F115 falls on its face when revved up

[tyson0317]

Guys - I have a 2002 Mercury 115 (effectively a Yamaha F115). EFI, 4-stroke, 4 cylinder.

The SELOC manual that I got for it, states something to the effect of, "Yamaha/Mercury do not release test specs for their sensors - good luck troubleshooting them."

My motor is having these symptoms:

Starts and idles without a problem.
When revved up, often dies as soon as it comes out of idle. When I feather the throttle really slowly, I can sometimes get it to rev up. With some recent troubleshooting, it seems to do more or less OK in my driveway. However, when taken on the lake and under load, the motor was having trouble exceeding 2000RPM.

- I've checked fuses.
- I checked fuel pressure at the rail - getting consistent ~35psi at idle and ~40PSI at WOT (when not in water).
- Got it fresh gas from an external tank, without much change.
- Changed the small fuel filter in the clear water separator housing.
- Changed plugs.
- Checked compression and leak-down.

Then I started looking into the electronics. Lack of OBD-style diagnostics, or data on how to test the components is hampering this effort greatly. Noticed that when I disconnected the Manifold Air Pressure sensor, the engine actually seemed to run much better. Figured that the computer, ignoring the MAP input and using TPS to guess at the fuel delivery was doing better at injecting, so figured that the MAP was bad. Replaced MAP sensor (week in the mail and $120) - virtually no change! Then tested the engine with MAP connected and TPS disconnected - again engine runs better. WTF?

One strange thing - when I disconnect the vacuum hose from the fuel pressure regulator, engine idle (which is usually smooth and good) starts running like crap. Plugging the hose with my thumb, returns it to normal. What does this mean? My understanding of MAP-based systems is that a vacuum leak into the manifold should not make the engine run rough. If anything, it should increase the revs, as the MAP ought to see the drop in vacuum in the manifold and the computer ought to balance it with more fuel. This is however, NOT happening. Certainly, fuel rail pressure is jumping up when I do this, but I don't think that it's a major problem (making things run a little more rich) because when plugging the hose with my thumb, the idle turns smooth again.

This tells me that one of the sensors required for the computer to do its job, is sending bad data. However, without an OBD scanner, or ohm-meter test procedures, I am at a loss on how to troubleshoot this without just semi-randomly swapping in sensors (which is going to get pricey).

Im in Seattle. Any ideas highly appreciated!

PS: one other and probably unrelated thing... I think my stator is going out. I would expect that when the motor is running, my voltage to be at 13-14v. It sits around 12-12.5v. I do not think that this is related, as I have a brand new battery and a charger connected to the system when doing all of this testing. But in the event that the computer is somehow wired through the stator...

[boscoe99]

When was the last time the injectors were cleaned?

The motor needs to go into a shop that has a well qualified marine mechanic.

[rodbolt17]

use the seloc in the outhouse.
buy the Yamaha manual.
every sensor is testable.
YDS will work on that engine although data is rather limited.

[fairdeal]

one other...unrelated thing... I think my stator is going out. I would expect that when the motor is running, my voltage to be at 13-14v. It sits around 12-12.5v.

How often does a stator actually fail?

given that its just copper wire wound into interesting shapes - I would think rarely.

Regulator/rectifiers on the other hand - heat sensitive solid state gizmos -

those I would expect to lead a short and unhappy life

[tyson0317]

Update:
Spent the day flushing injectors, ripping things apart and testing components. Suddenly, disconnected the water temp sensor and the engine started running PERFECTLY! Need to dip it in water to test under load, but the throttle works primo now!!

Assuming the sensor is bad? Anyone know what the computer does when the sensor is disconnected? Does it simply delete it from calculation? Or does it go into some sort of limp mode?

[99yam40]

need to find out if it was dragging the reference voltage down causing other reading to be bad also.

should be some tests that can be done on it,
if it is bad replace it.
I would not try running without it

[tyson0317]

For posterity:

Oddly enough, seems like my issue was caused by a bad water temp sensor. On car engines, water temp is used to do fine trim to fuel supply to account for the very minor change in tolerances due to metal expansion from heat. That and better vaporization of fuel in a hotter engine. The adjustment is usually to make mixture leaner with a hot engine and in totality the change is not very great. I would have never expected it to account for this much of a difference.

On this Yamaha however, the trim clearly affects acceleration enrichment and higher RPM operation.

With the sensor completely disconnected, the motor ran quite well in ~62F water. There was some roughness, but it was very minor and did not prevent me from revving, or taking the engine to 6000 rpm. I expect that when the replacement sensor gets here, this will be resolved.

One interesting thing to note (which was pleasantly surprising) is that the "limp map" on the computer is quite good! When disconnecting the MAP sensor, the computer automatically goes into limp mode and uses the TPS to take a rough guess at fuel supply. Usually, this makes a motor barely run; maybe put-put roughly to get you home. This Yamaha however, has a very good TPS base map and so long as you are gentle on the throttle (move it slowly to allow the fuel map to catch up to reality), the fuel supply is decent enough to get it running quite well through the entire RPM band. This is good to note if a sensor/electronic problem develops while far from home. Disconnecting the MAP and forcing the computer into TPS-Limp Mode seems to be a good temporary fix to get the boat to planing speed.

My next project on this thing is the stator - but for this I will make a separate post.