So you don't think it is possible to establish a base of something like, "Gee I can be pretty much sure its not going to be out of acceptable spec for 5000 inches of travel," or "since the bearings in the spindle have never gone out of acceptable spec in less than 500 hours I should never have to worry about bearings if I change them every 500 hours"?
I think my brother in law would disagree with you. I certainly disagree with you.
OT: As far as I know Boeing doesn't make any of those parts except design and prototypes. They contract them out to companies like the one where my brother in law works. Even a lot of the prototype stuff they send out to job shops. If you wonder why I said this reread what I actually wrote.
Back on topic. Like I said. If you want to predict the exact moment of failure you will fail. In this small and very narrow precise definition I agree with you. That is not MY definition or goal.
However for example, if I keep track of how long it is before spindle bearings start to increase runout beyond what I am willing to settle for, and wear out four sets of spindle bearings with a life of:
(Purely for illustration purposes)
750 Hours
1000 Hours
680 Hours
1100 Hours
I can establish a baseline that I should never have any problems unless there is a major crash to cause them in less than 500 hours. I don't care if its a $99 Bosch Colt Router or a $10K 10 Kw water cooled 3 phase spindle. If in my example I replace bearings every 500 hours I should never have a problem with bearings under "normal" use. I should of course note and log any major crashes or replace any components that I feel may be negatively impacted. The figures 500 hours and 5000 inches are used purely for illustrative purposes.
I can also use this to establish "value" to determine over time if a stock bearing, hybrid sealed bearing, or full ceramic bearing is a better long term value when comparing cost to life expectancy. Two sealed full ceramic bearings for a Bosch Colt router cost almost 2 and a half times the price of the router, but if my "safe" limit is 10K hours out of them and get a "safe" limit of 500 hours out of the stock bearings they're more than worth it.
While lead screws and nuts like the crappy V leads and pinch nuts on the Taig might seem harder predict the same data gathering can be performed over time. I may not be able to predict that it will go bad at a particular time, I can certainly find a life/inches of travel where it almost never goes beyond my acceptable tolerance, and use that as the baseline to adjust the pinch nuts, gibbs, and bearings per axis.
I might add that I have already established a "feel" for some of this. The Taig has been on 3 different computers now so I can't give you the total number of hours its been powered up, but after Hood pointed me in the right direction I discovered that this computer has been running Mach 3 for over 2300 hours. I estimate that is about 87-90% actual running time. I do not leave Mach open when I am not running a job. It was on its previous computer for a substantially longer calendar period, so its probably got around 4000 hours of actual run time on it.
In the past when I was just using the machine to play and learn it was no big deal, but now I am more and more concerned with preventable crashes and failures as I run it longer at a time and much more often. Its just a matter of whether trying to push it to the longer limits is worth more than the wasted time when it fails and destroys a work piece VS changing the components and making adjustments within the established "safe" period.
I'll continue to try not to destroy work pieces if I can prevent it in my way. I think that this type of data and record keeping can help work to establish a tolerance life with a Taig, a Haas, a Hurco, or a building size custom built machine. It does require some diligence and recording keeping. As a hobbyist being miserly with my time I may not want to do that, but I firmly believe it can be done and its is a useful tool. If I start making money and find a niche business then it becomes much more important.
If you still disagree so be it, then we are just going to have to disagree then. Mach 3 does have a time and distance log. Somebody besides me must have thought it was useful.