Cutting a perfect triangle on the TS

[Brian Behrens]

I'm building a sled for my table saw, and adding a piece of plywood cut as a right triangle for miter cuts. It will attach when needed via star knobs to the kreg track I put on the sled.


My goal is to cut a perfect right triangle for this attachment. Running through it in my head, I can easily square the 90 corner on my sled, but I can't figure how to accurately cut the hypotenuse. I've done the 5 cut method to test the sled and it is square to .002 over a 20" cut, so I'd like to get the miter attachment highly accurate also .


Yes, I can cut it easily with a miter gauge, but I'm looking for a tip to make the 45s perfect. I want to be able to cut any miter from either side of the jig, not have to keep track of which piece was cut on which side in order to cancel out any error in the jig. I can't see a way to cut the hypotenuse to where the right angle legs are the same length within a few thousandths.

This is the exact sled I am working on. I was considering using William Ng's method of cutting the miter base, but that would change the design, and I want to learn a new trick if anyone has one. Thank you for any help

[James Tibbetts]

Half of a square piece will do it. Whatever method works for you.. miter gauge on a table saw.. hand saw.. straight edge with a circular saw

[johnny means]

You could attach it to the sled in place and rip it off using the sled sideways.

[Pat Barry]

Once you can cut a perfectly square or rectangular piece you have your saw set up. Only two cuts required then to make a 90 degree corner.

[Brian Behrens]

So once I have my perfect 90 angle, how do I cut the long side of the triangle and have the two legs be exactly the same length. To a few .000 tolerance. That is the part I can't figure out. Any of the methods mentioned will likely result in one side being longer than the other, and defeat my goal of true 45s on the corners .I don't want a 44.9 and a 45.1

[Art Moore]

How about a Drafting triangle to align your miter gauge to the blade?

[Bradley Gray]

If you can cut a perfect square then draw the diagonal cut line. Temporarily screw or tape another piece of ply with one edge on the line and run the opposite edge of piece 2 against the table saw fence.

[Steve Demuth]

If you nail the right angle at the apex, then:

1. Getting the hypotenuse perfect is a matter of getting the two short legs identical in length and perfectly seated against the main fence. I would focus on that. In fact, for your application, I'd focus on getting that right with an inch or so of the hypotenuse on each end straight and 45 degrees true and undercut the rest of the hypotenuse - makes it less likely that a woodchip will make it difficult to seat the miter block against the fence.

2. Even if you don't get the hypotenuse perfect - say it's a few tenths of a degree off - you can still cut perfect miters as long as you always cut your pieces with the show side up (or down - just needs to be the same for every piece). This will cause you to cut the left miter of every piece on the right side of the jig and right miter on the left side, so they will always add to 90 degrees. A couple tenths of a degree variance in "rotation" won't impact the length of the miter cut enough to be noticable.

[Jamie Buxton]

Cut it as best you can using the saw. Then use a handplane to adjust the triangle. Set the plane to take off only a fine shaving. Take an extra stroke or two or three at one end of the hypotenuse, and one or two long strokes to clean up.

Or if you don't like handplaning plywood, cut the plywood roughly to size, edgeband the hypotenuse with a little solid lumber, then trim that. Or if you're averse to hand planing, but have a power jointer, use it to adjust the triangle.

[Brian Behrens]

Thanks for all the tips. A member sent me a video that shows a great technique for doing this to a high level of precision. It is a very interesting method. I'm not sure if it's proper to post the name of the video here, but I can pm it to anyone interested.

[Pat Barry]

Thanks for all the tips. A member sent me a video that shows a great technique for doing this to a high level of precision. It is a very interesting method. I'm not sure if it's proper to post the name of the video here, but I can pm it to anyone interested.
Not sure why both legs need to be exactly thesame length. You could put a series of equidistant markings along each rail ... ala ruler for example. That way you could make any lengths necessary

[Steve Demuth]

Not sure why both legs need to be exactly thesame length. You could put a series of equidistant markings along each rail ... ala ruler for example. That way you could make any lengths necessary

I was referring to the legs of the triangular insert. For the sled to cut perfect miters, that triangle needs to isoceles.

[Simon MacGowen]

I think I know the video you are talking about. All the suggestions given here are no match to that video. By the way, all those suggestions will not give the results you are looking for, even though many are good in theory.

Any method that relies on pencil marking, handplane (even Paul Sellers won't get the kind of tolerance you want) and what not is not fool proof at all. That video will get what you want.

Simon

[Steve Demuth]

Any method that relies on pencil marking, handplane (even Paul Sellers won't get the kind of tolerance you want) and what not is not fool proof at all. That video will get what you want.
Simon

I don't know the video, since it wasn't posted in this thread, but if as he says Brian already has the right angle on his jig a highly accurate 90 degrees, then measuring the short legs to be equal and adjusting the hypotenuse with a handplane will get him quickly and easily as close to perfection as the material in his triangular insert (wood, after all) permits. You can easily plane to within less than .01" variance at the ends of the hypotenuse, and a hundredth there will translate into less than a thousandth gap in an actual miter - or it would, if he were working with a material in which measurements less than a thousandth of an inch were meaningful. Pencil marks are beside the point - you can use an actual rule to measure the legs rather than plane to a mark, or just cut a stick to length and use that. And again, you don't need the actual hypotenuse perfect - you only need the ends to seat the jig so the short legs of the triangle are truly isosceles. You can plan a swale into the entire rest of the hypotenuse's length as long as you get those seating points where the jig insert sits against the sled fence right.

[Simon MacGowen]

0.01" variance is a lot when compared to the video the OP refers to. I watched that video a while back and my memory might be wrong and I think that video woodworker was talking about 0.001" tolerance or something like that.

I am also a hand tool user and all I can say is, other than hit or miss/by luck, there is no way a handplaned approach could match that youtube method (verified by a digital caliper).

If we are talking good enough mitres, the suggestions we see in this thread are good enough. But, if I understand the OP's requirements correctly, none of them will meet his needs.

Simon

[Pat Barry]

I was referring to the legs of the triangular insert. For the sled to cut perfect miters, that triangle needs to isoceles.
you need to mount the 90 fence exactly 45 degrees and precisely with the 90 apex at the centerline of the saw blade kerf. To do this you need a bit of adjustability to the 90 degree fence. That is what the system you have pictured allows. You can loosen the knobs skew the fence and lock it down when it is spot on.

[Martin Wasner]

Make another sled with a crotch to put 90º corner into. I've got one for making corner blocks. I just used a miter box to make the crotch in a scrap of melamine. It rides against the fence and I hang onto the triangle I'm cutting, pushing both through the saw.

[Bill Space]

I am having trouble understanding why some information (link, name or whatever) that would allow me to view that video, which apparently others on the forum have viewed, is not included in this thread.

I would enjoy viewing it, if for no other reason I would understand what others are referring to.

I suppose it is fair to assume that since the member PMed the link to the video to the OP, that he does not want it viewed by the rest of us... Not sure why though...

Bit confused here, but without question do respect the wishes of the video's producer...:)

[Steve Demuth]

I am also a hand tool user and all I can say is, other than hit or miss/by luck, there is no way a handplaned approach could match that youtube method (verified by a digital caliper).


Is there a reason no one is posting the video link? Kinda an odd discussion we're having, with the actual method not yet in evidence.

[scott kinninger]

https://youtu.be/AgVthkUE4AU

[Steve Demuth]

0.01" variance is a lot when compared to the video the OP refers to. I watched that video a while back and my memory might be wrong and I think that video woodworker was talking about 0.001" tolerance or something like that.

I am also a hand tool user and all I can say is, other than hit or miss/by luck, there is no way a handplaned approach could match that youtube method (verified by a digital caliper).

If we are talking good enough mitres, the suggestions we see in this thread are good enough. But, if I understand the OP's requirements correctly, none of them will meet his needs.

Simon

So, having seen the video now, it's a clever, although terribly involved way of getting two legs of an isosceles right triangle identical. But note that in his demo, he would get close to .001" agreement on this test piece if he was off by the .01" I suggested would be good enough in creating his jig. I could duplicate (with respect to the jig) what he did with a good ruler and a hand plane quite easily.

At any rate, if the OP builds the sled he showed (which was taken from a youtube video as well), then his accuracy in his actual miter joints is going to be determined as much by how accurately he seats the insert dead center in the blade, and carefully against the fence (to create actual isosceles triangle) as it will .001" accuracy in construction.

[Simon MacGowen]

So, having seen the video now, it's a clever, although terribly involved way of getting two legs of an isosceles right triangle identical. But note that in his demo, he would get close to .001" agreement on this test piece if he was off by the .01" I suggested would be good enough in creating his jig. I could duplicate (with respect to the jig) what he did with a good ruler and a hand plane quite easily.

At any rate, if the OP builds the sled he showed (which was taken from a youtube video as well), then his accuracy in his actual miter joints is going to be determined as much by how accurately he seats the insert dead center in the blade, and carefully against the fence (to create actual isosceles triangle) as it will .001" accuracy in construction.

Was that the one the OP referring to? This is the first time I watched it. The OP can confirm if he was talking about the same video.

The one I watched -- clever than Ng's, I would say -- was a different one. I will see if I can find that one again.

Simon

[john bateman]

It seems to me you could just tweak the hold downs until you get your perfect miters, then glue a shim to the end of the leg that is short, so that it always contacts the fence equidistant.

[Sam Murdoch]

Thanks for all the tips. A member sent me a video that shows a great technique for doing this to a high level of precision. It is a very interesting method. I'm not sure if it's proper to post the name of the video here, but I can pm it to anyone interested.

Don't know why it would be a secret that can't be posted in the thread - but will be happy to get a PM with that video link. Thanks.

[Brian Behrens]

I wanted to make sure it wasn't against the rules or that people thought I was promoting the video . I guess it's ok. Just search YouTube for Kings Fine Woodworking extreme crosscut sled . The triangle method is towards the end of the video

[Simon MacGowen]

I wanted to make sure it wasn't against the rules or that people thought I was promoting the video . I guess it's ok. Just search YouTube for Kings Fine Woodworking extreme crosscut sled . The triangle method is towards the end of the video
That is exactly the one I watched. I liked the way he taped the blocks to set up the cut. As long as you can the perfect squares (much easier than a perfect triangle), you can create a perfect triangle using his approach.

Thanx for the search key.

Simon

[Brian Behrens]

This is the video that answered my question. Skip to around 32:30.
https://youtu.be/njsD5W6fcI0

[Charles Lent]

For very accurate angles I use my MiterSet jig to set my miter gauge to exactly the angle needed. Add a jig to the miter gauge to hold the piece being cut and make the cut.

Charley

[William Young]

I haven't seen the video mentioned above, but why not keep it simple with a little trial and error. Start with a plywood scrap with one straight edge. This will be your hypotenuse. Now set your miter gauge to 45* as accurately as you reasonably can. Set the plywood hypotenuse against the miter gauge fence and cut one short leg of the triangle. Now flip the plywood blank and cut the other short edge. The angle between the two cut legs should be nearly 90*. Check it with your square and, if necessary, adjust the miter gauge and make the cuts again (shave just a little off of each leg). Check and adjust until you get the angle to the precision you want. Now with the miter gauge precisely set, cut the final triangle from a fresh blank.