help squaring up "crosscut table"

[mound]

Years ago I made a large crosscut sled (20" width capacity) and I remember the process of dialing it in using the 5-cut method, and it has served well since. Haven't really thought about that process again since.

For my kitchen project now I need to do lots of repeatable cross-cuts on parts wider than my 20" capacity, so I figured rather than making a huge and unwieldy sled for the table saw, I'd instead try to make a crosscut "table" for my track saw.  I whipped up this design in SketchUp and then built it:

[attachment=40490]

I'm having a hell of a time getting it precisely square though and am unsure why.  I do have a protractor attachment for my track which works great for one-offs, but wanted a setup like this for the stop-block feature so I wouldn't have to measure or mark each piece for repeated cuts. I'd think this approach should work fine.

The concept is probably pretty self explanatory from the image, the stop-block on a fence with a measuring tape, I've dialed that all in well. Rest the workpiece against the stop for length and tight to the fence. Then pull the track tight to each of the blocks that are mounted to the top of the fences, and it should be square if everything is put together right.

My process for testing this is to rest the workpiece (originally ripped on the table saw) as shown in the image, cut off about 1/2" from the end,  then rotate the workpiece counter clockwise 180* and then cut 1/2" off the other end. 

When I do this, if I measure the length on each side, they are the same. But if I measure corner to corner, they are off by about 1/16"

So I installed a small drywall screw in  the right-hand side of the stop block at the far end, allowing me to micro-adjust where the track lands, in case the far end of the track needs to be slightly further to the right.  I do this and repeat the test, and while the lengths are still the same on both sides, the diagonals are still off!

Is this just a matter of making tiny tweaks in my "micro adjustment screw" in the right direction until it's dead on or am I missing something fundamental in that this isn't using a "5 cut"  method?

Thanks!

[BrentDH]

Are both edges of your track parallel?  I looks like you are referencing the left edge of the track to the stops but then referencing the right edge to make the cut.  Maybe I am missing something but just throwing out an idea.

[Roly]

Place a board on the table as shown but not against the stop block and draw or scribe a line on the board from the saw guide, rotate the board and draw another line close to the other one, if parallel the guide is square is square, if not adjust guide.   If the saw blade is not square to the guide the front of the blade may cut a slightly different path than the back.   When you complete a cut the front of the blade may be cutting a different path than the back when the drop off occurs .     Roly

[jdonhowe]

You're creating a parallelogram (opposite sides equal, diagonals not so), and *almost* square. Tweaking should square it up. What I find incredibly useful in squaring things up is a large (12") clear plastic drafting triangle- cheap, dead-on accurate, and easy to see any gap in the angle you're testing.

[mound]

Thanks all, I did figure it out.. It did come down to tweaking my micro-adjustment screw until things dialed in. ALSO, and this was probably as much a problem, I wasn't paying attention to saw-dust building up between the fence and workpiece.. duh, I should have looked at that right away.. Now with it dialed in and making sure to clear dust between cuts, it's working well!

[MstrCarpenter]

Years ago I made a large crosscut sled (20" width capacity) and I remember the process of dialing it in using the 5-cut method, and it has served well since. Haven't really thought about that process again since.

For my kitchen project now I need to do lots of repeatable cross-cuts on parts wider than my 20" capacity, so I figured rather than making a huge and unwieldy sled for the table saw, I'd instead try to make a crosscut "table" for my track saw.  I whipped up this design in SketchUp and then built it:



I'm having a hell of a time getting it precisely square though and am unsure why.  I do have a protractor attachment for my track which works great for one-offs, but wanted a setup like this for the stop-block feature so I wouldn't have to measure or mark each piece for repeated cuts. I'd think this approach should work fine.

The concept is probably pretty self explanatory from the image, the stop-block on a fence with a measuring tape, I've dialed that all in well. Rest the workpiece against the stop for length and tight to the fence. Then pull the track tight to each of the blocks that are mounted to the top of the fences, and it should be square if everything is put together right.

My process for testing this is to rest the workpiece (originally ripped on the table saw) as shown in the image, cut off about 1/2" from the end,  then rotate the workpiece counter clockwise 180* and then cut 1/2" off the other end. 

When I do this, if I measure the length on each side, they are the same. But if I measure corner to corner, they are off by about 1/16"

So I installed a small drywall screw in  the right-hand side of the stop block at the far end, allowing me to micro-adjust where the track lands, in case the far end of the track needs to be slightly further to the right.  I do this and repeat the test, and while the lengths are still the same on both sides, the diagonals are still off!

Is this just a matter of making tiny tweaks in my "micro adjustment screw" in the right direction until it's dead on or am I missing something fundamental in that this isn't using a "5 cut"  method?

Thanks!

A bit simpler than the five cut method to check 90* is to just make a cut on a wide board with parallel edges (just as shown on your jig sketch)  than flip one piece over, slide them together, and see if there's a gap. If there is one, it's double the amount you're out of square for that width board; the accuracy gets better as the board get wider.

[mound]

A bit simpler than the five cut method to check 90* is to just make a cut on a wide board with parallel edges (just as shown on your jig sketch)  than flip one piece over, slide them together, and see if there's a gap. If there is one, it's double the amount you're out of square for that width board; the accuracy gets better as the board get wider.

clever!

[rwe2156]

Nice set up.

I use a panel cutting sled I can square up to 30" wide panels. If you're not familiar, its a sled that rides to the left of the blade with a fence on the far side and a couple toggle clamps. I have one runner in the miter slot, and another that rides along the table edge.

But I really like your set up it will be better if you're squaring up a panel longer than 48".

Wondering what you think about using a T track along one edge with a runner screwed to the bottom of the track so it would slide like a table saw fence?

[mound]

Nice set up.

I use a panel cutting sled I can square up to 30" wide panels.  If you're not familiar, its a sled that rides to the left of the blade with a fence on the far side and a couple toggle clamps.  I have one runner in the miter slot, and another that rides along the table edge.

But I really like your set up it will be better if you're squaring up a panel longer than 48".

Wondering what you think about using a T track along one edge with a runner screwed to the bottom of the track so it would slide like a table saw fence?

Thanks. Yah I chose this way because I'm crosscutting parts out of 96" long plywood rips and the track-saw is perfect since that large piece of plywood can sit still during the cut. 

I thought about that idea too, at first, essentially a movable track, but really the idea here was that the blade remains in one location laterally while the stop-block moves the way a table saw fence would.  Having that one and only kerf in the fence also means I can position a mark right at it for cutting if I don't want to use the ruler/stop-block for some reason..  A moveable track would chew up the fence and base over time.

[Alan S]

What I like about the 5-cut method is that you don't depend on anything being straight or parallel before the test. I like to adjust my sleds and leave space for sawdust below a screwed on face to the glued fence by shimming behind with tape if needed to make it precise and straight.