WTB cylinder square

[marka12161]

I picked up some angle plates at an auction that i'd like to evaluate for perpendicularlity. I may want to scrape some of them flat and square but we'll see. I have a nice grade A surface plate but would like a cylinder square in the 12-18 inch height range. I'm not opposed to an import or shop made if it's accurate. Not looking for anything rusty. Please PM me if you have something you think may work.

 

[TGTool]

I picked up some angle plates at an auction that i'd like to evaluate for perpendicularlity. I may want to scrape some of them flat and square but we'll see. I have a nice grade A surface plate but would like a cylinder square in the 12-18 inch height range. I'm not opposed to an import or shop made if it's accurate. Not looking for anything rusty. Please PM me if you have something you think may work.

You do know don't you that you can verify squareness using only parallels and your surface gauge to the accuracy of your best test indicator?

A cylinder square is nice for some things but a direct comparison to the angle plates doesn't give you a quantitative number for how far out of square they might be. You can use it to zero your surface gauge and indicator for comparison, but you can also skip that step and measure directly.

 

[marka12161]

You do know don't you that you can verify squareness using only parallels and your surface gauge to the accuracy of your best test indicator?

A cylinder square is nice for some things but a direct comparison to the angle plates doesn't give you a quantitative number for how far out of square they might be. You can use it to zero your surface gauge and indicator for comparison, but you can also skip that step and measure directly.

My intent was to use the cylinder square as a truth source to zero the surface gage/indicator as you mentioned and then measure the angle plates using that set up. I'm not familiar with using parallels and i don't understand how a direct measurement without first establishing perpendicularity from some truth source. I'm eager to learn something new.

 

[jwmelvin]

With a parallel held vertical, you can check both surfaces. Because the surfaces are parallel, the squareness error on one is opposite on the other. So zeroed squareness (true perpendicular) is halfway between the two faces of the parallel. Just be sure to check the parallel for parallelism and mount it stably so you can gauge both surfaces.

I use a solid 123 block, which is easy to flip for checking both sides.

To use a cylindrical square with a squareness gauge, I thought one needs to do basically the same procedure? I guess you can trust it to be square and just zero to one side.

 

[marka12161]

With a parallel held vertical, you can check both surfaces. Because the surfaces are parallel, the squareness error on one is opposite on the other. So zeroed squareness (true perpendicular) is halfway between the two faces of the parallel. Just be sure to check the parallel for parallelism and mount it stably so you can gauge both surfaces.

I use a solid 123 block, which is easy to flip for checking both sides.

To use a cylindrical square with a squareness gauge, I thought one needs to do basically the same procedure? I guess you can trust it to be square and just zero to one side.

Ahh, makes sense. Thanks. I think you guys just saved me about $500.

 

[TGTool]

In principle, if you imagine a sheet nominally vertical that you know is an even thickness (both sides parallel), you can bring your surface gauge up to first one side and then the other. The indicator will show double the discrepancy from perfectly square so you have both error direction and amount. The physical setup you can manage with what's available.

 

[marka12161]

In principle, if you imagine a sheet nominally vertical that you know is an even thickness (both sides parallel), you can bring your surface gauge up to first one side and then the other. The indicator will show double the discrepancy from perfectly square so you have both error direction and amount. The physical setup you can manage with what's available.

Thanks, I got that from jwmelvin's reply. In my case, if I establish the edges of one of my angle plates to be parallel, I can then orient these edges in the vertical plane and verify it doesn't rock on the surface plate, these parallel edges can then be used in the manner you describe.

Thanks again