Build Dilemma

[BuckeyeDennis]

Hi Dennis,

Actually this was a second attempt. The first was way off and I recut all the bevels after recalibrating my 45 degree stops.

For the shorter pieces which are 17 long by 14 wide, I just guided the short edges on the rip fence set on an floating table which was fairly close to the main table.

For the long pieces which are 48 x 14, I did notice the flex issue which was worse with all the weight supported on the rip fence which was almost to the floor on my first attempt. You are right that there's no good way to support this. I tried the adjustable legs that come with the 520, but they didn't really make a difference because I couldn't get them close enough to the low end to do much good. Ultimately, when I recut the bevels on the longer pieces, I switched to my Incra Miter 5000 sled (thanks again, Jack Wilson) and extended the fence and stop to support the bottom corner and clamped the piece firmly in place in two places along the long edge.

I like your tip on cutting slightly over acute. Maybe next time. Or a Festool.

This will solve your problem for only $200.

Dewalt GP.jpg (26.7 KiB) Viewed 4035 times

I know you think you don't have room for a RAS. But they are actually very space-efficient, for two reasons. One, you no longer need clearance room for cross-cutting long pieces on your table saw (or Shopsmith). Two, when you aren't using it, just swing the arm back out of the way, and it's a workbench. Moving the arm can cause repeatability problems with newer/cheaper RAS's, but these vintage cast-iron DeWalts are rigid, precise, and repeatable. And besides, this one is bigger than you actually need.

That said, I'd also like to have a Festool track saw, for cutting sheet goods. But when the time comes, I'll probably get the Makita. According to all the reviews I've seen, they are very close to being as good as the Festool, at about half the price.

[JPG]

I vote for undercut as well. Especially if corner blocks can be used.

[algale]

So, just how much angular error do you need to get say a cumulative 1/16" gap on the last joint in a box that is 48" x 14" and has 8 bevel miters?

I think it is something on the order of 0.003 degrees error per cut, but my math may be off.

[rpd]

[youtube]https://www.youtube.com/watch?v=N4L6Bks_zLg[/youtube]

[algale]

[youtube]https://www.youtube.com/watch?v=N4L6Bks_zLg[/youtube]

On plywood?

[BuckeyeDennis]

So, just how much angular error do you need to get say a cumulative 1/16" gap on the last joint in a box that is 48" x 14" and has 8 bevel miters?

I think it is something on the order of 0.003 degrees error per cut, but my math may be off.

Assuming that you clamp three joints up tight, and leave the one with a 1/16" gap unclamped, then the bevel error is 0.036 degrees, each bevel (assuming that your wood is perfectly straight). That's a mighty small angle, and I'm not sure that's even the geometry you're talking about.

In the photos, you have all four joints clamped, with zero gap on the inside of the joints. So the gap only accumulates across the width of the bevel face. In this case, if the stock is 3/4" thick, a 1/16" gap at the outside of the joint corresponds to a bevel error of 1.7 degrees, again each bevel.

[rpd]

On plywood?

Sure. plywood is wood. Hand planes can cut long grain and end grain.
A miter shooting board could be used to just do the final trim to clean/true up the 45 deg. cut you make with the saw.

[JPG]

1/16 over 3/4 is approximately a 4.8 degree angle.

Contributed from 8 sources = 0.6 degree each.

[JPG]

On plywood?

Sure. plywood is wood. Hand planes can cut long grain and end grain.
A miter shooting board could be used to just do the final trim to clean/true up the 45 deg. cut you make with the saw.

Plywood is also full of abrasive glue.

[BuckeyeDennis]

1/16 over 3/4 is approximately a 4.8 degree angle.

Contributed from 8 sources = 0.6 degree each.

True enough. But the face width of the bevel is 0.75 * SQRT(2), which is 1.06", giving a 3.4 degree total angle.

With all four corners clamped, I was assuming that the carcase was squared up, and each joint had that 1/16" gap. Thus only the two bevels local to a given joint would contribute to its gap.