I've taken measurements from Troy selectors, as we do so with every selector we can find so we have as complete a reference table as possible. Troy selectors dimensions are very similar to ours, except the selector center.
Our selector center probably has no equivalent, in that it's slightly longer to accommodate a wide variety of receiver thicknesses. In the earliest days of its development, we tried to design a center that would fit everyone's receiver, while minimizing the gap between the lever and receiver, until we came to the conclusion that it's impossible with the 5.56mm and 7.62 ARs, as the range of thicknesses is rather wide, even within Colt there is a wide margin. With forged receivers, this is I think inevitable, while billet receivers can hold tighter tolerances. But then there are many billet receiver manufacturers, not all of them are within specs, more info here https://www.m4carbine.net/showthread.php?t=81908
We published the article regarding the gap, as it's an often asked question. Our center's length is based on a Colt SP1 and on the thick side, it works with receivers as thick as 0.910.
Whereas a selector lever would be flush on both sides on a in spec receiver, the gap would be present on thinner receivers. Conversely, this same selector has the potential of binding its levers on thicker receivers.
With the 15-22, as S&W is the only receiver manufacturer for the MP15-22 receiver, so we didn't have to worry about a wide range of thicknesses, and made the levers as flush to the receiver as possible.
Re: Noveske receivers, every one of them we've come across has been right on the money and within milspecs in terms of thickness (tolerances are between 0.878 to .0895, Noveske receivers are right in the middle), so I'm curious as to why the Troy wouldn't work on it? 
The only thing I can think of is that Troy uses a dovetailed design for lever attachment, much like our BAD-CASS series. On the BAD-CASS series (M16, semi auto 90 degree, and semi auto 45 degree, and soon 0-50-18 short throw M16), the center is longer than the BAD-ASS so the dovetail interface will not bind on the receiver.
On the left is the BAD-ASS, on the right is the BAD-CASS (dovetailed), photo courtesy of JConsiglio
Roger Wang
Forward Controls Design
Simplicity is the sign of truth
Bookmarks