I think you mis-understand the concept of proofing. Here's a mechanical engineers take on the subject. Proofing was developed as a "necessary evil" to test firearms over a century ago. With the rather primitive manufacturing and metallurgical methods available, proofing was devised as a way of ascertaining poor metallurgical properties, and in particular, flaws that were present in critical parts of the gun. These flaws would not necessarily be visible to the naked eye. A flaw of sufficient size would lead to failure on proof, hence the term destructive testing.
Where proofing falls short is when smaller flaws are present. These have the ability to propagate cracks under cyclic loads. The magnitude of the load, and size of the crack, determines how many cycles to failure. This phenomenon is very well understood in machines such as jet engines. Non-destructive examination (radiography, ultra-sound, etc) is used extensively to survey parts for integrity. NDE is also used in the manufacture of firearms. Proofing has been retained as a legal requirement in some countries, but is also employed to test the basic function of the firearm.
Your proof tests are inconclusive and possibly dangerous as you dont know the magnitude of the stress you are imposing. If you are substantially above the endurance limit, you are in the grey area of finite cycle fatigue. Testing with one or two proof loads at the right pressure is sufficient to test for large flaws, anything beyond that can be contributing to fatigue.(ie you have reduced the service life of the gun, even with normal loads.)
I understand you're a young fella with a keen interest in firearms - we were all like that at one time. Just be mindful that inductive learning can have much more negative consequence that deductive learning. Good luck with your endeavours, whatever they are.
