Calibration plate calibration...

I think calibrating optics is not necessary as often as calibrating tips, but you should calibrate offsets between optic and ttp at each calibration of one of them (if you use both of them).
For the calibration of artifacts, I think it's not necessary, if you can check them by comparison (calibrate on the sphere and measure the ring, and calibrate on another sphere and measure the sphere).
The problem of re-calibration of a calibration sphere is that the measured deviation is often less than the uncertainty of calibration, so you change the diameter of an "uncertain" value.
Checking spheres by comparison is enough on most of cmms. Maybe it's not the case for very accurate ones...
For those different points, it's just my opinion !

Yes, as we use both camera and probes, we calibrate the offset each week too. But that glass plate... it takes quite some time... and i would like to prove my superior it is not necessary (except for a shock of the head for example) but don't know how to argue...
For the artifacts calibration, i admit it's a disturbing question, some years ago Hexagon were not able to give us an answer, they said something like "valid over 10 years"

Thank you anyways !

JEFMAN pretty much covered it.

You do not need to calibrate the optics very often. There are two reasons you would want to calibrate with any regularity:

1. You crash the machine and have to physically alter the camera.
2. you do a lot of min/max magnification settings within your programs.

The biggest difference between the camera and probe is the probe in mechanical where the camera isn't. Since there is force being applied to the probe it has to be re-compensated periodically. Since there is no physical force on the camera, it's less necessary. (again, unless you do a lot of mag in/out.)

The offsets should be done regularly, depending on your tolerances, i'd say once a week as a rule-of-thumb.

Regarding the artifacts...there's no reason for them to be calibrated. They are simply a known value. Unless you drop them, break them, expose them to extreme temperatures, there's no real reason to worry that they have changed. I have seen plenty of customers send them out for calibration or do them in house...that's your decision but it's not really necessary.

Thank you, that confirm my way at looking at things