View Single Post
Old July 5, 2019, 08:34 AM   #10
45 Dragoon
Senior Member
Join Date: June 14, 2013
Posts: 648
The "taper" that Denster refers to is present in some revolvers, not in all revolvers . . even the originals. Full contact of the forward arbor slot and the wedge can and will allow the wedge to "walk" and loosen. Denster and I had a disagreement about this in the past and apparently we still do. And that's OK.
The proper set-up for the wedge to work is simple triangulation. The rear surface of the wedge contacts the rear slot in the barrel assy. ( no contact with the arbor slot). This establishes a wide and parallel "base" with the forward "point" being a contact patch with the forward surface of the arbor slot. This has been verified by my own revolvers as well as many customers and even from a thorough investigation of a '60 Army produced in 1862. The original's arbor was broached without taper and the contact point of the wedge was just inside the entrance of wedge slot (easily verifiable). This is the typical area of contact in most open tops (tapered or not) but it doesn't have to be as long as it's not full contact. Pietta uses or did in the past, a process that looked as if the forward arbor slot surface was ground on both sides leaving the center area as a contact point. The main point in this particular discussion is that vibration has less effect when contact between surfaces are smaller rather than larger.

As far as arbor length correction, I much prefer a single S.S. plug anchored in the arbor hole. I don't like adding anything the end of the arbor which denotes an obvious "repair" and looks less "professional" . . . to me anyway. It also allows for the installation of a 1/4" smooth set screw to act as an adjustable bearing for the wedge. This set-up allows the user to "customize" wedge placement as well as allowing for any wear over time. Thanks noelf2!!

Follow me on Instagram @ goonsgunworks

Last edited by 45 Dragoon; July 5, 2019 at 09:17 AM.
45 Dragoon is offline  
Page generated in 0.03046 seconds with 8 queries