A 240EFI troubleshooting saga...

Thought I'd post my recent troubleshooting experience in case it helps someone else. This is a series of writeups that I was posting as I worked my way through this problem.

-----
This engine was running perfectly just a few days ago. Premium non-ethanol fuel, steady diet of Quickleen, occasional Seafoam, etc.
Today, the engine starts instantly and idles smoothly, and runs smoothly up to about 2500 RPM. Anything above that and the engine starts running VERY rough. Lowering the throttle to idle fixes it every time, but you have to go all the way down to idle. Bringing the throttle down slowly keeps the engine running rough until you get all the way down to idle.

Whether you creep up to 2500, or go there quickly, it's the same. It's like there's a threshold right around 2500 that kicks in this problem, and once it's happening only going to idle fixes it.

It seemed to me at first startup this morning that there was more smoke than usual, but it cleared like normal within a minute or so of engine warmup. Don't know if that's related or not.

I inspected and cleaned all plugs (which were new this season, as usual). The engine may be due for another decarboning but otherwise the plugs look fine. Besides, I wouldn't expect their behavior to change based on RPM's.

More info....

I used a spark tester and confirmed that all six plugs are getting a spark. For each cylinder, I then watched the spark tester while my wife ramped up the RPM's. At RPM's below ~2500 all six plugs showed normal sparking. When the problem started, ALL SIX CYLINDERS had intermittent sparking.

The service manual describes how the 240EFI's ignition system wires sets of CDM's together. If one CDM fails, it can take cause others not to function properly. It appears that the maximum number of interdependent CDM's is three. I read everything and studied the schematics and I can't figure any way that a CDM can cause this problem on all six plugs at once.

I have two spare CDM's on hand, so as a test I went through and replaced all six CDM's, testing the boat in the water between each replacement. Never made any difference at all. The original CDM's are back in the boat.

Mercury Corporate suggested the stator might have a problem. They claimed there are two sets of CDM charging coils, a low speed set and a high speed set. However, when I looked in the manual to get the wire colors so I'd know what to test, I found that's true for the 210 carb version. The 240EFI version has only a single set of charging coils, so if the engine will run at any speed it pretty much proves the stator is OK.

Likewise for the trigger coils. There's no reason for them to be speed sensitive. If the engine will run at all, the trigger coils have to be OK.

I'm at a complete loss to explain what's going on. It sort of feels like an RPM limiter, but the manual explains that the ECU handles RPM limiting by first backing off the timing on a single cylinder and then stopping the spark on the same cylinder. It proceeds to additional cylinders as necessary until the RPM's come down below 7000 RPM. That's not what's happening here - ALL cylinders are having their spark affected simultaneously and the RPM's are only in the mid 2000's.

Even more info...

Thinking that the problem must be common to all cylinders, I checked the battery voltage. If the voltage is wrong, it could affect the ECU's operation. Voltage at the battery terminals with the engine off is 13.02 VDC. Voltage with the engine idling is 14.22 VDC. The Mercury service manual says the target voltage is 14.5 VDC so it's a little bit low, but no tolerance spec is provided so I don't know for sure.

I tried disconnecting the regulator outputs so the engine could run exclusively on the battery, in case the voltage output of the regulators was noisy. The engine will crank but won't start with the regulators disconnected. Reconnecting them permitted the engine to start instantly again.

I have also tested the two head temperature sensors. The port sensor, which reports temperature as resistance, showed 721 ohms which according to the service manual means around 90F - just about ambient temperature right now. The starboard sensor is a switch and at the moment it's showing "open" with the DMM leads going one way and 7 megohms with the leads reversed. I presume there's a diode in the sensor that explains the sensitivity to polarity.

SUCCESS!

After testing countless things suggested by various Mercury techs and other helpful people, and finding no answers, I finally decided to try my other brainstorm (mentioned earlier): Disconnecting the rev limiter.

I left the rev limiter's power (purple) and ground (black) leads connected. I disconnected the lead that runs to pin 3 (or C, depending upon where you look in the manual) of CDM #2 and the lead that runs to pin 2 (or B) of all six CDM's. This took the rev limiter out of the system by preventing it from having any effect on the CDM's.

My wife then restarted the engine, and presto - problem fixed! We were careful to keep the engine below 5000 RPM's because it had no overrev protection, but otherwise the engine ran just like it always has. Started instantly, purred like a kitten.

(Note: I have had this problem very, very intermittently for a couple of years now. I've never been able to figure it out because stopping and restarting the engine always fixed it. I could never get it to occur on demand, until now. If it's the rev limiter, perhaps it was flaky all along and finally failed completely.)

What isn't immediately logical is how/why the rev limiter affected ALL six CDM's. A careful reading of the manual strongly implies, but stops just short of actually saying, that it operates by retarding spark on CDM #2:

"With the engine under load and at maximum RPM, the timing on #2 cylinder should be within specifications. As the engine rpm reaches approximately 7000 +/- 100 rpm, timing on #2 cylinder will retard, as needed, to a maximum of 30 degrees and then spark to the cylinder will stop." (Mercury Service Manual 90-877837, page 2A-46).

If true, that means the single "special" connection it has to CDM #2 is the rev limiter's output. That in turn means its input would be the common connection it has to pin 2 of all the CDM's. Presumably it would sense engine RPM's by aggregating the signals from all six CDM's and then alter CDM #2 via its special connection.

What's weird about the above scenario is the common CDM connection goes only two places: The rev limiter as described above, and the dashboard console connector where it connects to the kill switch and key switch (?!?). THIS suggests pin 2 is used to kill the ignition. Digging into the CDM schematic, we find that the common connection to all CDM's goes to their pin 2, which is used to discharge the capacitor.

So it's virtually certain that pin 2 is an "input" to the CDM's meant to prevent sparking. That reverses the rev limiter scenario theorized above and makes the unique connection to CDM #2's pin 3 the rev limiter's **input**. The collective connection to pin 2 on all CDM's then becomes the rev limiter's **output** and explains why I was seeing intermittent spark on all cylinders simultaneously: The rev limiter was limiting engine RPM by discharging all six CDM capacitors temporarily.

This is how the key switch and kill switch shut down the engine. The rev limiter doesn't quite shut down the engine because it stops grounding the capacitors in time for the energy stored in the flywheel to keep the crankshaft turning.

This all makes sense, except for that "special" connection to CDM #2 and the manual's fixation on measuring cylinder #2. That misled me for a long time and kept me from investigating the rev limiter earlier. Had the description of the rev limiter's operation been more clear/accurate, or even less incorrectly detailed, I might have saved a lot of time.

My next task is to see if a local Mercury service shop has an 821889A55 rev limiter module in stock. Otherwise, I'll order one online. They cost about $60. Then I'll decarbon the engine (since it's had quite a bit of suppressed spark operation, meaning unburned fuel and oil have been passing through the cylinders) and hopefully that will be the end of this little adventure.