Steve,

Thank you for posting the information on the solution for the spark problem.  The results are of particular interest to me since the routing of my ground to the starter sounds like it is identical to yours.  The test I made that showed similar symptoms was based on the possibility that the EC2 and the CAS return were grounded in two different places and that noise generated in the ground between these two locations was degrading the CAS signal.  If I understand correctly, the noise generated due to the starter current passing through the engine assembly is being injected into the CAS signals even though the CAS circuits and the engine block are electrically isolated.    

Benefiting from your troubles, I am fabricating a longer ground cable for installation in my plane.  My inclination is to add this cable to the starter and retain the original one to the front cover.  Removing the cable to the front cover would result in the ground for alternator passing through the big Oreo cookie.   Although the alternator currents are not as large as the starter current, they are still significant, noisy, and more likely to cause problems during normal operation rather than just during starting.

Steve Boese