounds like it leaves only injectors and regulator suspected. But here is another test that I conducted: with all components attached and in working state I primed the fuel pump, pinched off the return line, turned off the fuel pump - and the fuel pressure holds! As I remove the pinch off pliers from return line, it drains immediately. To my understanding this excludes the injectors out of suspicion, am I right?

 

[RalphP]

Probably; that really sounds like the regulator's fried.

What brand are you using? I'd try a different brand, even if it's a lifetime warranty.

RwP

 

[KarimJGobbar]

The brand of regulators? I used two brands: Sorensen and Standard. Both - to no help. So regulator is out of the list, too, IMHO.
That's why I point my finger to regulator socket - a thing where the regulator is inserted and bolted to. Made of aluminum or whatever alloy, it looks pretty thin and suspicious. Its ears (or how do you call it - where the mounting bracket is hooked to) are slightly curved. But the thing is I never heard of such a thing like regulator socket nor read any post about it, not even sure if this term is correct...

 

[RalphP]

Not sure how the Caravan is set up; I'd bet it's part of the fuel rail. But I don't know how THAT would bleed down unless it was spraying fuel around.

RwP

 

[KarimJGobbar]

True, it is part of fuel rail, at the end (downstream) of it, right at the place where return line begins. No, it does not spray around, not even smell of gas near it, but bleeds internally, into the return line. This, to my understanding, follows from my test with pinching the return line. The main question is why it bleeds, the regulator itself is new and good. Is it indeed the socket?

 

[RalphP]

No, it'd still be the regulator.

Again, the socket only connects it to the fuel rail.

RwP

 

[KarimJGobbar]

But how it can be a regulator if two new ones were in place?
Yes, socket only connects, but what if the problem is right at the place of connection?

 

[RalphP]

Because the regulation is inside the regulator.

That's actually outside it.

And yes, it can be two new ones; most of today's parts are crap compared to the older OEM stuff.

Bill Freiberger wrote an editorial about it back when he was editor of Hot Rod Magazine, even.

I'd try a different brand.

RwP

 

[KarimJGobbar]

I absolutely agree with you about the quality of today's parts. And it, IMHO, holds true for everything we buy today: from food to software
I cannot replace the regulator right away, but, I hope, I will do it in few weeks.
Thank you very much for your brainstorming!