Merc 350 MAG MPI fuel rail/injector removal and inspection

stephenm27

Active Member
Nov 3, 2009
526
Milton, GA
Boat Info
2021 Jeanneau NC 1095
Engines
2 x Yamaha F300
I am diagnosing a general problem with my starboard engine (intake pop/backfire under increasing load). I have replaced a number of likely culprits over the off-season but to no success so far. I have now also verified there are no error codes in the ECM and live data captured appears "normal" as compared to the port engine. So I'm now moving my focus to failures or conditions that wouldn't be detected by the computer and its sensors.

The question I currently have is, is the dummy fuel regulator on the fuel rail able to cause fuel delivery issues? I noticed tonight that the rubber vacuum line to it was nearly blocked due to being bent at 90 degrees under a wiring harness. This vacuum line originates on the intake manifold and is teed to this dummy regulator as well as the CF3-mounted regulator.

Would starving the dummy of vacuum prevent it from properly adjusting fuel flow under increasing RPM/load?

Thanks in advance for insight/experience...
 
In a word, no.

That regulator on the rail is a leftover from the automotive application of this motor. From what I understand, it doesn’t do anything, and doesn’t control fuel pressure in any way.
 
In a word, no.

That regulator on the rail is a leftover from the automotive application of this motor. From what I understand, it doesn’t do anything, and doesn’t control fuel pressure in any way.

That is my understanding from reading as well, but when I look at the fuel rail, it appears to be connected in such a way that fuel would need to at least flow through it?

So the question becomes, is the "not used" aspect simply because it's downstream of another regulator which is doing the regulating or is it because fuel doesn't even need to flow through it--but around it in the fuel circuit?

It's relevant mainly because this engine had the CF3 paint peeling issue in the fuel cooler. I'm trying to determine if I should examine this regulator for flow restriction.
 
Fuel doesn’t flow through that regulator, as it’s at a “dead end”. It’s located at the end of the fuel rail with no return line.

In an automotive application, with the regulator located at the end of a fuel rail, there would be A return line attached to the regulator to allow excess pressure to bleed off and be returned to the tank

OR

The regulator could be located at the point where pressurized fuel enters the fuel rail, thus only allowing 43psi (or whatever it’s set for) to enter the fuel rail. No return line would be required with this second setup.

If you have the paint issue, you need to do a complete cleaning of the regulator in The CF3 module, fuel line up to the fuel rail, fuel rail and injectors also need to be cleaned.
 
...If you have the paint issue, you need to do a complete cleaning of the regulator in The CF3 module, fuel line up to the fuel rail, fuel rail and injectors also need to be cleaned.

Thanks for clarifying. I've looked again at the parts diagram and can understand how it's dead-ended on this design.

In my case, the CF3 module cap and regulator has already been replaced along with all fuel filters. I was optimistic that perhaps the rail and injectors might have escaped without seeing contamination. I'm not sure I'll be that lucky...I was on the port engine but perhaps not this one.
 
Unfortunately the paint issue occurs after the filter. So any paint chips flow through the fuel line and the next “filter” that catches them is the screen inside the injectors.
 
Unfortunately the paint issue occurs after the filter. So any paint chips flow through the fuel line and the next “filter” that catches them is the screen inside the injectors.
Sadly, I am very aware of this design. I have two (port, starboard) CF3 caps with peeled paint sitting on my workbench. It appears that I will now be removing the fuel rail and injectors for cleaning.

I would assume there is no consumer-level injector cleaning that can be done once they're removed? Best left to an automotive injector specialty shop in your experience?
 
I clean mine. Built a contraption to house the injector in the lid of a glass jar. wired a switch to give power to the injector, and finally a big can of cleaner, whose nozzle is attached to a brass sleeve I put over the "intake" of the injector.
I can either open the injector and spray away, or build up pressure on the top, and then use the switch to observe the spray pattern going into the jar. easy enough to flip it over and spray backwards, blowing any junk off the intake screen.
takes longer to get the intake and rail off than to clean them, and I've got that down in spades. What I can't do is measure pressure, or spray, like a pro shop would be tooled up for.
 
I clean mine. Built a contraption to house the injector in the lid of a glass jar. wired a switch to give power to the injector, and finally a big can of cleaner, whose nozzle is attached to a brass sleeve I put over the "intake" of the injector.
I can either open the injector and spray away, or build up pressure on the top, and then use the switch to observe the spray pattern going into the jar. easy enough to flip it over and spray backwards, blowing any junk off the intake screen.
takes longer to get the intake and rail off than to clean them, and I've got that down in spades. What I can't do is measure pressure, or spray, like a pro shop would be tooled up for.
Very helpful. This might be worth attempting. I have no evidence to suggest I would need to characterize pressure or spray pattern at this point so doing it here in the shop sounds doable.

Do I need to ensure injectors are re-installed to their original cylinder?
 
Not questioning the path your taking but it might want to pull your spark plugs to see if they give you any input and when you do that go ahead and do a compression test to eliminate any problems that could cause these same kind of problems with the head.
 
...do a compression test to eliminate any problems that could cause these same kind of problems with the head.

I like your thinking. It confirms another angle of troubleshooting that I'm likely to do as well (compression test). I've delayed doing it based only on the assumption that the simplest, most likely answer is fuel rail contamination given the CF3 issue this engine experienced. Being a fuel-rail newbie, I have put it off in favor of all the easier things to throw at the problem:
  • Replaced 2-year old plugs with new ones in late fall to be certain I didn't have any issues there. They all appeared clean/good coloring upon removal.
  • Plug wires replaced.
  • Rotor replaced.
  • Ignition coil has been replaced.
  • Tested with the anti-siphon valve in bypass to eliminate a fuel flow issue caused by it if failing.
  • Fuel sample taken from bottom of tank to evaluate for presence of water (none found).
  • Crank position sensor has been swapped with the port engine for next outing.
All of this work has been completed over the course of lake trips to test, make changes/replacements during the week in the garage, repeat. A slow, methodical process, but I have a detailed log that I record findings in. Unfortunately, I am unable to make the problem (intake pop) occur on the hose here at the house. I attribute that to the absence of a "load" that is required--like when pushing the boat up on plane. At 1500 RPM on the hose, no problems are detectable.

I think my next steps are:
  • Fuel rail disassembly and injector cleaning (diy-style).
  • Compression test.
All opinions/insight are certainly appreciated and welcomed...I'm nearing the boundaries of my experience.
 
I like your thinking. It confirms another angle of troubleshooting that I'm likely to do as well (compression test). I've delayed doing it based only on the assumption that the simplest, most likely answer is fuel rail contamination given the CF3 issue this engine experienced. Being a fuel-rail newbie, I have put it off in favor of all the easier things to throw at the problem:
  • Replaced 2-year old plugs with new ones in late fall to be certain I didn't have any issues there. They all appeared clean/good coloring upon removal.
  • Plug wires replaced.
  • Rotor replaced.
  • Ignition coil has been replaced.
  • Tested with the anti-siphon valve in bypass to eliminate a fuel flow issue caused by it if failing.
  • Fuel sample taken from bottom of tank to evaluate for presence of water (none found).
  • Crank position sensor has been swapped with the port engine for next outing.
All of this work has been completed over the course of lake trips to test, make changes/replacements during the week in the garage, repeat. A slow, methodical process, but I have a detailed log that I record findings in. Unfortunately, I am unable to make the problem (intake pop) occur on the hose here at the house. I attribute that to the absence of a "load" that is required--like when pushing the boat up on plane. At 1500 RPM on the hose, no problems are detectable.

I think my next steps are:
  • Fuel rail disassembly and injector cleaning (diy-style).
  • Compression test.
All opinions/insight are certainly appreciated and welcomed...I'm nearing the boundaries of my experience.
I don’t see the distributor cap being listed as replaced.

2 years ago I had Popping under load in my 350 mag.

A new cap and rotor resolved the issue.
 
I don’t see the distributor cap being listed as replaced.

Sorry, my oversight. The distributor cap was also replaced at the same time as the rotor. It's predecessor was only a couple of years old and didn't show evidence of an issue, but easier to replace and remove the possibility.

I did *not* replace the distributor itself, nor the camshaft sensor. I would hope a failing camshaft sensor would show up in some form/effect in the ECM data download and it didn't, so I moved it to the bottom of the list, so to speak.
 
backfire is conventionally ignition related. Double check that timing.

My somewhat limited understanding of this engine is that timing is entirely controlled by the computer. As (digital) throttle position is increased, the crank position sensor reports increased RPM, the manifold pressure sensor reports increased vacuum, and the computer begins adjusting spark advance.

Using the Rinda diagnostic software I was able to verify the spark advance on the starboard engine approximately mirrors that of the port so that exhausted my knowledge regarding timing adjustment.

However, I’ve read that a sticking intake valve may present similarly and the computer wouldn’t detect it. I assume a compression test would help identify this condition...?

Again, thanks to all for the continued input. It’s very helpful to hear ideas and learn from others’ experiences...and maybe my eventual success will help someone else.
 
So I’ve decided to clean the fuel rail and injectors first, before doing a compression test.

I have the fuel line depressurized and disconnected and all 4 rail attachment points (studs with nuts) removed. But, as nothing is ever easy, the rail assembly doesn’t want to budge. I feel a little wiggle on the one side, but am afraid to apply too much force.

Anyone done this have any clues or suggestions? Should I de-clip the rail from the injectors and work on removing the injectors individually? I’m sorta past the point of turning back :).
 
Update: rail and injectors begrudgingly came out. Paint chips found in rail upon initial emptying. More chips found in at least one injector inlet.

uc


uc
 
Last edited:
Clean, clean, clean. Then have the injectors serviced and screens changed.
 
Clean, clean, clean. Then have the injectors serviced and screens changed.
I have bench-flushed and ran the injectors today with cleaner. Visually, the spray looks good and consistent. Of course, I don’t have the equipment to precisely measure atomization or pulse width.

Surprisingly, this model of Bosch (the manufacturer based on my research) injector has no screen protecting the input so pulsing them using cleaner from output to input is the only way to clear micro-contaminants in the intake.

It’s really no wonder that *any* paint flake from the cooler has the ability to foul these things—there is NO filtration for any contaminants past the CF3 filter. Nothing.
 

Forum statistics

Threads
112,942
Messages
1,422,701
Members
60,927
Latest member
Jaguar65
Back
Top