Construction a cylinder...........best method.

[D-Train]

What would be the best method for constructing a cylinder? I am making a program for a gage rod, and I am asked to measure the diameter at the ends of it to see if the part has any wear. Instead I took lots of hits all up and down it in sections. But would it be better to construct circles with those sections, then construct a cylinder with all the circles, or better to just construct a cylinder with all the hits?

Thanks guys.

[Hi-Tech Redneck]

I asked myself the same question a while back...so I did it both ways...Both methods gave the exact same results...

If you have a cylinder constructed from 4 circles of varying diameter, the cylinder diameter will be the average of the 4 circles diameters...

If your looking for wear in certain areas, why not just dimension the circles individually...?

[WolfMan]

It really depends. For example if you want to see the location or concentricity of one circle To another, it is better to measure a bunch of circles (constructed from points or just meas circles).

If you just want a cylinder and you not worried about location of it along the cyl, then measuring a bunch of points and constructing a cyl is the way to go.

Also by having multiple circles you can visually see the form, of the cylinder and have a better visual representation of the form. Imo if you are looking at the wear, it is better to measure or construct a bunch of circles because it will be easier to see at what distance the cylinder is wearing more, also it will be easier to see if the cyl is bending by looking at the location of the circles

[sealevel]

maybe pepper the end(s) in question with lots of hits or use auto-cyl. construct min/circumscribe cyl and graphically report cylindricity.
you didn't specify tolerance, but assume it is fairly tight? what about micrometer in many places? is it battleship drive shaft or a spine rod?
large O.D? small O.D.?
guess it really depends on your gage drawing as to how to approach.
personally, i like auto-cyl + graphical analysis. (mucho hits, mucho levels)

[WolfMan]

Say you measured an auto cyl, and created an analysys, then you see rhat it is much more wear somwhere in the middle, but how do you know exactly where??? Sure you can extract it, or if you measure the circles you will know exactly where the wear is. Just my 02

maybe pepper the end(s) in question with lots of hits or use auto-cyl. construct min/circumscribe cyl and graphically report cylindricity.
you didn't specify tolerance, but assume it is fairly tight? what about micrometer in many places? is it battleship drive shaft or a spine rod?
large O.D? small O.D.?
guess it really depends on your gage drawing as to how to approach.
personally, i like auto-cyl + graphical analysis. (mucho hits, mucho levels)

[WolfMan]

You can measure one circle, then simply by using a paste with pattern , paste 40 more cires.

[Michael Johnston]

If ya really want to split hairs... Measure each segment twice. Set a local zero in X and Y after the first pass then measure a second time for your analysis. Maybe a little overkill but in my opinion it would be the best way to eliminate as much error as you possibly can in each segment.

[dph51]

Say you measured an auto cyl, and created an analysys, then you see rhat it is much more wear somwhere in the middle, but how do you know exactly where??? Sure you can extract it, or if you measure the circles you will know exactly where the wear is. Just my 02

He is looking at wear on the ENDS of the pin.

[dph51]

Measure (or SCAN if possible) circles at each end at various levels.

[JEFMAN]

He is looking at wear on the ENDS of the pin.

I would use auto cylinder to level the pin, then measure circles at each end...

[D-Train]

I asked myself the same question a while back...so I did it both ways...Both methods gave the exact same results...

If you have a cylinder constructed from 4 circles of varying diameter, the cylinder diameter will be the average of the 4 circles diameters...

If your looking for wear in certain areas, why not just dimension the circles individually...?

Good point, I'll bring that up. Thank you.

[D-Train]

It really depends. For example if you want to see the location or concentricity of one circle To another, it is better to measure a bunch of circles (constructed from points or just meas circles).

If you just want a cylinder and you not worried about location of it along the cyl, then measuring a bunch of points and constructing a cyl is the way to go.

Also by having multiple circles you can visually see the form, of the cylinder and have a better visual representation of the form. Imo if you are looking at the wear, it is better to measure or construct a bunch of circles because it will be easier to see at what distance the cylinder is wearing more, also it will be easier to see if the cyl is bending by looking at the location of the circles

no not worried about location or anything, just the diameter of the shaft to make sure it will still fit where it needs to.

[D-Train]

maybe pepper the end(s) in question with lots of hits or use auto-cyl. construct min/circumscribe cyl and graphically report cylindricity.
you didn't specify tolerance, but assume it is fairly tight? what about micrometer in many places? is it battleship drive shaft or a spine rod?
large O.D? small O.D.?
guess it really depends on your gage drawing as to how to approach.
personally, i like auto-cyl + graphical analysis. (mucho hits, mucho levels)

It's just a gage rod, that is 372mm long and 60mm in diameter, and comes to a point at one end, looks like a missile...lol

[D-Train]

If ya really want to split hairs... Measure each segment twice. Set a local zero in X and Y after the first pass then measure a second time for your analysis. Maybe a little overkill but in my opinion it would be the best way to eliminate as much error as you possibly can in each segment.

That's good stuff there Mike, I'll have to do that. Thank you.

[D-Train]

Thanks guys, as always, a huge help.