Don’t do cosolvent without a C1D1/2 setup. Supercritical ethanol would be a bad thing to leak.
For typical THC runs you just need a solvent density of about 800 kg/m^3. If you’re going for THCA (a little less efficient) you’ll want to back down to subcritical.
Looks like you’re running ~899 kg/m^3 already so you should be good t go for a THC run.
@ExTek90 had a method of true fractional extraction where he would run a light fraction first pulling off terpenes. Following this would be a quick heavy fraction run pulling off waxes, and finally the last heavy fraction would pull the remaining acidic cannabinoids. Although, if you read further and continue his procedure, the ultimate goal was to produce crystallized THCa, I do not see how this would not work for what you are trying to do.
Your calculated densities seem sound, although, I don’t know if there is any info out there that would elicit parameters geared towards fractionating CBD off from the rest of the THC and other cannabinoids. Might be out of the hands of CO2 extraction and more applicable for fractional distillation or chromatography.
Thank you all for those answers. I have now plenty of stuff to read and try. I will try to come back with some personal conclusiones from my own trials in a bit in this post. Cheers
CBD is a little less soluble than THC in CO2, but not by much. Good luck optimizing your extraction, it can be quite an ordeal to get clean cannabinoid fractions.