Nice work Steve. I read your report. Great stuff!
I have had a sewn corner tie out experience some damage but it was still pretty far away from 'failing'. I'll see if I can explain this clearly. In this instance, a nylon reinforcement patch was sewn to the corner of the cuben by sewing along just the perimeter of the nylon patch. The guyout was then sewn at the corner through both the nylon and cuben. This design led to damage of the cuben because the nylon has some stretch to it. When a force was applied to the corner, the nylon would stretch and thus leave the non-stretchy cuben to bear the brunt of the force. In effect, the cuben wasn't really reinforced because the nylon was only sewn around the perimeter and it was able to stretch and leave the cuben to bare the force.
As a result, the cuben started tearing where the guyout was sewn. However, since the guyout was also sewn through the nylon, once the cuben had somewhat failed the nylon was then left to bare the brunt of the force. The nylon would then stretch and disperse the force between all the stitches along it's perimeter which seemed to be working. I also don't know for sure, but the cuben seemed to be CT1K.08 rather than CT2K.08. The manufacturer claimed 0.6oz cuben but having used CT2K.08, I would say this was a lighter weight.
Regarding Lynn's failure, I have a photo of this spot:
In the initial design the guyout was attached to the loop of nylon at the bottom. Since the guyout pulls somewhat perpendicularly to the cuben, this put the stress on only a few stitches at the bottom leading to Lynn's failure. After reading about Lynn's failure I used a needle to re-route the guyline through the nylon as shown. I believe this method disperses the force over a much larger number of stiches and thus is less likely to fail.
Regarding your tests Steve, I would say that the sewn method is more prone to experience gradual failure than the other methods. While the sewing failed suddenly in your trial, I believe it can experience stress under normal use and the stitching holes can gradually enlarge and slowly bring it closer and closer to failing. I don't think this 'gradual fail' is really a possibility with the bonding methods.
It seems to me the results of your tests show that if you want a stronger corner, instead of sweating about what method to use, the best method is to use a larger reinforcement patch. Since the samples normally failed along the edge of the reinforcement, using a larger reinforcement would spread the forces over a larger area of the cuben.