Thanks for the input. The Celcon had a formula and a table for a few flavors of their materials. (The formula has F in it but never defines F, I assumed it was f, which was defined but not used). The sabic website has a good calculator, but the allowable stress went up as the desired cycles went up, (the design fails at 1000 cycles but passes at 1 million) so I am a little skeptical.
I'll keep looking and post back if I find anything better.