custard
Established Member
In an effort to get more accuracy into my work I'm trying to square and calibrate my kit on a more regular basis. So I've been looking for simpler and quicker routes to greater precision.
The "Five Cut Method" is well documented (here's a good account http://www.youtube.com/watch?v=UbG-n--LFgQ ), but it's a bit slow and not that intuitive. I found a recommendation from David Best on using a square and a dial test indicator in order to get the fence square to the trajectory of a sliding table, and decided to give it a go.
Here's a video that compares the two methods for squaring a cross cut sled (grit your teeth when you see the lack of blade guard!) http://www.youtube.com/watch?v=UC_iF5pZxmI
I used a flat 375mm x 750mm square from Beta Tools, which checked accurately against my control square and straight edge. However you could use a good quality plastic square, or prop up the thin blade of a normal woodworker's square to keep it flat.
The specifications call for no more than 0.1mm of deviation in 250mm of travel. I put two felt pen marks on the square 250mm apart and measured between these two points.
Here's the set up at the start.
And here it is at the end, showing 0.03mm of deviation.
This method quickly translates into a calibrated adjustment method, but even though it's possible to adjust to zero deviation I find that moving the fence for mitre cuts means that I can't retain much better than 0.03mm deviation over 250mm of travel. The good news is that this error doesn't subsequently move more than that, fluctuating between zero and 0.03mm, which is well inside manufacturers allowed tolerances.
The proof of the pudding however is in the final cut. And even with this 0.03mm deviation the saw produces perfect 90 degree angles as best as I can measure.
For thinner components, like drawer sides, I can just about match this accuracy with a shooting board; but for anything much thicker than about 18mm I find I need to cut the components on a table saw in order to, for example, achieve cross cuts accurate enough to assemble into perfect, gap free squares with Dominos. Using a dial test indicator and an accurate square seems to provide a quick, fuss-free route to maximum accuracy. I'm well please with the technique and can recommend it.
The "Five Cut Method" is well documented (here's a good account http://www.youtube.com/watch?v=UbG-n--LFgQ ), but it's a bit slow and not that intuitive. I found a recommendation from David Best on using a square and a dial test indicator in order to get the fence square to the trajectory of a sliding table, and decided to give it a go.
Here's a video that compares the two methods for squaring a cross cut sled (grit your teeth when you see the lack of blade guard!) http://www.youtube.com/watch?v=UC_iF5pZxmI
I used a flat 375mm x 750mm square from Beta Tools, which checked accurately against my control square and straight edge. However you could use a good quality plastic square, or prop up the thin blade of a normal woodworker's square to keep it flat.
The specifications call for no more than 0.1mm of deviation in 250mm of travel. I put two felt pen marks on the square 250mm apart and measured between these two points.
Here's the set up at the start.
And here it is at the end, showing 0.03mm of deviation.
This method quickly translates into a calibrated adjustment method, but even though it's possible to adjust to zero deviation I find that moving the fence for mitre cuts means that I can't retain much better than 0.03mm deviation over 250mm of travel. The good news is that this error doesn't subsequently move more than that, fluctuating between zero and 0.03mm, which is well inside manufacturers allowed tolerances.
The proof of the pudding however is in the final cut. And even with this 0.03mm deviation the saw produces perfect 90 degree angles as best as I can measure.
For thinner components, like drawer sides, I can just about match this accuracy with a shooting board; but for anything much thicker than about 18mm I find I need to cut the components on a table saw in order to, for example, achieve cross cuts accurate enough to assemble into perfect, gap free squares with Dominos. Using a dial test indicator and an accurate square seems to provide a quick, fuss-free route to maximum accuracy. I'm well please with the technique and can recommend it.
