I finally decided to make a table saw sled. I was familiar with the five cut method from watching many YouTube videos, some several times. I was sure I had the knowledge to create a really accurate one. That is, until I found each iteration was making things worse…
A table saw sled is a wooden platform that a woodworker builds to make it simple to make cuts at exactly 90° on a table saw. The five cut method simplifies the task of fixing a fence at that exact angle. This post is intended for people who are somewhat familiar with the method, and possibly having problems getting it to work.
I had double-checked both William Ng’s video, 5 Cuts to a “Perfect” Cross-Cut Sled which I had first viewed a number of years ago, and Jonathan Katz-Moses’s video Ultra Square Medium Sized Crosscut Sled which I saw relatively recently, and copied down the formula from the latter, as it was more complete, including instructions for which way to move the fence, depending on the sign of the result from a calculation. Both calculations were the same, with Jonathan crediting the method to William.
The great advantage of the 5-cut method is that it not only tells you how accurate your table saw sled fence is, it tells you exactly how much to move it, and in which direction, when it’s not right. It should result in a fence that is out by no more than a few thousandths of an inch over the length of the fence – typically around two feet. Each time you run a test it provides a new amount to move, and these movements should get smaller each time. Most people nail it in one or two iterations. I planned to get close in three or four, max.
My first run was not bad – the formulae indicated a modest adjustment. After I made the indicated change, my next test was worse. I put this down to a natural talent for general clumsiness, but remember, the beauty of this method is that each iteration provides the total correction required. It doesn’t matter whether it’s the first iteration or the tenth. My third demanded a movement that was beyond the range of my feeler gauges.
I took the night off and had a think. I re-watched the Katz-Moses video, rechecked my measurements and calculations, and decided to start over using a set square for the first test in the new series.
Two more iterations after the fresh start, and I was back in the same mess, and the large rectangle that I was using for the 5 cuts was now the size of a DVD case.
I went back and looked at William Ng’s video. Now, while Jonathan specifically states which way to move the fence to correct either a positive or negative error, William contents himself with merely stating whether the fence is too high or too low.
William Ng: “If it’s a negative number, our fence is too low“
Jonathan Katz-Moses: “If number is (-): move fence towards user“
I took a moment or three to process this because I was sure I was wrong. But eventually, I worked my way through the logic and accepted that my two heroes were giving me diametrically opposite instructions. Surely, though, there would be a correction by now. William’s video has had over 2M views since 2012, and Jonathan’s 86K views in three months. Wouldn’t someone have noticed an error which would guarantee doubling the error on each iteration?
But if neither video was wrong, how could they both be right?
Can you guess? The difference was that Jonathan did his cuts on the right side of the blade, and William used the left! I don’t recall how I worked this out, but I do know I was still doubting myself that I had the answer until I visualized a single cut being made with a very out of square fence. Notice the wedge-shaped off-cut is reversed, depending on which side of the blade it comes from.
Either method will tell you to move the left end of the fence up.
Assuming the fence was too close to the operator, Jonathan’s last cut would be wider at the top (A) than the bottom (B) which gives a positive result (move the fence away from the user), and William’s the opposite, A is smaller, negative result, “fence is too low”, so move the fence away from the user.
Anyone who followed either video exactly, placing the wood as the video author placed it, would succeed. Me, I used Jonathan’s instructions, and just automatically placed the wood on the left side of the blade which seemed to be more natural. I used William’s actions, and Jonathan’s process.
After I worked this out, two more adjustments, and my fence was set up!
I would like to think I haven’t been the only one to follow the Katz-Moses video without noticing he used the right-hand side of the table. I’m sure neither expert thought it necessary to specify where to place the wood, but if they had, I would have saved a whole lot of head scratching! It just goes to show that one can listen to a video, read the instructions and still get it wrong by not watching closely enough.
