Table saw alignment done. I think?

[keith ouellette]

I attached my dial indicator to my miter bar and picked a tooth on the blade and marked it. I tried to do everything the same front to back. I even tried to keep the tip of the indicator on the same part of the carbide.

When I was set I moved the dial so it would read zero at that point and then moved the dial and the marked tooth to the front of the saw and noted the difference.

it was off by 3.5/ 1000ths.

I then set up the indicator so that it touched the smooth part of the blade (below the teeth) and turned the dial so it read zero. Then I rotated the blade and watched the needle.

Doing it this way it changed by 2/1000ths.

Assuming 3.5/1000ths is the correct reading do you think this is enough to make me move the table?

would I be able to move the table on a jet 3hp cabinet so slightly to be close to perfect or would I more than likely make things worse?

[Jason Beam]

that wholely depends on which way it's out.

You will want to ALWAYS avoid anything that pinches toward the fence at the back. :)

[keith ouellette]

Is that at the front of the saw where I stand or the back of the saw?

I will have to do it again because I didn't note wether it was plus or minus 3.5/1000ths.

[Lee Schierer]

I attached my dial indicator to my miter bar and picked a tooth on the blade and marked it. I tried to do everything the same front to back. I even tried to keep the tip of the indicator on the same part of the carbide.

it was off by 3.5/ 1000ths.

I then set up the indicator so that it touched the smooth part of the blade (below the teeth) and turned the dial so it read zero. Then I rotated the blade and watched the needle.

Doing it this way it changed by 2/1000ths.

Most likely the carbide on the teeth is ground at a slight angle on the sides. Your most accurate reading will be the body of the blade or the "flat area" below the teeth. .002" is pretty good and it may not be worth the trouble to try to get closer. Even if the .002" is angling toward the fence at the back of the blade, I doubt it will ever cause a kick back, though you may see some slight teeth marks on the edges of the boards. The fence should be at zero relative to the miter slot or tail away slightly (.002-.005") at the back of the table. If you are worried about getting the saw blade alignment worse by trying to adjust it, then intentionally set your fence so the back is .002" away from the miter slot at the back of the table.

[Jason Beam]

It's most important not to pinch at the back of the saw - where a piece having already been cut could get caught on the uprising teeth at the back of the blade. A couple thou isn't much, but I'm firmly of the position that NO pinching should be allowed. In fact, if you can't get it dead parallel to within .001", then toe AWAY from the fence if you have to.

[Tom Veatch]

To be quite honest, I don't think a 0.0035 taper over almost 10", is really that much of a problem, but it might be noticable. Personally, I'm just anal enough to work at getting it down to less than half that.

I have a Jet cabinet saw, and adjusting the table alignment is really pretty simple. There are 4 cap screws, one at each corner of the cabinet that holds the table. If you back off 3 of those screws and loosen the 4th one, but leave it snug, you can use a dead blow hammer to lightly tap the table top into near perfect alignment.

Once you've got the alignment at 90° is set to your satisfaction, rack it over to 45° and check it again. If the trunnion tilt axis isn't dead parallel with the table top, the alignment can be right on at 90° and off at 45°.

If you find it's off with the blade tilted, you'll need to add or remove shims under the table at the bolt locations to make the correction. If you add shims at the front or remove shims at the rear it will effectively move the front of the blade in the direction of the blade tilt. Likewise adding shims at the rear or removing shims from the front will move the rear of the blade in the direction of the tilt.

So, as an example, if the 45° alignment check shows that the front of the blade is too far right on a left tilt saw, you'll need to add shims at the front, or remove shims at the rear, to move the front of the blade back to the left and correct the alignment. You can avoid messing up the 90° alignment if you remove or loosen only the bolts on the side of table you're going to shim. Leave the other two torqued down.

I used a couple of long thin wooden wedges to make the adjustments. I gently tapped a wedge in beside the two bolts that were loosened/removed so that both wedges"felt" about the same tightness until the tilt alignment is good. Then I used a feeler gauge to get the total thickness of shim needed at each bolt location. They won't necessarily be the same thickness, but should be fairly close.

You can buy shim washers or flat stock in various thicknesses from local sources or online, or you can sacrifice an old aluminum beverage can and cut as many pieces as necessary to build up to the thickness needed. I found that when I torqued down the bolts with a stackup of shims in place, they compressed and threw off the alighnment slightly. So you might want to use a slightly thicker shim to account for that.

I know there are a lot of "weasle-words" like "fairly close", "about the same tightness", "slightly", etc., but that's about as precise as I can be in what is largely a trial and error process. The process can get a little tedious and frustrating, but I found the performance of the saw improved enough that I'm very glad I took the time and made the effort to get the alignment at 90° and 45° almost dead on.

[keith ouellette]

Thanks for the help.

Tom; thanks for the explanation.

OK I did it again and noted that it read 0 at the front (what I call the front is where I stand ) and it read just under 4/1000ths at the back ( where the wood goes out)

Which corner would i tap and which cap screw would I leave snug?

[Lee Schierer]

Thanks for the help.

Tom; thanks for the explanation.

OK I did it again and noted that it read 0 at the front (what I call the front is where I stand ) and it read just under 4/1000ths at the back ( where the wood goes out)

Which corner would i tap and which cap screw would I leave snug?

I would loosen the two at the back and the one to the right at the front. Leave the one at the left front snug. (All directions as you face the saw as if to make a cut)

[Tom Veatch]

... it read 0 at the front (what I call the front is where I stand ) and it read just under 4/1000ths at the back ( where the wood goes out)

Which corner would i tap and which cap screw would I leave snug?

That's what I call the front, also.

Can't tell from your report which direction you need to move the table. I assume that the back of the blade is ~0.004" farther from the miter slot than the front of the blade - but, I don't know which miter slot you're measuring from.

OK, lets assume that the measurement was from the left miter slot and the back of the blade is 0.004 too far to the right. That means the table top must be rotated clockwise to match the miter slot to the blade. If those assumptions are wrong and the top should be rotated counter clockwise, switch left for right in the following directions. As a side note, almost everyone talks in terms of aligning the blade to the miter slot, but with a cabinet saw, you actually align the miter slot to the blade. With a contractor style saw, you do move the trunnion to match the blade to the slot.


Back off the bolts on the back, and the one on the left front. Loosen, but leave snug, the bolt on the right front.
Lightly tap the left front corner of the table to nudge the table in a clockwise direction. Check the results. If no change, you'll need to tap a little harder.
Repeat the tap and check until the readings are where you want them.
Torque up the bolt that you left snug. Recheck the alignment to be sure you didn't move the table when you torqued the bolt. Then torque up the other three loose bolts and verify the alignment didn't change when you torqued up the bolts. If it did, and the misalignment is greater than your tolerance go to 1. above and repeat.
You should now be the proud owner of a table saw with a blade as perfectly aligned to the table as you chose to make it.


EDIT: Just had a thought. Every time I've aligned mine, it's been after moving the saw and the front and rear rails were removed. If the rails are mounted on yours, it might take a pretty strong whack to get any movement of the top, so don't be misled by my instruction to "Lightly tap the....".

[Myk Rian]

Here's a couple pics. If the reading is higher at the back of the blade, adjust CCW.

[Tom Veatch]

Here's a couple pics. If the reading is higher at the back of the blade, adjust CCW.

Depends on which side you measure from.

In the pictures shown, if the distance from the left miter slot to the back of the blade is greater than the distance to the front of the blade, the table of a cabinet saw has to rotate clockwise to move the back of the table to the right and reduce the distance from the left miter slot to the back of the blade. The trunnion of a contractor/hybrid saw must rotate counter clockwise to move the back of the blade toward the left miter slot.

If the distance from the right miter slot to the back of the blade is greater than the distance to the front of the blade, the table of a cabinet saw has to rotate counter clockwise to move the back of the table to the left and reduce the distance from the right miter slot to the back of the blade. The trunnion of a contractor/hybrid saw must rotate clockwise to move the back of the blade toward the right miter slot.

Direction of all rotations taken in a view looking down on the table.

[Myk Rian]

Depends on which side you measure from.

That's why I put the text on the left.

[Howard Acheson]

>> it was off by 3.5/ 1000ths.

That is too much. A same tooth measurement should be a minimum of 0.02" and even better is 0.01" . Either can be attained with a little patience and care. The closer you get to 0.00, the better your saw will cut.

Your process is fine but it is best if you determine which tooth is closest to the miter slot with the measurement taken at the front of the saw. Mark that tooth. Next, find the tooth that is furthest from the slot. If there is more than a couple of thousands difference, the blade is warped and should be replaced or the arbor face is dirty or defective. Loosen the blade and rotate it about 90 degrees, tighten the nut and again determine the closest and furthest tooth. It should be the same tooth or close to it. If not, you have an arbor problem. Now move the marked tooth to the back of the saw and again measure. Now adjust your table for slot parallelism. There is no value in measuring the variation in the blade plate. It's only the blade tips that do the cutting.

Use a flat tip on your dial indicator so you are sure you are measuring from the same spot on each tooth. In fact, it's best to use a recently sharpened blade when making your measurements. Good sharpeners have close tolerance equipment and blade sides are as close being in a perfect plane as possible.

[Lee Schierer]

>> it was off by 3.5/ 1000ths.

That is too much. A same tooth measurement should be a minimum of 0.02" and even better is 0.01" . Either can be attained with a little patience and care. The closer you get to 0.00, the better your saw will cut.

Your post seems to contain several errors.

I hope you don't think that 3.5/1000ths is bigger than .02" It actually comes out to .0035" which is considerably small that 0.01" or 0.02"

Also it doesn't matter if the blade is dull or sharp because you shouldn't try to measure off the teeth when attempting to align the saw. Always make the measurement off the flat part of the disk and mark that spot.

I've never heard of a blade sharpener service that grinds the blade disks, some may grind the side of the teeth although even this is not necessary for good sharpening. Normally on the the face and possibly the top of the tooth is touched. Grinding the sides would reduce the kerf width and clearance between the blade body and the teeth.

It doesn't matter if the blade is warped for saw alingment purposes you should mark and use the same area of the blade front and rear. Any other location on the blade will throw extra error into your measurements.

[Tom Veatch]

That's why I put the text on the left.

In that case, Myk, assuming a cabinet style saw, I respectfully suggest that you have reversed the required table rotation.

For a cabinet style saw, you have to rotate the table to make the miter slot parallel to the blade. If the rear of the blade is too far to the right (+ measurement relative to the left miter slot at the rear of the blade) the table top has to rotate CW instead of CCW to bring the miter slot parallel to the blade.

For a contractor style saws, you have to rotate the trunnion to make the blade parallel to the miter slot. In the case where the rear of the blade is too far to the right, the trunnion has to rotate CCW to bring the blade parallel to the slot.

[Howard Acheson]

>> I hope you don't think that 3.5/1000ths is bigger than .02" It actually comes out to .0035" which is considerably small that 0.01" or 0.02"

My bad, I meant to type 0.001" and 0.002".

Also, whether you measure off the side of the blade tip or under the tip will make no difference to the accuracy of the measurement. Of course, you must use the same blade tip front and back or the same place unter the same blade tip. Either will give you the info needed to determine parallelism.

[michael veach]

many years ago before dial indicators and the like were common place, you would set your fence to the miter slot. Then run a good straight board through the saw making sure it stayed tight against the fence. If the board between the fence and blade had teeth marks coming toward you, the blade was closer at the back. If they curved away, it was closer at the front. If the marks were even both directions, it just meant your blade was not perfect. Still works pretty well today.