How do you get PRECISE bevels using the tablesaw?

[Dan Friedrichs]

I recently built a 12-sided bucket, requiring 24 cuts at 15 degrees to be made on the tablesaw. I thought I had the blade set at pretty close to 15 degrees, but with that many angles to line up, even a few hundredths of a degree can result in a gap.

The bevel angle indicator on my saw is useless - I might be able to get within 1 degree, at best, using it. I have a Rockler "tilt-box"-type digital angle gauge, but it's a piece of junk - it won't produce consistent measurements. Holding a protractor to the blade is difficult to do. So....if you needed to cut a bevel at, say, exactly 15.00 degrees, how would you go about setting the blade?

My only thought is that I regularly set the blade to 90 degrees using a machinist's square. I trust my miter saw to cut accurate angles, so I could use it to put a 15 degree cut on a block of wood, and use that to adjust the table saw blade bevel, in the same way that I use the machinist's square. Is there a better method?

[Howard Olena]

If your table saw is very level, you can use a digital level or digital protractor to dial in the blade. If the table is flat (one would assume), but not necessarily level, you can use a dial protractor. Starret makes a 505A which is relatively cheap, or you can get into really expensive Mitutoyo's (Mitutoyo 187-906 reads accurately to 5 minutes). I'm not sure why you find holding a dial protractor to the blade difficult... If it is because the teeth of the blade are wider than the blade itself, you can make a small notch with a file, in the arm of the dial protractor to clear the teeth.

To confirm the angle, you can bevel both sides of a piece of wood, then cut the piece of wood into 12 (or however many sides you actually want) blocks with a chopsaw. When you assemble your test 'bucket', you'll be able to see if the bevel is off.

Although I've never done it with a 12 sided figure, I can usually dial in a 4 sided figure, using the test method above and the bevel angle indicator on the saw, within 20 minutes or so, but I'm pretty slow...

That said, I don't see why you couldn't use a block of wood as you describe, but I'd still test it a couple times!

[Van Huskey]

Those "cuts" are begging to be done on a router table. Look at the bottom of the page in the link:

http://mlcswoodworking.com/orderstatus/html/smarthtml/pages/bt_door.html

[John Coloccia]

Use a protractor of some kind to set a sliding bevel gauge. Take the gauge to the table saw and set it like that. You should be able to nail it.

[Peter Quinn]

I set a bevel square up to a protractor to set the TS angle, or use an angled block as you describe. It will still require test cuts. I also have an angle cube I bought from Rockler on sale, not sure the brand, but it seems to be pretty darn accurate. I've used it at work to set up three different machines for a job to put the same odd angle on some work from different references; one right tilt saw, one left tilt saw and a right tilt slider. But this was not a 12 sided object. Any time you multiply your error times 12 as you know even small errors become significant. And sometimes the saw adds a little error to the equation, so what its set at and what it actually cuts may be two slightly different things. Test pieces are really required.

[Sylvain Deschamps]

Dan I've had the same problem with making multiple angled cuts for a couple of projects. I used a draftsman protractor which very accurate but consistancy was hard to achieve. I went back to my Wixley "tilt-box"-type digital angle gauge, and as Peter did I tested it's accuracy on different machines and it is quit precise. So I looked at what I was doing wrong at the tablesaw and came up wit h these errors I was doing.
When zeroing the digital guage on the surface of the TS, make sure surfaces are clean without saw dust in between, do it as close as possible to the blade, and make sure the guage is perpendicular to the blade. When you stick it on the blade, place it as close as possible to the teath without touching them. Also make sure it is squared to the blade and place it on a complete flat surface of the blade. What I mean by this is that some blades, as the Freud thin kerf' have a thicker outer rim, and if you position the guage magnets on these two different thicknesses, then you will not have an accurate reading. After doing all this, I practiced and improved my cuts by applying a consistant pressure to the blade as to not influance the angle at all. Good luck!

[Jerry Bruette]

Dan

I just built 5 kaleidoscopes for Christmas presents and needed 22.5 degree angles. Used my Wixey as Sylvain describes. Make sure your tilt-box is plumb to the table when you attach it to the blade.

One other thing to check is if your table is flat and your ZCI is level to the table.

Jerry

[Alan Schaffter]

Dan- Here is one (easy?) way to make multi-sided boxes, etc. if they have an even number of sides like 12-

First, like has already been suggested, accurately set your tilt with a digital gauge like a Wixey which should get you within .5 deg. (Using my mount works well. :) ). Then make a test cut and use a good device (digital protractor type angle gauge, etc.) to verify it is set correctly and cutting the correct angle.

The key, however, is to use a two step assembly. Glue your project (bucket) up in two halves. Depending on the size, run each half over a jointer making both edges of both halves co-planar. Make sure you make the same number of passes with each half or else they will be different sizes (different diameters). Then glue the halves together. With this technique the initial angle cuts do not need to be perfect.

Jointing the edges of the halves can be a little difficult if the project is large. You either need a wide jointer, make a cradle so you can run each half through a planer, or my least favorite, glue sandpaper to a flat table and rub each half over it to sand the edges co-planar.

[Dan Friedrichs]

I like that, Alan. Except it's pretty hard to clamp up half of a bucket :) (with the whole bucket, I can just run a ratchet strap around it)

[Alan Schaffter]

I like that, Alan. Except it's pretty hard to clamp up half of a bucket :) (with the whole bucket, I can just run a ratchet strap around it)

Clamp each half to a board face down. You may need a temporary male mold until all the staves are in place and you may want to add little stops to keep the sides from splaying, then use your ratchet strap.

[Howard Acheson]

This will do it:

http://www.amazon.com/Wixey-WR300-Digital-Angle-Gauge/sim/B001PTGBRQ/2?o=9

[Wayne Hendrix]

I like that, Alan. Except it's pretty hard to clamp up half of a bucket :) (with the whole bucket, I can just run a ratchet strap around it)

Masking or packing tape.

[Alan Schaffter]

This will do it:

http://www.amazon.com/Wixey-WR300-Digital-Angle-Gauge/sim/B001PTGBRQ/2?o=9

Just wait a few days and do it better with this . . .


http://www.ncwoodworker.net/pp/data/500/Wixey.jpg http://www.ncwoodworker.net/pp/data/500/Wixey-2.jpg

[Michael Peet]

Just wait a few days and do it better with this . . .


http://www.ncwoodworker.net/pp/data/500/Wixey.jpg http://www.ncwoodworker.net/pp/data/500/Wixey-2.jpg

Alan, what is that? I can't find anything on Wixey's website.

Mike

[Alan Schaffter]

Alan, what is that? I can't find anything on Wixey's website.

Mike

Not yet, but VERY, VERY soon and a little birdie told me a few may be available even sooner right here on SMC! Notice the new batteries? AAA vs button? How about 2 years before changing? See the mid-case line under the display? Could that mean a tilting display- could be!! I'll hold off with the other neat new features and see if you can guess what they are by looking at the display! I should have one in my hot little hands in a few days- I'll post pics of the real thing.

Disclaimer: product by an SMC advertiser, no financial connection to me.