How to set odd anagles on miter gauge?

[dusty]

INTERPOLATE

Michael, if you are using the Shopsmith miter gauge the protractor is about a 3" arc. It is at the edge of that 3" arc where you are reading the degree markings. Hard, to say the least, to set the pointer midway between two marks (22° and 23°). But if the arc was say 12" long, it would be somewhat easier to distinguish the midway point.

To achieve this wider arc place a rigid strange edge (12" steel ruler) to the face of the protractor. With the protractor set at 22° place a mark at the far end of the ruler. Now move the protractor to 23° and place another mark. Midway between those two marks is 22.5°. This won't be perfect but it will be more accurate than what you started with. To achieve even greater accuracy - use a 3' straight edge or somewhere in between.

splitting hairs.png (85.41 KiB) Viewed 8192 times

I used a 5 degree separation here for display purposes. A 1 degree separation would obviously be more difficult however that can be overcome by making the measurement further away from the pivot point.

[reible]

This idea is sounds good, the "math" makes sense and has been used for years and years and years....... so what if we take it to the shop and give it a go.

I pulled out my incra express with the 1000HD miter gauge. The miter gauge can be set to 1/10 degree. I added a woodpecker straight edge just to make this as accurate as I could.

I taped a sheet of white paper down so I could see the lines clearly and then used a pentel .5mm pencil to draw a line at 29 degrees. A second line was drawn at 30 degrees. At a foot out I then added a short line at what I thought was mid way between, visual bisection if you will.

The then I set the miter gauge to 29.5 degrees and added a another line. Not a perfect match but the lines touched, that being my guess at 29.5 and the miter gauge line at 29.5.

Let me say that the error was less then 1/10 degree, far less.

So if you have say the V120 with one degree settings getting to a half degree using this method is very doable. Since the degrees lock in you don't have the visual potential errors on the stock miter gauge getting the two actual degree marks to bisect. My gut tells me the error will be larger but still more then likely acceptable for a lot of projects.

Do keep in mind over a very short distance a half degree isn't much but when you are out 12" is getting to be a lot.

Way to go dusty!

Ed

INTERPOLATE

Michael, if you are using the Shopsmith miter gauge the protractor is about a 3" arc. It is at the edge of that 3" arc where you are reading the degree markings. Hard, to say the least, to set the pointer midway between two marks (22° and 23°). But if the arc was say 12" long, it would be somewhat easier to distinguish the midway point.

To achieve this wider arc place a rigid strange edge (12" steel ruler) to the face of the protractor. With the protractor set at 22° place a mark at the far end of the ruler. Now move the protractor to 23° and place another mark. Midway between those two marks is 22.5°. This won't be perfect but it will be more accurate than what you started with. To achieve even greater accuracy - use a 3' straight edge or somewhere in between.

splitting hairs.png

I used a 5 degree separation here for display purposes. A 1 degree separation would obviously be more difficult however that can be overcome by making the measurement further away from the pivot point.

[ChrisNeilan]

Dr., My brain hurts!

[dusty]

Thanks for taking the time to validate this approach, Ed.

The white paper approach is an improvement over what I did. I used a fine point magic marker and scribed the lines on the table. I then measured the separation between the two lines at the edge of the table and marked the midpoint. Then I set the miter gauge (V120) using that center mark and read the angle on the V120 protractor. "Dead on for wood working purposes". Maybe not dead on enough for segmented turnings or inlays but pretty darn close.

Chris - don't think about...go out in the shop and do it. It will come clear with no additional pain.

[thunderbirdbat]

I was thinking that graft paper would work a little better since it would be easier to figure out the midpoint between angle lines.

[dynabase]

Cool solution Dusty and Ed. I have a standard SS miter and a V120 miter so your solution will work perfectly. To expand the concept a bit I was thinking about using a piece of paper and drawing an X from corner to corner as an alignment line to reference when you tape the paper down. If you did that you could use the same paper for left and right angles and you could move and re-tape the paper and maintain the reference alignment with the miter.

Michael

[dusty]

Cool solution Dusty and Ed. I have a standard SS miter and a V120 miter so your solution will work perfectly. To expand the concept a bit I was thinking about using a piece of paper and drawing an X from corner to corner as an alignment line to reference when you tape the paper down. If you did that you could use the same paper for left and right angles and you could move and re-tape the paper and maintain the reference alignment with the miter.

Michael

I've play with this a little bit more. It does not work if done exactly as I described. The measurements cannot be taken at the tables edge or at the edge of the miter slot. Taking the measurements (reference points) there introduces a very slight amount of error.

The measurements need to be taken on an arc. An arc of any size will work and the vertex of the arc needs to be the center of a miter track (the pivot point of the protractor in your miter gauge.

Ed Reible experienced some minor error that he experienced and I believe he took his measurements as I described. Had he taken those measurements on the arc I expect that the error will be even further minimized.

Interesting application of simple mathematics (geometry).

[edflorence]

In the pre-CAD days there was a draftsman's trick that still works great for finding any angle. I have used this method when laying out the angles for cutting compound miters with good results. For example, when cutting crown molding, the miter gage needs to be set at 54 ¾ degrees, and this method makes finding that angle easy.

You start by making a template, which consists of laying out a right triangle with a base leg of 10 inches. I have used cardboard and plywood for template material. Then, to determine the height of the adjacent leg of the triangle you divide 10 by the tangent of the desired angle. Pretty much any pocket calculator, even the one on a cell phone, will give you the trig functions. Using that height, lay out the second leg of the triangle at right angles to the first and then connect the ends of the legs to create the third side (hypotenuse) of the triangle. Place the template against the saw blade, bring the miter gage up to the template and lock the gage at the required angle.

In the case of the 9 ½ angle, I figure the template will be a triangle with a base of 10 inches and a height of 1 11/16ths.

Granted, this method requires care in laying out the template, but if made of sturdy material it will only need to be made once. And it works for any and all angles.

[JPG]

An 'arc' is unnecessary. a point on both lines equidistant along each line from the 'vertex' will suffice. Connect those two points and bisect that line and draw the final line(half degree) from the vertex to the bisecting point.

Actually dividing the line into fractional lengths will give fractional partial degrees.

[dusty]

An 'arc' is unnecessary. a point on both lines equidistant along each line from the 'vertex' will suffice. Connect those two points and bisect that line and draw the final line(half degree) from the vertex to the bisecting point.

Actually dividing the line into fractional lengths will give fractional partial degrees.

Good point. The arc is not required. But if the measurements are made at points "equal distance from the vertex" are those not the same points as those taken on an arc whose radius is that same length. I think this is apples and apples.

Bottom line: I was wrong in my initial comments about taking measurements at the tables edge.