Bill,

 

If you think the problem is with the EC2 signals, an example of examining those signals is shown in the following web site:

www.rotarycopilot.com/

in the Start Up Sequence discussion in the Documents section.  Although this discussion dealt with an issue that has since been resolved, the data collection and examination methodology may be helpful for your situation.  Using this would depend on being able to collect data during the time that the problem occurred which might take some effort if the problem doesn't happen very often.

 

It may also be worthwhile to consider setting up the monitoring system described in the Software section of the same website.  This system was developed to troubleshoot my EC2 installation and enables one to verify the operation of the EC2 controls (among other things).  In my case, there was a cold solder joint on one of the pins of the mixture control potentiometer of the control board resulting in an intermittent connection.  Watching the EC2 output as the mixture control setting was changed made the problem obvious.  The engine didn't even need to be running to find the problem.