Tablesaw alignment thought about a different way

[Don Crawford]

I have jigs, calipers, and dials for this fairly regular job of aligning the table saw blade to the miter slot.

However, for the armchair engineers out there, throw rocks at the following suggestion.

Lets say you remove the blade, loosen what you need to loosen (either the table nuts or the trunions depending on type). Raise the arbor to the top of its travel.

And you have a device which fits on the arbor like a blade (maybe thicker for more rigidity) but it is shaped like an L with a miter bar at the end of the L. Then you would lower the arbor slowly with everything loose until the miter bar is in the slot. Then you would have everything perfectly square and you lock it all down.

Why wouldn't that work?

[Tony Bilello]

That will probably work if you could get the "L" at very exactly 90* and as long as you can get your hands/fingers in where they need to be to tighten everything up.
I don't know if this would be any easier that using a dial caliper. I suggest you make one and see for yourself. then let us know how it went.

[Mike Kees]

I think that in theory this could work. However making this piece would probably need to be done in a machine shop to obtain anywhere near the accuracy needed. Also I cannot see the saw staying in place as it is tightened. With Tony on this one,make it and let us know if it works.

[Don Crawford]

I think that in theory this could work. However making this piece would probably need to be done in a machine shop to obtain anywhere near the accuracy needed. Also I cannot see the saw staying in place as it is tightened. With Tony on this one,make it and let us know if it works.

I am pretty sure I can get this done with a couple bolts and aluminum extrusion I have at home. I will report back.

[Mike Kreinhop]

If you had an old blade that you weren't too fond of, you could drill a hole in it to attach the horizontal bar that mates with the miter track.

[Dan Friedrichs]

All the pieces would need to be ridiculously rigid for this to work. Put a dial indicator on a saw blade while in the saw and push against the blade with your finger - you'll see ten-thousandths of movement. You're typically trying to align a saw to a few thousandths. I don't think you'll get even close to the needed precision (to say nothing about temperature stability, robustness, etc) with aluminum extrusions...

[Frank Pratt]

I don't know that any of the suggested methods would have the rigidity needed for this. It doesn't take much side force to cause a few thou flex in a jig like this unless it was massively strong. It probably takes less than half an hour to get things aligned using a dial indicator & do it old school. Even after making this jig, you still have to do the actual aligning.

Seems like a dubious solution looking for a problem.

[Kyle Iwamoto]

Im an actual armchair engineer. I've used a very expensive production jig to align my saw, and got it close. Nothing but a test cut will tell you if your alignment is spot on. Get close, run some scrap through and adjust accordingly. Your jig, or any of the others will get you close. The final adjustments, run a test cut. Several, I should say. Good luck!

[Andrew Hughes]

I’m trying to figure out what the heck you guys are taking about.
Does that mean I’m not a Engineer :confused:

[Bill Carey]

I’m trying to figure out what the heck you guys are taking about.
Does that mean I’m not a Engineer :confused:

When you find out Andrew, please let me know. I had no idea a TS could be this complex. LOL :cool:

[Andrew Seemann]

All the pieces would need to be ridiculously rigid for this to work. Put a dial indicator on a saw blade while in the saw and push against the blade with your finger - you'll see ten-thousandths of movement. You're typically trying to align a saw to a few thousandths. I don't think you'll get even close to the needed precision (to say nothing about temperature stability, robustness, etc) with aluminum extrusions...

The other thing is that the saw itself doesn't have the rigidity. You'll get flex and slop between the end of the arbor and the top. You can deflect a saw blade by about 5 thousands by just moderate finger pressure. The main thing is that the saw needs to run properly after setup. Honestly, you can do it with just a quality combo square.

When you tighten the corner bolts for the top on a Unisaw with a dial indicator on the blade, you can see it moving by thousandths. Part of the skill/art of setting up a saw is being able to tighten the bolts without throwing the saw out of alignment. That and knowing when to stop adjusting. It is temping to try to set up a saw more accurately than it was manufactured; you spend a lot of time chasing machining tolerances doing that.

[Tom M King]

One year, I needed to spend some money on tools right at the end of the year, and one of the things I splurged on was the Master Plate. I don't know if it's worth the cost, but I was able to get all my table saws dead on with it much easier than trying to use a blade. I would buy one of those before making, or having made, something more elaborate.

http://mastergage.com/display_product.asp?id=4

[Bert McMahan]

You can get a 6" x 12" piece of MIC6 aluminum (super flat) from McMaster for $35 if you want to make your own Master Plate, though at $49 I'd probably just buy theirs since it's got a nice square hole through it already.

[Joseph Quattro]

I have a masterplate bought years ago - I don't remember how much it cost, but have used it for just about every saw I own to good effect.

[Bill Dufour]

Problem is you probably do not have accurate enough measuring equipment to locate the miterbar exactly the distance over from the blade flange. You have to go up 2 inches or so and then over about 6" and measure this within 1/1000 of an inch and then make the jig to that same accurately. I bet the blade flange has more wobble then you think. If you use a blade stabilizer you need to make another jig for that spacing.
Bill D.

[Mark Hennebury]

You don't need any jigs, calipers, dial indicators or custom machined super flat plates;
All you need is an understanding of what you are hoping to achieve, how to achieve it and a block of wood and a big hammer.
Cut the block of wood, examine it.
Smack the table with a hammer to adjust it. ( if its an expensive saw, use the block of wood in between the hammer and table)
Rinse and repeat until it cuts as accurately as you want it to.

[Don Crawford]

One year, I needed to spend some money on tools right at the end of the year, and one of the things I splurged on was the Master Plate. I don't know if it's worth the cost, but I was able to get all my table saws dead on with it much easier than trying to use a blade. I would buy one of those before making, or having made, something more elaborate.

http://mastergage.com/display_product.asp?id=4


This is actually close to what I am talking about right?

[Dan Friedrichs]

This is actually close to what I am talking about right?

It's just a (very) flat piece of metal that bolts onto the arbor. You still use a dial indicator mounted to something running in the miter slot to measure how parallel it is to the miter slot.

[David Buchhauser]

It's just a (very) flat piece of metal that bolts onto the arbor. You still use a dial indicator mounted to something running in the miter slot to measure how parallel it is to the miter slot.

Or you can install a straight, flat piece of aluminum bar into the miter slot, then take measurements with a pair of dial calipers (or digital calipers) at each end between the Master Plate and the aluminum bar.
David

[Andrew Hughes]

I have one of those and it works fine. But so does a combination square up against one side of the miter slot. Rubbing on the same tooth front and back.
Heres a pic just in case your a engineer :)

[David Buchhauser]

Nice Andrew! I like it. I think your method can get you to within 0.005" or less - plenty close for most work, or to use the "5-cut method" to fine tune it.
David

[Mark Hennebury]

I guess i just don't get it; why wouldn't you just clamp block of wood to your mitre fence and cut it and examine the cut surface. You can tell how well aligned it is by the resulting scratches from the teeth; if you leave the wood clamped and run it back and forth past the blade , you can tell by the feel , if its touching the front and back teeth, you can tell by the feel, you can tell by the sound, you can do it with the saw running or off, and run the block past the teeth just the same as your adjustable square or dial indicator. I don't see why you need anything.

The "5 cut method" will only determine the error in perpendicularity of the mitre fence to the axis of travel, it won't tell you anything about the alignment of the mitre slot to the blade. They are two different things. The scratch pattern of the teeth will tell you if the axis of travel is parallel to the face of the blade. If you have the mitre slot correctly aligned with the blade, you will have scratches going down, from the front of the sawblade and scratches going up from the back of the sawblade. If you have the slot misaligned one way the back teeth wont touch the cut wood, or misaligned the other way will burn the wood.

There is an easier and more accurate way of checking the miter perpendicularity; Simply cut a 40" long 2x2 in half and stand it on your saw table top, then slide the two pieces together you will see and can measure the error, it will be magnified 20 times. and as an added bonus, it will also tell you if your blade is perpendicular to your table top. So the one cut method gives you error readings in two dimensions and magnified 20 fold.

[Andrew Seemann]

Nice Andrew! I like it. I think your method can get you to within 0.005" or less - plenty close for most work, or to use the "5-cut method" to fine tune it.
David

I use a digital caliper from the miter slot (mostly for speed and convinience), but I check it with the combo square method. You set it so it just barely drags against one of the front right pointing teeth and run it against one of the back right pointing teeth and listen if the sound and feel of the drag is the same. That should get you within a few thou, which is really more accurate than the tolerance of the saw as a system. My guess is that it will match on some of the teeth but not all; again the method of measurement is probably more accurate than the arbor run out, arbor arm arc tolerance, blade run out, miter slot straightness, etc. taken all together.

And then after we get the blade to within a gnat's whisker of parallel to the miter slot, we go and bolt on huge @ss Biesemeyer fence extensions to the top, put it on a mobile base and bounce it around the shop floor, change the blade, run 8/4 white oak on it for a few hours, and then find all our parallelness and squareness we spent so much time on are all off by 5 thou here and 10 thou there, and yet, the saw works and cuts just fine. . . . .

[David Buchhauser]

Here is a nice video from Rockler on this subject.
David

https://www.youtube.com/watch?v=jvjZKFSXkNA

411940
(https://www.youtube.com/watch?v=jvjZKFSXkNA)

[David Buchhauser]

The "5 cut method" will only determine the error in perpendicularity of the mitre fence to the axis of travel, it won't tell you anything about the alignment of the mitre slot to the blade. They are two different things.



Hi Mark,
Of course - you are exactly right. My mistake. I don't know what I was thinking - I guess I need to wear my "thinking cap" more often.
Thanks for you post.
David