Re: 1/12 scale R-40-23 reefer Dreadnaught end


Dennis Storzek
 

On Thu, Dec 12, 2019 at 08:45 AM, Andy Carlson wrote:
I think that in some very important ways the old-school methods of panto reducing down over size patterns with a pantograph can produce parts which the best of CNC tooling has not shown an even match.
-Andy
And I totally disagree. Anything that can be cut by a 3D pantograph can be cut by CNC. In fact, you'll be hard pressed to even find a pantograph in a modern day tool shop, unless the owner is almost retirement age and never gets rid of anything. Toolmaker time is expensive, and a pantograph is a manual machine. The question becomes do you want your toolmaker to spend hours sitting at the pantograph, or be doing something else while the CNC machining center cuts the part unattended.

Pantographs can't reverse the pattern; that is they can't cut a cavity directly from the model of the part. The usual way that was dealt with was to wax the part model, and then pour epoxy around it, thereby producing a pattern of the cavity.

The real problem is one of visualization. It is difficult to see contours in a cavity that are very evident on the part.  In the case of the 1/12 scale end, the pattern maker had the positive pattern to compare to the drawings, and most likely photos of the real end. In my toolmaking work, I cut graphite EDM electrodes, that, while small, are a positive copy of the part, so if contours are strange, they are evident, at least under magnification.

The problem with the IM end is it was most likely cut as a cavity directly into the mold plate, relying on the computer graphics to inspect the shape of the surface, and obviously the programmer got it wrong and didn't catch it. This could have been avoided by using the opposite side of the surface geometry to cut a positive part in plastic or wax. Since CNC files are scaleable, the positive check piece could have been made oversize for easier inspection.

Dennis Storzek

Join main@RealSTMFC.groups.io to automatically receive all group messages.