On Oct 30, 10:51 pm, Bill Schwab wrote:
Hello all,

I have a design problem requiring a reasonably rigid structure in a
tight space. Forces are not at all large, and structural failure is not
a great threat. However, the parts need to be securely connected to
each other in the sense that we cannot afford to have parts spinning out
of alignment when disturbed. The result is a scientific instrument
package not unlike an indicator holder: it needs to not fall apart every
time it is touched, but does not have to be very strong.

To provide rotational support at the connections[*], I have been
planning for machine screws in pairs. Space is tight, so that has
pushed me toward 4-40, which is a pain to tap. I am thinking of using
one 8-32 screw and making a tight-fitting slot to prevent rotation at
each connection. Any preference or other ideas?

In such a situation, how close to the edge of an Al part would you
consider it "safe" to tap for 4-40 and 8-32, respectively? Again, the
concern is not so much ultimate strength as robustness and common sense.
Ordinarily, I try never to get closer than 0.1" from the OD of the
hole to the edge, and usually think more about staying clear of
parallels, so this is new to me.

Thanks,

Bill

[*] It pretty much has to be made in pieces. Some of you could probably
remove a joint or two, but as Dirty Harry said "A man's got to know his
limitations."

Proper mechanical design treats threaded fasteners as items to fasten
two pieces together, not as locators for parts. If you need to locate
two parts in a precision fashion, you need to match drill, ream and
dowel pin them, slot and key them or machine mating areas on the
parts. Threaded fasteners are supposed to be only used in tension,
you see that one violated all the time, too. If you need to support
shear forces, you need to put in pins or machine shoulders to take
those forces.

Stan