Shopsmith Forums

Posted: **Sun Aug 22, 2010 2:31 am**

My apologies Mike. Maybe this just comes too easily to me. I will try to elaborate.

If your table saw is set up in reasonable alignment, that is to say that when you set the fence next to the blade, it touches both sides of the blade, (I don't use dial gauges) then when you rip, rotate and rip you will get something very close to a square. Capisce?

Now you mark your diagonal. Do you agree that a diagonal from corner to corner of a one mile by one mile square is likely to create more accurate 45's than a 6" long miter gauge? OK a yard by a yard? Do you understand what I'm saying? If you waver fairly evenly left and right of the line while sawing but come out at the corner, a piece clamped to the resulting wavy line will still be accurately 45deg.
I have bandsawn a lot of miles of curves and straight lines sometimes with changing bevels (this takes two people) so to me cutting a dead on line doesn't seem to be much of an issue but the point I was trying to make is that the longer the diagonal the more accurate the angle.
To be a degree off on a long diagonal you have to miss the line by a lot at the far end. to be off by a degree on a 6" long perfectly set up miter guage, you only need an insignificant wood chip between one end of the gauge and the piece.

I'm all elaborated out. I hope this explains it and sorry I wasn't clearer the first time.

Paul M

Posted: **Sun Aug 22, 2010 1:02 pm**

shipwright wrote:My apologies Mike. Maybe this just comes too easily to me. I will try to elaborate.

If your table saw is set up in reasonable alignment, that is to say that when you set the fence next to the blade, it touches both sides of the blade, (I don't use dial gauges) then when you rip, rotate and rip you will get something very close to a square. Capisce? YES

Now you mark your diagonal. Do you agree that a diagonal from corner to corner of a one mile by one mile square is likely to create more accurate 45's than a 6" long miter gauge? YES. NO COMPOUNDING OF ERROR. OK a yard by a yard? Do you understand what I'm saying? YES If you waver fairly evenly left and right of the line while sawing but come out at the corner, a piece clamped to the resulting wavy line will still be accurately 45deg. A WAIVER IN THE MIDDLE WOULD BE YOUR WORST ENEMY BEST CORRECTED BY PLANING DOWN THE 'PROUD' AREA TO THE LINE IF THE ERROR IS TOO GREAT. TOO GREAT WOULD ADMITTIDLY HAVE TO BE PRETTY BAD ON A LONGER SURFACE. (I.E. 3/16" ERROR AT 12" WOULD BE 1 DEGREE) I have bandsawn a lot of miles of curves and straight lines sometimes with changing bevels (this takes two people) so to me cutting a dead on line doesn't seem to be much of an issue but the point I was trying to make is that the longer the diagonal the more accurate the angle. UNDERSTAND COMPLETELY
To be a degree off on a long diagonal you have to miss the line by a lot at the far end. to be off by a degree on a 6" long perfectly set up miter guage, you only need an insignificant wood chip between one end of the gauge and the piece.

I'm all elaborated out. I hope this explains it and sorry I wasn't clearer the first time.

Paul M

Absolutely no apology needed. You just cracked me up (literally) with your 'detailed' explanation. Reminded me of a recent post on how to network a printer.........in 1 sentence.

My comments above.

What REALLY had peaked my interest was you saying 'The 45 deg is very accurate if you just rip a square'. As a newbie, I've alway assumed an accurate square cut involving using either miter gage alone OR combination of using the fence first followed up by the miter gage. I thought you had a method that assured accuracy using just a ripping op and I was very interested in hearing how. I see now that it assumes your initial cut edge is reletively square to the edge you reference after you rotate 90 and that the piece is long enough where the error won't matter.

I initially thought JPG's post discussing the parallelogram clarified but then THAT created some issues since a diagonal across corner will not measure 45 degree BUT, if the parallelogram is reasonably close to a square and it is large enough, the error won't matter to get an accurate 45.

So, bottom line, I'm good. (and wiser). Thanks for being so long winded.

Posted: **Sun Aug 22, 2010 1:20 pm**

I did take for granted that if one were to want to rip a square on a table saw that one would start with a piece that had one "factory" corner or a verified square corner.

Paul M

Posted: **Sun Aug 22, 2010 10:48 pm**

shipwright wrote:I did take for granted that if one were to want to rip a square on a table saw that one would start with a piece that had one "factory" corner or a verified square corner.

Paul M

Thank You for that! For an instant, I thought a square might be possible without a 'verified'(or assumed) corner!;)

Just make sure one side of that corner is against the fence on the first cut, and the other side of that 'corner' is against the fence on the second cut.

Details Details Details!

Shopsmith Forums

Cutting 45 degree angles on large stock?