On Tuesday, November 6, 2018 at 11:44:31 AM UTC-8, wrote:
Andy, when I was laying out my ASW 24 panel many years ago (OK, over a quarter century ago), I wasn't just arranging paper cutouts on the blank panel to see what would fit. There were other consideration such as clearance behind the panel, minimum distance from the edges for structural integrity, minimum distance between two instruments for structural reasons as well as interference between them, obstruction of the top edge of the panel by the glare shield (I have a Streifeneder panel with higher knee cutouts and a three-flats--vs. rounded--cover), instruments that fit 57mm or 80 mm holes but had larger dimensions behind the panel (e.g., the Cambridge and ClearNav ADCs), switches and fuses, ELT annunciator, sightline behind the control stick, etc. And now we have the larger, non-standard panel displays mounted on stalks, in the panel, behind the panel, etc. Making sure everything fit was an iterative process to a certain extent. I cut the first hole, test fitted it, then proceeded, making adjustments as necessary.

Any quick thoughts to share with the group on how you dealt with these other issues? I assume that's what's meant by cutting a trial panel and testing for fit. I've changed my instrumentation a bit in the last few years so am thinking about a new panel. The idea of unpacking a brand new, beautifully cut panel and discovering my favorite instrument doesn't clear something by a few mm is daunting.

Chip Bearden

I'd suggest making up the layout you like in CAD, then have a piece of waste material (G10, acrylic, or whatever) cut to that plan. Then assemble all the actual instruments onto the test piece, hold it in place, and see if it all really fits. Then cut the edges right down close to the instruments and see if the test piece fits inside the real panel. This tests for clearance to the radius of the molded piece. Do not depend on the AS drawings if you are pushing things to the edge. It also allows you to locate the test piece (and therefore the CNC) to the real panel (which has no reliable center line and will be asymmetric to some extent).