I've edited that part and designed it to have a diameter of 25,7mm (after several trials) and when printed it became with a diameter of 24,7mm, again the 1,0mm systematic error. Now I can finally place my 25,0mm aluminium tube as tight as yours but still be able to move it easily enough without risk of braking the piece due to the designed split! Nice!
About calibrating my printer, if I make e.g. a 100mm sided cube like you seem to suggest, and then measuring the differences between what each axis should be measuring and what it really measures, and then get those differences, would you be kind to tell me (us) what should be done mechanically to reduce those differences? I ask you because I presume that you being a Master of 3d printing that designs CNC/3D printing machines (my respect for that) you could have some short and easy answer for that. :)
Thank you very much and keep up the good work!