CCD Inspector does it's best work when it's in a starfield, not so much on an individual star. For what it's worth, here's my protocol:
At the beginning of each session, slew to a known good star at the same altitude as my target for the evening and use Frank's MetaGuide for a collimation check: it will tell me if I'm good or not pretty quickly. SharpCap can also work, I like MetaGuide
If things look off, I will bring the OTA in and use a Hotech SCT collimator (with the OTA at the same angle as the general altitude) to get it pretty well dialed in as well as ensuring the secondary is centered on the optical center. You can also use a Hubble artificial star, just use a couple of convex mirrors to reduce the "dot" down to a few microns (I use that with my 8's and 6's as the Hotech doesn't work lower than 9.25")
At night I'll find a nice cluster somewhere with at least 20 good stars (again at the same general altitude) and use CCD Inspector to dial it in. You want to get it so that the "red" corners are all uniform and balanced
A few notes ...
- Make sure that you collimate at the angle that you will take pictures at ... many people collimate when the OTA is horizontal, then wonder why it looks different when the OTA is nearly vertical in use ...
- The C14's (and C11's to a lesser extent) suffer from mirror flop which can throw off the collimation. Once I have finished collimating I use some blue painters tape to mark the "down" position of the collimation and simply don't let the OTA go vertical or inverted from that mark (so, no meridian flips) ... problem solved. If you don't do this, be aware that your collimation may get goofy as you go across the zenith
- Any kind of impact on the OTA can also throw off the collimation