Sooooo
If that’s true it seems to imply conventional wisdom about retorquing may be in error.
Specifically that once you correctly torque something and make a witness mark and go about your business. At a later date if you need to remove a scope and reinstall with a torque wrench it’s commonly known that you’ll go past your witness mark before the wrench indicates desired torque value.
Conventional wisdom (ie most gunsmiths Have told me) has been to just go to the original mark. But you’re saying the bolt stretched and isn’t elastic and needs to be stretched again so you should continue til the wrench says stop.
Not necessarily.
Let's talk about fracture mechanics a little bit.
Fracture mechanics is, essentially, the study of what it takes for a material to break under applied stress. There are many terms used in this field (I know, say it ain't so for an engineering subject!).
Fracture toughness is the ability of a material to resist brittle fracture by undergoing plastic deformation...by "bending", in other words. A couple terms for this bending are "elastic deformation" and "inelastic deformation".
"Elastic deformation" is the ability of a metal to deform and then return to its original shape after the forces which caused the deformation are removed. If you exceed the maximum allowable force for this, then the metal will not be able to return to it's original shape...it will remain deformed. This would be "inelastic deformation". How much deformation will exist is a function of the material's fracture toughness and the forces applied to it.
If you stay within the design limits of the bolt for a given torque range, then the bolt will only experience "elastic deformation". If the torque range is going to approach the design strength of the bolt, then it's time to consider a differnent bolt material and/or a higher grade bolt.
A "witness mark" (or "match mark" as I'm used to calling it) is useful in and of itself. But it's no substitute for an actual measured torque value in critical jobs where torquing is very important. Many things other than bolt deformation can affect the witness marks upon a subsequent torquing event: lubrication, contact surface areas between the bolt head and the component, temperature variations between the time of the original torque and the newly applied torque, thread damage (even microscopic galling of CRES components), improperly applied or wrong type of lubricant, use of a different torque wrench (even if they were both in calibration), and probably a few other things.
I typically use match marks where it may be important to note whether the bolting hardware has subsequently moved during a period of operation after the torquing event. I did this, for example, on a particular generator repair job on an aircraft carrier after that ship had repeated problems with the hardware mysteriously coming loose, allowing damage to a slip ring assembly. We had done several repair jobs in a short amount of time and it was my firm belief that the problem was NOT hardware related. (Read: personnel related.). This was ONLY happening on ONE particular generator on ONE carrier out of the entire fleet. Strangely enough, once we match marked the hardware such that they were easily visible upon inspections during operation and before/after maintenance evolutions, they stopped having a problem with hardware "mysteriously" allowing components to move.
Things that make you say "hmmm...".
Some engineering instructions I've had to work to direct a re-torque upon completion. So, if it's in the drawing/technical manual, that's what you do. You may or may not observe actual movement during this and there could be many reasons for doing so. Perhaps it's a function of the gasket material being used. Perhaps it's a function of incremental torquing for a given torque sequence which requires a final torquing upon completion to verify the incremental torquing didn't affect the applied torques which came before.
For me, when I'm working on my car, a re-torquing is simply an idiot check I do at the end to make sure I didn't miss anything. Life would suck after an oil change, for example, if I only THOUGHT I properly tightened down the oil pan drain plug or only THOUGHT I got all the lug nuts on all my wheels after a brake job.