More On Disappointed In MiterSet

[dusty]

The table needs to be braced with an extension table and tubes also.
Angle AND length need to be right on for perfect fits of course
For my needs I'm looking into a small adjustable sled just for my octagon pieces for my clocks

If the source of the problem is in the SS, then wouldn't a sled require the same bracing?

I don't mean to be speaking for Cliffy but when dealing with this instability thing you must acknowledge that everything is referenced to the blade. If the carriage and headstock are effectively locked to the Way Tubes much of that instability is overcome. Even more so if you tie the may main to an extension table (or two).

I seldom need the precision that is being discussed in this thread but nonetheless I tie the main table to the right hand extension table ALMOST all of the time.

The only improvement brought by having a sled is that the fences (on the sled) are set in relationship to one another. BUT, if that sled moves because the table moved some level of precision is lost. The relationship between the sled and the blade has changed. If the angle is not altered the length will be.

If extreme precision is required, the Shopsmith just may not be the right tool.

[DLB]

One thing about the MiterSet that I keep going back to: I don't have a MiterSet, so I 'zero' my miter gauge per the SS procedure, or some improvement on it. Meaning I set it 90 degrees to the plane of rotation as best I can. Before I do that I've set my table so the miter slots are as near to parallel to the plane of rotation as I can. But, there is some small angle of deviation between the slots and the plane. So when I set my miter gauge to the plane of rotation, I am offsetting for that small deviation angle. But if I use a MiterSet I'll zero my miter gauge relative to the miter bar, and slot, and therefore not to offset for the small deviation angle. Any way I look at it, it is two different things even though the difference is small. The only way to make the difference zero is perfect alignment of the table slots to the plane of rotation.

Because of that, IIWM I'd try the MiterSet in a different way than the demonstrators used it. I'd cut my pieces using the same reference edge and face for all cuts (just as I do on MiterPro). Thinking of each board as having an "A" and a "B" miter, I'd cut all of the "A"s at 22.5 and all of the "B"s at -22.5. (Again, same idea as the MiterPro.) I'm not saying this will work, it still depends on the MiterSet, along with everything else, having a very high degree of accuracy. But, in the manner of the MiterPro and similar jigs or sleds, it allows some of the errors to cancel. The MiterPro doesn't work by cutting two perfect opposing 45 degree angles, it works by cutting two complementary angles where errors cancel.

Note - I'm using 22.5 degrees for convenience above, and because that is how I think the angles have been referred to in this thread. IIRC this is 67.5 degrees for an octagon.

- David