5.0L TBI running rich.

[Chris Dauth]

This one has got me stumped !

First up a bit of history. This is a new boat to our marina, a 2005 Champion 925 with twin 5.0L TBI Mercs driving B1’s Raw water cooled, and it’s the owners first “big” boat. It was shipped 1000 miles up here by road. The owner did have a sea trial and “it seemed to go OK”. Past history is lacking.

We are a small boating community and between us have bought most of the Mercruiser tools, including a Rinda scan tool. A “real” Mercruiser dealer with techs are 500 miles away !

As I’m the “old man” of the marina, the manager asked me if I would “take a look” and see what is going on. Well, we started the port motor up, and it is rich, I mean heaps of black smoke and within about 30 seconds there was ¼ “ thick soot floating around the back of the boat !

My first pick was the ECT sensor (engine coolant temperature) which will tell the ECU (MEFI 3) the motor is cold and therefore give her more fuel. So I grabbed the scan tool and it showed the correct temp and we could see it change as the motor warmed up. It settled out at 155 F. Also, on the scan tool, there were no codes set and all other engine parameters were within spec., almost spot on per the Mercruiser manual # 24 which covers the 305 and 350 TBI & MPI.

Over the course of 2 days, we have changed all the sensors from the starboard engine (which runs perfectly) and the fuel filter. The IAC and the TP sensor have been changed as well, and there has been no change.

We pulled the two injectors and took them up the road to have them tested by the local fuel injection specialist. So, this is what we are up against, the specialist tells us, “never seen anything like them before in my life.” We picked them up next day, nothing wrong with them, no charge ! I’m wondering if anything was done ??

My thoughts are now, as the marine electronic fuel injection (MEFI) is open loop, no oxygen sensor, the extra fuel getting in is coming in thru the worn injectors or the wrong injectors. Our next move is to swap the injectors out of the good starboard motor. The only thing we haven’t checked is the 30psi fuel pressure. We have to make up a gauge to do that.

Anybody have this problem ??
Is there somthing I have missed ??

Oh for a 4 barrel carb.

 

[Fun Times]

You may have a bad MAP sensor (Manifold Absolute Pressure Sensor).

I remember I once had a small block TBI engine idling really rich with a lot of smoke coming out of the exhaust area and the spark plugs where really rich looking.

It turned out to be the MAP sensors top cover of the sensor was lifting up off of the base of the sensor and making the engine run really rich and rough with lots of dark smoke, and if I pushed down on the top of sensor the engine would run normal again.

So you may want to try switching the MAP sensors between the two engines for a test.

For fuel pressure testing the TBI engine, This link may help you out a little.:huh::smt001

http://www.marineengine.com/boat-forum/showthread.php?384521-fuel-pressure-test-port

I wish you good luck with it.:thumbsup:

 

[Chris Dauth]

Yep, we changed over the MAP sensor.

I can handle the pressure test OK. I have a gauge for the 454 MPI, but I'll have to make up a fitting to suit the TBI system. There is no test point on the TBI whereas the 454 has the schrader valve on the front of the block.

We also checked the vacuum line from the pressure regulator on the fuel cooler, and that is OK. I was hopeing to find a ruptured diaphram but no such luck.

 

[JediJD]

Just musing...do we know that the base engine operation is solid...steady vacuum at smooth idle...similar numbers between both engines? Did the plugs look like incomplete combustion? Just so I understand, you swapped the injectors between engines too?

If so, full rich has to be the ECU thinking that the engine is ice cold or WOT. If the ECU is seeing normal inputs, and base engine is good, and all the outputs have been exchanged, it has to be fuel delivery.

 

[Jersey Mike]

This one has got me stumped !

Anybody have this problem ??
Is there somthing I have missed ??


You've done really good so far! Check fuel pressure next, check for fuel dilution of the crankcase oil (from running so rich for so long. Don't forget about basic ignition and fouled plugs after running so rich for so long.

 

[Chris Dauth]

I went through the manual again tonight (Friday) and checked off all the tests we have done. Only ones left are the fuel pressure and changing out the injectors.

At the end of every test procedure and the problem still exists, they say to replace the ECU. I think they're about $AU3000 here, so I suppose I had better swap that over too.

Some one is not going to be a "Happy Chappy" if it is the ECU.

I'm still hoping I have missed something.

BTW, if I disconnect either injector, the engine settles right down, idles nice and smooth and no smoke. At first the motor almost dies but the IAC brings it back to 650 RPM. It's as though the injectors are pumping twice as much fuel that the motor needs

 

[Chris Dauth]

I took an oil sample today, and it was as black as ink and about thin as diesel, so we pumped the lot out and topped her off with fresh.

Fortunately the motor has not been under any load but who knows what as been going on with cam followers etc.

I have an idea the owner took it out for a run before I got to it. I would hate to think he'll need a rebuild just because of a crook fuel delivery system.

Time will tell.

 

[JediJD]

It absolutely doesn't sound like ECM...the ECM is doing what its supposed to do if it is responding to inputs and providing outputs. The question is, why do we have incomplete combustion? It also doesn't sound like injectors...the odds of both injectors failing at once are ridiculous and disconnecting either one reduces fuel equally. I doubt its a traditional fuel pressure problem either...higher fuel pressure would increase fuel volume, not lower pressure.

Please forgive me if this is stupid...I'm just not there and not sure what we are working with. Is there any crankcase recirculation system in place (such as PCV) or EGR?

 

[Chris Dauth]

Yep, this motor has PCV and EGR, and it was a thought I had 2 days ago because when you look at the fuel flowing onto the throttle plates on both motors, they look about the same. My thoughts then were that you should be able to see a difference, I mean, with the amount of black smoke, you would think that port motor was sucking fuel by the bucket full !

Then, also, the pulses on the injectors are at 2.4 milliseconds, and the manual says that is in spec at idle, 2 to 3 milliseconds.

What you say makes a lot of sence, could be worth a check. Don't PCV valves get stuck and clogged on autos ???

I see the EGR line comes from the front of the starboard exhaust manifold and disappears some where in the back of the inlet manifold.

A few extra brains on this problem is sure a help :thumbsup:

 

[fc3]

The Rinda tester should also show fuel consumption. If the ECU "believes" that it needs to dump more fuel, it should show far more than about a gallon per hour of consumption at idle. If it shows a reasonable number, then it's probably the injectors.

EDIT: Never mind. You posted at the same time I was writing this. 2.4 ms indicates that the MEFI-3 module is computing the correct A/F ratio and that all of the sensor readings and the lookup-tables in the MEFI unit are correct. Now you just need to confirm that 2.4 ms pulses are what's going out to the injectors. If so, then the injectors are bad.

Personally, I'd put an oscilloscope on the injector's wiring. You can tell if the pulse duration is correct with the 'scope. If the trace is good, then it's the injector. If the trace is bad, then it's the ECU. The MEFI-3 ECU is not available any more. You have to convert to the current system which requires new harnesses. I don't know if that works with TBI.

Best regards,
Frank

 

[Chris Dauth]

I had similar thoughts about the ECU, it could be calculating 2.4ms pulses and showing that on the scan tool but if the output transistors were going bad, the pulse on the injector could be greater. Then, like I said above, I thought I would be able to see the extra fuel going in.

I'm starting to think the problem is outside the electronics, like maybe the PCV or EGR, as I have swapped all the sensors. And I have still that spare ECU sitting on the starboard motor :thumbsup:

Since the PCV, EGR has been mentioned, I checked the manuals and they don't get a mention. Manual # 24 covers TBI and MPI up to 2001 and the next one I have is #31 which covers MPI only and 2005 and later, and no mention of PCV or EGR. So I guess there is a suppliment some where. BUMMER !

BTW, I have a dual trace Techtronix in the workshop.

 

[fc3]

Cool. I would go with the scope test next since that's got to be easier than swapping the TBI.

I have a Tektronix scope too.

This one is from 1968. It's called a mainframe 'scope. LOL! It works fine, though.

But mostly I use the HP. Much more portable.

 

[Chris Dauth]

It's Saturday morning, and me and the Admiral are taking Solitude for a 40 mile run up the Western side of Fraser Island to go whale watching, so I won't get back here 'till Monday morning, that will be Sunday afternoon your time.

Anyway, I'll chew all this over while I'm away and make another start next week.

Frank, I remember those "Mainframes" well, we had a collection of plug in modules, vertical amps and time bases, must be getting old, 1968 ??? BTW, I got my ham radio licence in 1963, VK4OO, ..._ _._ ...._ --- --- :thumbsup: (still current)

Guys, Have a great weekend, thanks so much for the input, I'll be back soon.

 

[Chris Dauth]

Ok, I'm back 7:30 Sunday night ! Had a great trip :thumbsup:

So, here is my thinking, heaps of black smoke = too much fuel OR not enough oxygen.

If we assume the electronics are OK and the injectors are OK. And there is no way of measuring the O2 in the exhaust as the system is open loop with no O2 sensor.

What happens if there is an unrestricted flow of exhaust gas up the EGR line from the exhaust manifold to the inlet manifold ???

Here is what I figured out, could be way off track here, you experts might like to put in 2c worth and tell me what is missing.

It seems as though the system comes into balance but the system doesn't know the O2 is missing. I would imagine the lookup tables in the ECU take account that the normal air in the system would contain 25% oxygen but in this case the inlet manifold has a fair bit of exhaust gas in there and the IAC opens just enough to let in enough fresh air to keep the engine at 650 rpm, even though it is as rich as all hell.

Years ago, I had a car with EGR and the EGR valve was stuck open with gunk. It was vacuum operated, hi vacuum shut the valve and only when the throttle was opened did the valve open. I don't know if this Merc. has an EGR valve, but I'll look tomorrow.

UPDATE:-
I have been playing around with Google and EGR, and found this at http://www.wisegeek.com/how-can-i-tell-if-my-egr-valve-needs-to-be-cleaned-or-replaced.htm


If the EGR valve is stuck open, it will essentially cause a vacuum leak, leading to inefficient combustion, rough idling, hesitation, and sometimes stalling in extreme cases. This is because the car cannot combust on carbon dioxide from the tailpipe — it needs atmospheric oxygen. If the combustion chamber is flooded with exhaust from an open EGR valve, it will not function properly. To check and see if the EGR valve is stuck open, have someone idle a parked vehicle with the brake on while you examine the plunger shaft to see if it is stuck open.

A big thank you for JediJD, could have put me on the right track !!

 

[JediJD]

In an automobile, EGR is designed simply to cool the combusion a little to reduce the formation of oxides of nitrogen. If I recall correctly, the goal was seven percent exhaust added to the air/fuel mixture at cruise (and not at cold start). The EGR valve sat on top of the exhaust manifold and channeled directly to the intake manifold. When it failed, the pintle in the exhaust either froze up, or the diaphragm in the unit failed. Standard check was propane (or carb cleaner) around the EGR valve with and without manifold vacuum, to see if there was an idle change. If so, replace the EGR valve. It could be the problem, but the more I think about it, it shouldn't make anything run full rich...even in open loop. Use a hand held vacuum generator and move it up and down with some carb spray and see if you have substantial change in engine idle. With the piintle UP you should have some idle change (because you don't normally have EGR at idle). But down, if the carb spray makes a difference, you have a leak, or up and it makes no difference, it is stuck.

PCV, positive crankcase ventilation, took the realtively lower vacuum from the air cleaner cover and used it to try and circulate the fumes from the oil pan area (the crankcase). It was an emissions device only to the extent that without it, the fumes would escape unchecked into the atmosphere. Also could be a mild vacuum problem, but not a rich mixture issue.

If I recall correctly, your TBI has a fuel pressure regulator built into the throttle body. If I was a betting man, I'd bet that is where your problem is. Fuel delivery problems, think fuel...if you hear the sound of hoofbeats, think horses, not zebras.

 

[Chris Dauth]

That regulator in the Throttle Body is a pressure dampner, suppose to take pressure spikes out of the system. There is a little vent tube coming out of it, but with the motor running, there is no fuel coming out of the vent.

The main pressure regulator is down on the fuel pump / fuel cooler, the one I have yet to check.

I think you nailed it with the EGR valve, like I said, the fuel flow onto the throttle plates of both motors appears to be exactly the same, but that is only by eyeball.

BTW, did you see my update above, or did we post at the same time ???

Anyway, some more tests tomorrow.

 

[BonBini]

Don't think it's an EGR deal. If it were I would think as you revved the engine it would smooth out.

From reading through all your troubleshooting: You've confirmed Timing, sufficient air and combustion when you disabled one fuel nozzle and the engine's response was to smooth out....

Defiantly getting more fuel than expected via fuel nozzles. Now if they are not being left on too long (duty cycle), they must be flowing more fuel due to pressure.

Can't recall for the moment the exact operation of the fuel pressure regulator; maybe vacuum sensed for boosted fuel during cranking? Also perhaps a return line restriction! I do have an old corvette hot rod fuel systems book somewhere. I do recall when changing the fuel pressure to a GM TBI system in open loop, this is how to tune the fuel ratio after a performance build!

One more thing. Gm fuel systems always run in redundancy. meaning if you unplug one sensor, the system will use another to get things close. Allot of times when you unplug the bad sensor the engine will straighten up some. ie: unplugging a bad map sensor (telling the ECU the engine is under heavy load), the ECU will use the mass air flow, throttle position and rpms to get things close. This can sometimes be a quick troubleshooting aid. Keep up posted.

 

[JediJD]

Yes, sorry, I posted about when you did. Certainly could be EGR, and I hope it is, but I am with Bon Bini...if it smooths out off idle, it shouldn't be EGR.

Having said that, I have worked on engines long enough to know that sometimes the only solution to a weird drivability issue is ritual goat sacrifice (in the US we had some practical problems with the traditional virgins and volcanoes model). If EGR and FPR don't solve it...I'll bring the barbecue sauce.

 

[Chris Dauth]

I think tomorrow, I'll check the fuel pressure and then blank off the EGR line and see what happens. At least that will take a few variables out of the equation.

One good thing is that there is a perfectly good motor on the starboard side, with plenty of good spares.

BTW, We call that solution a "Greek Stock Take" over here, just drop a match in the bilge, problem solved !

 

[Chris Dauth]

Ron,
These Delphi MEFI ECU's are pretty primitive compared to the ones on the GM autos. They are lacking the O2 and Mass Air Flow sensors and you can't reprogram them without dumping the entire Eproms. Everything is in the lookup tables to the extent that the injectors are sized to suit each motor.

So if you have to replace a injector it must have the same Merc part number, eg. a $50 GM injector off eBay, won't work !

I think the newer PCM 555 ECU's don't have the problems the MEFI's 1,2, and 3's had. The 555's have overspeed limiter, and you can program idle speed etc.