657 engine quits at high speeds – solution

[micro222]

This was quite a head-scratcher for me, but I eventually figured it out. I’m posting this in case someone else runs into the same problem.

The engine would run perfectly until a specific RPM was reached, then it would suddenly quit. I could restart it immediately and resume, but it always happened at the same RPM.

I disconnected the wire from the MPEM to the ignition module and no more problem.

I couldn’t leave the wire disconnected because then the lanyard or stop button wouldn’t shut down the engine. To fix this, I installed a relay between the lanyard/stop wire and the ignition module. It now works perfectly.

 

[etemplet]

So the relay shuts down the engine?

 

[micro222]

Yes. When the relay is off (ie no current flowing through its coil) the relay contacts disable the ignition module by shorting the input, just as the MPEM would. When the lanyard is plugged in the relay turns on, the contacts no longer shorting the ignition. A useful thing to know about the ignition system is that the MPEM is not needed for the engine to run. The magneto powers the the CDI ignition module and the spark plugs fire.

 

[etemplet]

Hmmm... I thought the MPEM had a timing curve for the engines. Well.... glad you got it working !!

 

[micro222]

At first I assumed the MPEM handled some complicated engine control functions, but after disconnecting it and seeing that the engine ran exactly the same and that only one wire went to the ignition I had to rethink that idea.

 

[etemplet]

I know Matt Braley could set a timing curve on certain model ski when programming a key. I'm curious if it is as you say or it is not someone one would notice. In the end... Who Cares?? ... you got it running. LOL

 

[micro222]

From chatgpt:

For a 1994 Sea-Doo, the ignition timing depends on the engine model: most 1994 650/657 cc 2-stroke Rotax engines (used in GTX, XP, GTS, etc.) use a CDI ignition, which is fully electronic and not mechanically adjustable.

Some details:

• The stock CDI is pre-programmed for optimal timing; it advances automatically based on RPM.
• Typical timing at idle: ~20° BTDC
• Timing at full RPM: ~28–30° BTDC

 

[lagojaguarasp]

I wonder if the rev limiter is on the mpeem or on the CDI module in theese. Sounds sketchy to go full blast w/o knowing for sure if you also just did a rev limiter delete

 

[micro222]

I wonder if the rev limiter is on the mpeem or on the CDI module in theese. Sounds sketchy to go full blast w/o knowing for sure if you also just did a rev limiter delete

It's in the MPEM. By bypassing it the way I did there is no longer a rev limiter. I strongly suspect that the capacitors in the rev limiter were giving out and caused the problem I was having.

 

[lagojaguarasp]

It's in the MPEM. By bypassing it the way I did there is no longer a rev limiter. I strongly suspect that the capacitors in the rev limiter were giving out and caused the problem I was having.

That does make sense!

 

[etemplet]

I may have an MPEM that would be useable in the ski. I've dug into them previously.