Measuring the taper of a taper-sided....

[SGW]

I'm curious how you might go about measuring the taper angle of a taper-sided slot. Imagine a plate 1/16" thick. There's a slot cut through it that is about 1/16" wide on one side of the plate and maybe 1/8" wide on the other side of the plate, AND THE TAPER DOES NOT EXTEND THE FULL DEPTH OF THE SLOT.

Like this (side view through slot):

------.....-------
........\.../
........|..|
-------...-------

(Pretend the dots aren't there -- they're just to make the formatting work.)
Sort of like the slot is countersunk on one side. How would you determine with good accuracy what the angle of the tapered portion is?

Top view of the plate, the slot looks more or less like this:

.------
(........)
.------

The long dimension is maybe 1/4". All dimensions are approximate and are given only to give some sense of the scale involved.

My idea is to take something like silicone form-a-gasket that will peel off when it hardens, fill the slot, and when it hardens take it out, slice a thin vertical cross-section with a razor blade, and measure its angles with a toolmaker's microscope, which is available. But I'd like to know if anybody has any other ideas.


[This message has been edited by SGW (edited 07-08-2004).]

[rklopp]

You could mike over two sizes of small short cylinders (dowels) or balls that rest on the walls of the taper, then crunch some trig to work out the angle. Use a depth mike or a uni-mike if you can only access one side. This approach is analogous to measuring threads using the wire method.

[SGW]

This taper is maybe 3/64" deep, so such an operation would be pretty finicky. Perhaps do-able though, if I made a couple of short pins of appropriate diameter. With the toolmaker's microscope I can measure the width of the taper opening at the top and the bottom, so I can determine what diameters to make the pins.

I'll put that one on the to-try list; any other ideas?

[Joel]

An indicator in the (graduated) mill spindle?

I have a little angle gauge set made out of sheet stock that might work pretty well in this application. They were inexpensive.

[wierdscience]

put a piece of card stock on the bottom(small end) and fill the slot with either wax or idealy cerrosafe alloy,this will give you an accurate male profile which will be easier to measure.

If you use the cerro safe you could saw or mill out a thin cross section of the male projection and file or cut off the straight section,then take two strips of shimstock about 4" long and 3/4" wide with straight edges,overlapped and held in a V shape fit the angled portion of the sample into the intersection of the V until it fits tight with no light shining through,clothespin clamp the overlap and superglue in place.Then you will have a gauge with long enough legs to measure with a standard bevel protractor,it may have a +/- of a degree or two,but the angle is more or likely something nominal like 30* rather than 37.2 or other strange thing.I have used this method on several occasions with excellent results,since the only other good way is using an optical comparator$$$$$ it most of the time works perfect enough.

[John Garner]

SGW --

Your concept of using the groove as a mold has great merit, and there is a commercial catalytically-cured rubber product named Repro Rubber that is sold for this very application. The Repro Rubber cures very rapidly, doesn't stick, but won't be found in any "normal" hardware store and will carry a hefty price tag.

As wierdscience points out, Cerro Safe and its sister low-melting-point alloys are also often used for this purpose commercially but they have the price-and-availability limitations of the Repro Rubber.

A couple of inexpensive any-hardware-store-in any-town alternatives? I'd suggest Plaster of Paris or a quick-setting filled epoxy such as J B Kwik. The biggest drawback to both of these is that they will bond to the metal . . . so you better use a release agent.

One more thing to watch out for: Plaster and epoxy will both mechanically lock into undercuts.

John

[Paul Alciatore]

I like rklopp's dowel idea but instead of doing the math, just use a cad program to draw it in cross section and add circles for the dowels. Be sure you do draw them in contact with the lines that represent the tapered section. Use circles of a size that you have drill rod for. Then draw a line from the top of the circle to the surface and list it's length. That's the measurement you make with a depth gauge.

I used this CAD drawing and dowel method to make dovetails and it works like a charm. Also great for finding readings for three wires on an odd size thread. Did that last week on an English/metric thread that was about .75" - 1mm.

Paul A.

[Bruce Griffing]

If you have a mill, mount the part on an angle vise and put a DI in spindle. Adjust the angle until you can move the DI back and forth on the taper with no deflection. Now measure the angle of the plate with the table and you are done.

[SGW]

Thanks for all the ideas, folks. This forum is great!