Making a 36" straight edge from scratch - ideas?

[Lignin]

Greetings All

Well I haven't done this myself so...

I purchased just about all the various metalworking books Lindsay Publications put out, a sad loss to me when they closed-up.

The re-prints of older texts talk of generating surface plates, precision straight edges, and squares from scratch. They all use a three piece method. As I understand it, you get two to match fairly close and then take the third and match it to the others. You keep exchanging and comparing the three to each other, as you remove finer and finer discrepancies you will finally end up with a flat surface plate, matching straight edges and matching squares. The greater care you take, the finer the degree of agreement between the various items. In case it's not clear you do surface plate to surface plate, square to square, and straight edge to straight edge. I don't recall any of the of the old books stating this, but my thought is; if you did this with sufficient care then the straight edges and the limbs of the squares would also match each other and the surface plate also.

Matching shorter pieces of thread in a similar manner and then using that to generate a longer thread is the method that was used to make the first precision threads. Precision meaning that the threaded piece matched another threaded piece very closely.

all the best

Olin

Thanks. That seems to be the received wisdom. Those old timers were sure patient - a quality which is fast disappearing.

 

[oldgoaly]

I was borrowing a 3' to make some tests (cylinder heads), but I wanted one of my own. I happened across a 24" long 1/4"by 2" pc of cold rolled that was fairly close. How I was told was 3 panes of glass with fine sand paper (garnet paper?) the variations evened out the differences. I would check it against the 3' Starrett and it was good. It took time but then I had more of that than money. Years later I came upon a bargain for a Starrett pink granite 18 by 24, got the straight edge out and check it on that by golly it is as flat as the granite! btw I did the narrow 1/4" edge. A couple of oldtimers told me how to do it, I believed them, always thought I better listen so I don't have to go buy some pc of junk from a 3rd world country. Every time I use it I think of Bill and Greg, there are many things they showed me how to do and sometime I hope to show my son or any young folk who have more time and no money, it might just keep them off those silly video games for a few minutes! Go ahead and try it!

 

[Optic Eyes]

Greetings All

Well I haven't done this myself so...

I purchased just about all the various metalworking books Lindsay Publications put out, a sad loss to me when they closed-up.

The re-prints of older texts talk of generating surface plates, precision straight edges, and squares from scratch. They all use a three piece method. As I understand it, you get two to match fairly close and then take the third and match it to the others. You keep exchanging and comparing the three to each other, as you remove finer and finer discrepancies you will finally end up with a flat surface plate, matching straight edges and matching squares. The greater care you take, the finer the degree of agreement between the various items. In case it's not clear you do surface plate to surface plate, square to square, and straight edge to straight edge. I don't recall any of the of the old books stating this, but my thought is; if you did this with sufficient care then the straight edges and the limbs of the squares would also match each other and the surface plate also.

Matching shorter pieces of thread in a similar manner and then using that to generate a longer thread is the method that was used to make the first precision threads. Precision meaning that the threaded piece matched another threaded piece very closely.

all the best

Olin

This can be done, I have done this, it is not easy, but start with stress relieved steel