GPU Optical

Why not even unobstructed mirror systems?

Central obstruction is only one of the two problems of mirrors. Their other problem is their sensitivity to all disturbing effects like mis-alignment of parts, distortions by the gravity or mechanical problems, thermal problems, etc.
If a mirror of a Newton telescope is not perfectly collimated, the result is a huge amount of coma, ruining the image. The same problem in the case of an aplanatic lens causes practically very little image degradation, as these lenses are highly corrected for coma, they usually have only one significant off-axis aberration, which is astigmatism. They also have some focal plane curvature, but it is not important in the case of planetary applications. For this reason, refractors are not sensitive to a small amount of mis-collimation and you do not have to regularly collimate them. But the real problem is not the need to regularly collimate a Newton, the real problem is that there is no perfectly collimated telescope in this Universe. Even after the most careful collimation some mis-alignment remains, which will always interfere with the performance of a mirror system, but have practically zero effect in the case of an (aplanatic) refractor.
If a mirror suffers any mechanical deformation, (e.g. bends under a force), the resulting wavefront error is twice the amount of physical deformation of the mirror, as light travels once to the surface of the mirror, then back through a similar path. On the contrary, if a lens is deformed similarly, the wavefront suffers less damage, as the deformation of the glass surface changes only the length of path travelled by light through air v.s. through glass. As glass has about 1.5 refraction ratio, the difference between the wavefront of the deformed and undeformed condition is only half of the physical deformation of the optical surface. So, the deformation of a lens causes a wavefront error of HALF the amount of physical deformation, while the deformation of a mirror causes a wavefront error of TWICE the amount of physical deformation. This indicates that lenses are at least 4 times less sensitive to physical deformations.
Mirrors are also very problematic about thermal problems, as light goes through twice on the layer of warm air on ther surface, this effectively doubles the problem caused by warm optics. Also, mirrors are usually in the lower part of the telescope tube, so the warm air from them flows up the telescope tube, and light travels a long distance in this optically inhomogenous air.
So, even if we build an unobstructed mirror system (e.g. a Yolo, an off-axis Newton , etc.), it will still have a significant disadvantage over a refractor both in regards of error resistence against mechanical problems, and also regarding thermal problems. By applying ventillators, we can handle the thermal problems, but we can do nothing to make these systems more error resistant mechanically.
||