The Solar Splinter Saga continues in 2004.
After the partial success of last year solar splinter was upgraded several ways:

The Result:

It Flies well:

During flight I either had a control anomaly or the horizontal stabilizer failed.
Since the H Stab was found in two peices 50 ft apart I currently believe the H stab failed.

Update on crash theory:
one of the aero guys on RCSE suggested that my problem was a too flexible boom. The basic concept is as the aircraft speeds up the tail needs to provide more down force. This flexes the boom down reducting the H-stab angle of attack, pitiching the nose forward, increasing speed, flexing the boom etc.... Until something breaks. (The h-stab in this case) This fits better with my perception of the failure. It also fits better with my ground testing. The 2005 version will have a much stronger boom.
We will keep trying...
Work has begun on Solar 2005: see Cutting the tail ribs...
The Brain that runs the whole thing is a MOD5282 Embedded Ethernet Module from NetBurner.
I will publish schematic and source code details later on as I finalize the 2005 effort.