The printer is a RepRapPro Mendel legacy. The reason of the so frequent bed calibration, I think it depends on how the bed structure is done using 3 screws and springs perhaps too soft, and even my frequent and radical mechanical fix that screw-up the calibration.
As example yesterday I installed two belt tensioners, and I have to dismount the x-axe and played to much with the y-axe dismounting the belt. I will update some part for sure, I have seen the latest design solve some issues on the bed structure.
The I discovered that the X-axes was not going any more up and down. Once up, it blocked itself in one side and needed to be manually pushed down during movement. That was because the right Z-rod and the Z-smooth rods were not parallel, because the plastic of the structure holding the motor was 3mm on left of the ideal center line of the diameter of the motor and the one of the z-rod.
I dismounted everything trying to force the motor position in the pre-defined holes as much as possible to right to earn the 3mm. The ideal solution would have been a 5mm long track hole instead of the circular one for the motor screws, so that it could slightly be moved. The problem that so perfect and forced measure coincidences work well in the CAD, but often fail in the real life due to slight print and mount imprecisions.
In reality I am new on the 3D printing and I still have to choose if I will spend my time on PRINTING 3D objects or DISMOUNTING and MOUNTING the printer with continuous fixes and upgrades in a research of the holy grail of the perfect printer.