Re: Step wedges

Wayde Allen (allen@boulder.nist.gov)
Fri, 22 May 1998 12:31:00 -0600 (MDT)

On Fri, 22 May 1998, Peter Marshall wrote:

> I've done this before and it works more or less for a few steps. However
> the real problem if you haven't go a densitometer is in deciding what is
> density 0.15.
>
> When you get a lot piled up I think reflections on each surface start to
> cause a problem, as does dust, scratches etc.
>
> You would also need to develop so that your straight line has the correct
> gradient.

Then there are about a hundred other problems that you haven't even
thought of yet. This is at the root of basic metrology, has been the
basis for many peoples careers, and is the reason the national standards
labs exist.

> As I said before if you want a reasonably accurate wedge pay the 8
> dollars!

Yes, unless you want the challenge of designing and building a check
standard. There is certainly nothing the matter with that. In the end,
it all depends on what is "good enough". For most of us, a reasonable
level of precision is plenty adequate. Accuracy is seldom needed for
alt-process work.

- Wayde
(wallen@boulder.nist.gov)