As I said, I understand that you want to try the etching route; I was just curious if you had also explored the other options.
Some of the later responses have detailed some reasons why the results may not be crisp -- the problem of getting a good pattern transferred onto a round surface; the possible problem of undercutting if you go to much depth; the possible problem of getting an even etch at all points. The latter two points may be especially acute when trying to produce consistent, fine lines appropriate for machinist levels of precision -- particularly if you leave the lines "raised" against an etched background.
None of these are intended as arguments against trying it. On the contrary, I hope you do try it, and I hope that this discussion helps you as you anticipate and work through possible challenges. (The latter is part of the fun in inventing, right?) The key questions I will be interested in seeing answered are the following: 1) Does the method produce results as good as (or better than, or worse than) the standard approach? 2) Can acceptable results be produced as easily as (or more easily than, or less easily than) the standard approach?
I look forward to seeing your results.
--- In Homebrew_PCBs@yahoogroups.com, "Rick Sparber" <rgsparber@...> wrote:
>
> Andrew,
>
> By "new method" I mean chemical etching. My hobby is "inventing". I do have
> a lathe, mill, and shaper so could make a dial if I wanted but that is with
> standard methods. When I etch a circuit board, the features are very shape
> and detailed. So why would I not get crisp features on a dial?
>
> I can transfer toner to metal by using kitchen parchment paper and crazy
> glue. Works OK. But true etching would be far more interesting right now.
>
> Rick
>
> -----Original Message-----
> From: Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com]
> On Behalf Of Andrew
> Sent: Tuesday, March 19, 2013 1:53 PM
> To: Homebrew_PCBs@yahoogroups.com
> Subject: [Homebrew_PCBs] Re: etching the OD of a cylinder to create a
> graduated dial
>
> Rick,
>
> When you say "a new method," do you mean new compared to the
> mirror-image-on-clear-plastic that you described earlier, or are you also
> including new as compared to the standard machinist route using a dividing
> head and number stamps?
>
> I understand that you want to try an etching method, and if you succeed I
> will be interested in seeing your results. My suspicion is that it will not
> be as clear and crisp as the dividing head look ... but of course, I may be
> proved wrong.
>
> A year or so ago, a participant on the Practical Machinist forum described
> his rebuild of a large bandsaw (over a span of many, many posts); IIRC, one
> segment detailed his re-creation of some machine plates using an etching
> process. These were flat, however, rather than cylindrical -- a much easier
> proposition. Still, might be worth looking at for ideas, if you haven't
> already seen it.
>