[Metrology] Spotting a straight edge using a granite master square

[jeremysf]

Hello!

I’m working on machining and scraping a cast iron straight edge from a raw casting.

There is a ton of information out there about spotting one of the faces of the straightedge (like the shoe) using a surface plate for scraping.

When it comes to spotting to adjacent faces (like the shoe and the back) to scrape for squareness, is the common method to use a granite master square sitting on the surface plate?

If so, do I need to rig up some kind of way to clamp the granite square to the ledges of the surface plate? Or is it enough to just blue up the face of the granite square sitting on the surface plate and push and rub the straightedge up against it (while also resting on the surface plate).

I looked in the usual sources like the “Machine Reconditioning” book but wasn’t able to find info on scraping a casting square.

I know another technique would be to make three squares and scrape them square to each other, but that’s a lot more work than I was hoping to do.

Also, I don’t yet own a granite master square but was looking at some 4-face granite parallels (where all four of the long sides are flat, parallel and square to each other) as a “cheaper” alternative to buying a large 5-face square. $900 vs $1700+ In particular a set of 1.5x3x24”. My thinking is to rest the 1.5” face on the surface plate, and blue up and rub against the 3” face. Thoughts?

I tried buying a budget $250 granite square from Shars but only the two long 1” thick working edges are square (ie a two face square) and even those two faces weren’t very square, so I returned it.

Would love some thoughts from more experienced hands before I go investing in some higher end granite accessories.

Thank you for reading!

 

[rabler]

Random thoughts, not experience:
Tom Lipton has some youtube videos on squareness indicators, basically an indicator on a stand with a bumper on the bottom. Fairly trivial device to build. They work by zeroing them to a known right angle.

You could use one on a surface plate without a square reference, to assess the squareness of a "square" if you have to known parallel sides:
If you had 3 sides of something that was close to square, in order, sides, 1,2 and 3.
You can use a surface plate and height indicator to get sides 1 and 3 to parallel. Once sides one and three are parallel, and side 2 is flat but maybe slightly parallelogram, you could use one of Lipton's video's squareness indicator by testing side 2, first with side 1 down on a surface plate, then test side 2 with side 3 down on a surface plate. Comparing the readings from the squareness indicator with those two orientations would allow you to determine any slight deviation from square.

 

[jeremysf]

Good thoughts!

My issue is not being able to check for squareness. I have a bump indicator stand, tenths indicators, a steel Starrett master square, etc.

My issue is being able to correct for squareness and flatness on the order of 0.0001 or less by scraping.

For machining, I would use inspection on a surface plate with an indicator or steel square or steel master square to determine “not square” and then do a better job tramming, cleaning away tiny chips or otherwise indicating in the work. But for “really square and really flat”, other than surface grinding, it would seem like one would have to scrape squareness?

In other words, I know how to correct for flatness when scraping, but trying to figure out how folks correct for squareness when scraping. Like, just use a steel master square and/or indicator to see that something is out of square by a few tenths (or less) and then just sorta guesstimate several passes of scraping for squareness? Maybe re-scraping for flatness in between? Especially interested in any references (like “read chapter such and such in Machine Reconditioning”)

I realize the practical answer is “just buy an already highly precise square” or “you don’t really need to create a straightedge or cast iron square that is THAT square do you?”

But hey, it’s a hobby and I’m interested in learning how it’s done. Or even to learn “people don’t scrape things square”.

 

[Richard King 2]

I would suggest you buy a cast iron angle block to check squareness. Then use one or Toms methods to check it. It's your project but in my opinion you don't need to have the faces of your straight edge square. Those surfaces are what about 2" wide? In all my years Ive never scraped a straight edge like that square, just flat and making the angle an exact 45 degree is also not necessary. We don't scrape the 45 deg. and flat on a part at the same time. For example, on a lathe compound that has 2 dovetail and 2 flat ways next to each other, you scrape the flats first co-planer and then the dovetails and we tip up the straight-edge to meet the dovetails and check the dovetail parallelism with dowel pins and micrometer or indicator gage or a comparator.

 

[NC Rick]

I would want them square. Just from OCD. It seems like clamping the flat shoe part to a precision angle block and scraping would only work if one were to also include the reference square into the program (by scraping it too?