The attached article was developed to solve a problem that befuddled us for a while. how to measure diameters that were not a full diameter. I would be very interested in your reviews.
Thanks
Jim

Seems to me you'd have a hard time determining the point of where the arc changes to a line so you can measure the distance across the arc. If it was me (and I actually have to do this on my model railroad's curves) I would cut out a template with the minimum (or maximum) allowable radius and lay that on the curved surface. Sort of a go-no go gage.

Also, taking practice from the railroads, the traditional way to lay out (or to check) a curve (where you can't find the center or don't have the full diameter) is to take a 100 foot chord (or otherwise fixed distance), lay it across the curve, and then measure the angle between the chord and the tangent to the circle.

A resourceful way to solve a challenging situation.

Are there instruments designed to do this? I know I have seen instruments for measuring the radius of a sphere - a "tripod" that rests on the surface and a "micrometer" type probe at the center to measure the distance to the surface. The equivalent for a circle would be basically what Steve mentioned for the railroads - take a "beam" of known length and set it onto the curve. Attach a probe to the beam and to the curve.

( I tried to do and ASCII art drawing to show what I was thinking, but the spaces kept disappearing, ruining the picture. You'll just have to use your imaginations!)

If it doesn't exist, I sure think it would be a good item for someone to manufacture! This can't be the only shop that needs to do such measurements.

Tim F.


P.S. I like the "trial & error" spreadsheet approach for analyzing uncertainties. I could also solve the problem with calculus, but often the spreadsheet approach is just as quick and more intuitive. Calculus would have the advantage of giving a general equation that would work for different values of chord & height.

A resourceful way to solve a challenging situation.


( I tried to do and ASCII art drawing to show what I was thinking, but the spaces kept disappearing, ruining the picture. You'll just have to use your imaginations!)

If it doesn't exist, I sure think it would be a good item for someone to manufacture! This can't be the only shop that needs to do such measurements.

Tim F.


P.S. I like the "trial & error" spreadsheet approach for analyzing uncertainties. I could also solve the problem with calculus, but often the spreadsheet approach is just as quick and more intuitive. Calculus would have the advantage of giving a general equation that would work for different values of chord & height.
For drawing (to hold the spaces) - use real characters rather than spaces, then change the color of the characters which should NOT appear to "white"
(done in magenta on top "V" and white on bottom "V" here to visualize)


xxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxx

Thanks Steve, actually when we first became befuddled we used the OEM steel pins that were made for the coupler. These pins usually ran about .008" to .010" under the diameter of the bore. Our task was to determine how much wear had occured in the radius area. Engineering set the criteria at .005% of the OEM pin diameter. By inserting the pin into the "hook" area we could tell that wear had occurred but not how much wear. The degree of wear determined if we could rework or if it was rejected as scrap.
I do like your idea of making up a set of radius gages and I intend to explore this more!
Thanks
Again
Jim

A resourceful way to solve a challenging situation.

Are there instruments designed to do this? I know I have seen instruments for measuring the radius of a sphere - a "tripod" that rests on the surface and a "micrometer" type probe at the center to measure the distance to the surface. The equivalent for a circle would be basically what Steve mentioned for the railroads - take a "beam" of known length and set it onto the curve. Attach a probe to the beam and to the curve.

( I tried to do and ASCII art drawing to show what I was thinking, but the spaces kept disappearing, ruining the picture. You'll just have to use your imaginations!)

If it doesn't exist, I sure think it would be a good item for someone to manufacture! This can't be the only shop that needs to do such measurements.

Tim F.


P.S. I like the "trial & error" spreadsheet approach for analyzing uncertainties. I could also solve the problem with calculus, but often the spreadsheet approach is just as quick and more intuitive. Calculus would have the advantage of giving a general equation that would work for different values of chord & height.
Thanks Tim, I recently watched a demo of a CMM robot arm that could possibly do this in no time at all. the salesguy demonstrated that once set up touching the probe to 3 points on the curve would yield the radius. Cost was in Excess of $50000. My Engineering department is searching the cheaper versions.
PS. Although I studied calculus its been a few years ago. I felt more comfortable with the spreadsheet. Excel is great isn't it!
Jim :D

Would the "cord" have any bearing?

Al...

Our industry here is metal stamping, so I don't know if this will apply to you or not. Do you have an optical comparator available? That is my preferred method of measuring a radius. I have radius charts that I can lay over the image on the comparator, works pretty well. Personally, I prefer NOT to use a CMM for this type of measurement, it is too dependent upon the CMM operator touching on the radius and not getting too far up the side into the straight edge.

All the best!
Jonell

Our industry here is metal stamping, so I don't know if this will apply to you or not. Do you have an optical comparator available? That is my preferred method of measuring a radius. I have radius charts that I can lay over the image on the comparator, works pretty well. Personally, I prefer NOT to use a CMM for this type of measurement, it is too dependent upon the CMM operator touching on the radius and not getting too far up the side into the straight edge.

All the best!
Jonell

Excellent advice Jonell, We do not have an optical comparator but you can bet Its going on my wish list.
Thanks
Jim

Would the "cord" have any bearing?

Al...
Al, do you mean "chord"? If so then you are on the money :bigwave: because the chord in this case is "c" which is one dimension that we can measure.
Thanks
Jim

Yes, I meant to type chord. At times my mind is faster than my 3 typing fingers.:thanx:

Al...

Yes, I meant to type chord. At times my mind is faster than my 3 typing fingers.:thanx:

Al...Jim, see how handy that high school geometry course would have been if you had paid attention to the teacher instead of Suzie Derkins' profile as she sat in the 3rd seat of the first row?:bonk: :lmao:

Jim, see how handy that high school geometry course would have been if you had paid attention to the teacher instead of Suzie Derkins' profile as she sat in the 3rd seat of the first row?:bonk: :lmao:
Wes,
you are so right but her name was Rita! Thank god for the Machienry's Hand Book # 23. They have a nice little review of the things I missed while looking at Rita! :lmao: :lol:
Jim

Wes,
you are so right but her name was Rita! Thank god for the Machienry's Hand Book # 23. They have a nice little review of the things I missed while looking at Rita! :lmao: :lol:
Jimthe very edition I have! I think they must be up to about the 100th edition since I got mine.

Jim, see how handy that high school geometry course would have been if you had paid attention to the teacher instead of Suzie Derkins' profile as she sat in the 3rd seat of the first row?:bonk: :lmao:
Wes,

It looks like you were way ahead of your time. By the time you were in high school, you were already well-versed in the use of optical comparators to judge curves ! http://elsmar.com/Forums/images/smilies/embarrassed.gif

Tim F

Our industry here is metal stamping, so I don't know if this will apply to you or not. Do you have an optical comparator available? That is my preferred method of measuring a radius. I have radius charts that I can lay over the image on the comparator, works pretty well. Personally, I prefer NOT to use a CMM for this type of measurement, it is too dependent upon the CMM operator touching on the radius and not getting too far up the side into the straight edge.

All the best!
Jonell

Jonell, since our product lines are heavy and bulky (400lbs +) would it be best to render a tracing of the radius to the comparator or do they make comparators large enough to handle these bigger product lines? Not sure which way to approach. :confused:
Thanks
Jim

Jonell, since our product lines are heavy and bulky (400lbs +) would it be best to render a tracing of the radius to the comparator or do they make comparators large enough to handle these bigger product lines? Not sure which way to approach. :confused:
Thanks
Jim

Jim,

How big is the radius that you're trying to measure and what is the length, width and height of the product? :confused:

Jonell

Jim,

How big is the radius that you're trying to measure and what is the length, width and height of the product? :confused:

Jonell
Actually there are two radii intercepting each other (see photo in the article). Radius varies depending on machine size. I will approximate anywhere from 1.75" to 6.75". The length, width and height varies as well depending on machine size. lets say 1'x1'x2' up to 2.5'x2'x5'. These also vary depending on OEM model. Hope this helps!

Actually there are two radii intercepting each other (see photo in the article). Radius varies depending on machine size. I will approximate anywhere from 1.75" to 6.75". The length, width and height varies as well depending on machine size. lets say 1'x1'x2' up to 2.5'x2'x5'. These also vary depending on OEM model. Hope this helps!


I'm almost sure that you can get an Optical Comparator to fit your needs. I did a web search and found some that will accomadate 500 lbs. along with some pretty good table travel at this website:
http://www.dorseymetrology.com/section5/section5.html

The only question that you would need to answer would be if the product program life would be long enough to justify the cost of the machine, unless you have other products or potential products that you would use the machine for.

All the best!
Jonell

We encounter similar 'challenges' daily. If the exact contour is super-critical we use a CMM, otherwise we use a radius gage set and can always come within a few thousandths if not right on. The gages, made by Starrett, Mitutoyo, etc., are available from many suppliers, such as www.mcmaster.com.

Alex