LumberJocks Woodworking Forum banner

Square a Table Saw Sled

5778 Views 12 Replies 9 Participants Last post by  woodsmithshop
I've made a few table saw sleds since I got into woodworking a little over a year ago, some worked better than others, but none of them worked great. My biggest fear, and least favorite part, of making a sled is getting it square to the blade.

On my last sled, for cutting miter joints for boxes, I placed the sled on the runners as square as I could, made a cut 3/4's up the sled, then used a framing square to attached the fence. The problem was that since the blade was angled at 45 degrees, it made it difficult to align the square with the kerf. I felt that laying the square flat on the sled and "eye balling" the gap along the kerf wasn't accurate enough, especially as the square would slide around when trying to adjust the fence.

Are their better methods for aligning the sled fence to the blade? What's some of the techniques ya'll use to square your sleds? Any help would be a appreciated!


1 - 13 of 13 Posts
I did mine pretty much like Barry. His instructions are great.
For sleds I've made (or kits I've assembled) I have used the 4-corners test. I like it because it magnifies the error by 4, allowing you to set the angle set more precisely than a framing square, etc.

For example, for a cross-cut sled, you just take a scrap (I use MDF), say 10" square, with one straight edge. Mark that edge "A". With that first edge "A" against the fence, make a crosscut, and mark that new crosscut edge "B". Now rotate the piece 90 degrees, placing "B" against the fence, so you make a new crosscut "C" (on the edge that hasn't yet been cut). Rotate one more time, with "C" against the fence, to cut the last side "D". Now, rotate to place "D" against the fence, as if you are going to recut the original edge "A". Instead, move the piece over so you cut off about a 1/4" test strip.

If that strip is a perfect 1/4" thick all the way along, your angle is perfect. If it gets thinner towards the end of the cut, your angle must have been a little more than 90 degrees; if it gets thicker, the angle was less than 90. Tweak your sled setting in the correct direction, remembering that the angle error in your fence is only 1/4 the angle error in your test strip. Repeat the test. If you're obsessive like me, you could use a micrometer to compare strip thickness at the beginning and end of the cut.

If your jig is for cutting a 45 degree miter, you can still use the technique with a variation. Cut mitres on four different strips of scrap, bring them together like a frame, and any gap represents 4x the error. Tweak and retest, etc.

Hope that helped and wasn't too confusing!
See less See more
I follow the crowd to praise CessnaPilotBarry's method. Worked great for me. I think the key is the drafting triangle. It has a much more clean and readable edge than the framing square or the combination square I was using.

Mark's 4 corners test is a good way to verify your results, and also to see if you need to tweek the adjustment. The method Barry detailed to attach the fence is the perfect compliment to this check and verification.
Thanks for the link ot Barry's blog, I hadn't seen that yet. I would have never thought to clamp the fence down and tap it into place.

Thanks Mark for the info on the 4-corner test. I saw that mentioned the other day and wasn't sure what it ment. I'll definitly be trying that out when I make my new sled.
1st, let me apologize for the length of my 2-cents. Barry did it right. The only thing I did different than Barry, was, I didn't rabbit the runners. I cut them for a perfect fit width-wise, but about 1/16" less depth. I elevated the runners using flat washers of the same thickness, and double-stick tape. The washers raise the runners up nicely to set the sled on. I just lined up the edge of the sled with the edge of my saw table. After the tape and sled are stuck together, I flip the sled over and shoot some brads into the runners. They're not going anywhere. When attaching the fence, I bored the 1st hole to the left as a perfect fit to the screw/bolt, (I've done this on my router table fence also. It serves as a pivot point) and the other 3 (I used a total of 4), I oversized them to the next size drill bit. That should give enough play on the right to dial in the fence to square. P.S. Also, I keep a box of canning wax (very affordable) in the shop for makin things slick that need to be slick. Just rub em on the runners, and zoom, yer slidin that sled. I wouldn't be without a sled or 2. I actually have a large and a small version. One more thing, I also did do the 4-corner test that Mark talks about, and I've found that as long as you have a good square, you wouldn't have to do this. But, It is a very good way to "double-check"
See less See more
Guys, I know this is an old thread but the link mentioned to Barry's blog seems to have disappeared. Is it still available?? Thanks Pat
here is the link for the 5 cuts to accuracy on a sled.

very good video.
Smitty I'm familiar with this video was looking more for how to assemble the sled. If there were any differences than the wood whisperer version. Thanks for the reply, Pat
This is the method that I used:

Got me to within .00025 of an inch over approximately 21". Definitely worth a try.

Steven, what do you use to measure a gap of 0.00025?
JesseTutt: Good question. You probably think that I am blowing smoke, but here is how I came up with that result: I had a piece of plywood that measured approximately 21" x 21", I placed one edge of the plywood against the fence of the sled and made a cut (less than an 1/8 of an inch, just enough to square it up), then I placed that newly cut edge against the fence and made a cut on the second edge in the same way as the first. I did this to all four sides, referencing the newly cut edge against the fence each time. After the fourth cut (which brings me back to the original side), I cut a piece that is approximately 1" wide. I then measured each end of the cut-off with a dial caliper (being sure that everything is level and measured in the same way at either end). The difference was .001, which is 4 times the actual error due to the fact that the error is compounded over 4 cuts, so .001 divided by 4 is where I came up with .00025. Any way you look at it, that is very accurate.

If you want a better (and more detailed) explanation, check out the link.

See less See more
the link that I posted does show how to put the sled together, and apply the runners, then square it up
1 - 13 of 13 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.