tablesaw bevel angle error

[Michael Panis]

Hi Folks,

I'm seeing a problem with my tablesaw (Jet cabinet saw) that I cannot explain.

First, I'll say my tablesaw top is flat and the workpiece I'm cutting is flat. I've adjusted the blade's tilt with an engineering square such that the blade, relative to the table, is dead on 90 degrees. The square rests directly on the blade, not the teeth.

After making a cut, I notice a small gap at the bottom of the workpiece, as shown in the diagram below:
237222
If I put a square against the cut edge and look at the gap along the 6" edge, the gap is about 1/32", which is about 0.3 deg off of 90 deg. Since I'm using this edge as a reference for boring a hole that extends 6", I can't live with that variation.

Also, if I simply rest the piece against the blade after making a cut, I can see a small gap at the bottom.
And, if put both the workpiece and the engineering square used to setup the blade on the table, with the square against the cut edge, I can see the gap.

In short, the angle of the cut does not match the angle of the blade.

I've checked the arbor runout. It is zero. Shockingly zero. I'm impressed every time I've done that.
I've checked the flange that the blade sits against. It's runout is also zero.
The blade itself, a Forest Woodworker II small kerf, has about 5-10 mil variation across its surface, but I don't think that's the problem because:
a) I made the same cut with another blade and got the same result
b) Even if the surface of the blade has some variation, that doesn't mean the teeth have the same variation
c) Even if the teeth have some variation, I would expect that to cause some wobble or a messy cut, but not an incorrect angle. And the cut is clean.

Any ideas?

---Mike

[John Coloccia]

Are you ripping or cross cutting? Anyhow, check that the blade is square to the miter slot. If the blade is angled, it will cut an angle. Any angle will make a mini cove cut.

[John Coloccia]

Oh, and make sure you ZCI is level with the table and not causing a bump you can pivot on.

[phil harold]

Have you Checked all Adjustments?
http://www.woodmagazine.com/wood/pdf/delta/ds-038free.pdf

Without actually looking at the saw I am thinking the blade is not parallel to the mitre slot
Go over all adjustments and then make a test cut again
good luck!

[Michael Panis]

It is a crosscut using an Incra mitre guage. Good point about the mitre slot alignment. I've checked it before, but not recently. Thanks!
(And I did make sure the Zero clearance insert is not interfering)

[Bill Huber]

Phil, I don't see that the blade being parallel with the miter slot should have anything to do with it, if you were cutting a 45 degree cut the face to the edge should still be 90 degrees, could you give some more detail on that?

I had somewhat the same problem at one time, I was checking the blade with an 18" drafting square and things were just fine. The problem was my extension on the left side was high on the left edge. So with a long board things were fine but with something short it was not cutting square. I readjusted the extension and now both long and short are square.

[John Coloccia]

Phil, I don't see that the blade being parallel with the miter slot should have anything to do with it, if you were cutting a 45 degree cut the face to the edge should still be 90 degrees, could you give some more detail on that?

I had somewhat the same problem at one time, I was checking the blade with an 18" drafting square and things were just fine. The problem was my extension on the left side was high on the left edge. So with a long board things were fine but with something short it was not cutting square. I readjusted the extension and now both long and short are square.


Imagine the blade 90 degrees to the miter slot. You will get a 10" radius on the cut. As it approaches parallel to the slot, the cut becomes straight. The path the wood takes needs to be aligned with the blade.

[Lee Schierer]

Imagine the blade 90 degrees to the miter slot. You will get a 10" radius on the cut. As it approaches parallel to the slot, the cut becomes straight. The path the wood takes needs to be aligned with the blade.
Nicely described. I agree check the blade alignment to the miter slot. Then don't forget to align the fence once the blade is aligned to the slot.

[Dick Brown]

Make a cut on a disposable piece of wood with a fair amount of wood both sides of the cut, slide the cut edges together with them on a flat surface and see if there is a gap at the bottom. If there is no gap, your blade isn't @ 90 degrees.

[phil harold]

The path the wood takes needs to be aligned with the blade.

If you are sawing a dado 1/8 x 1/2 deep if the fence is parallel you get a groove an 1/8"
If the fence is 15 degrees out of whack with the saw blade you get a cove 1.25 inches wide by .5 inches deep
His 1/32" is either the back or front of the blade cutting a small arc
@ 1/32" it looks more straight than an arc


When having cutting problems with equipment always make sure all your alignment is correct in all directions, once set they rarely go out of adjustment,
then look for other problems warped boards, table extensions not flat, warped cast iron...

[bob hertle]

Imagine the blade 90 degrees to the miter slot. You will get a 10" radius on the cut. As it approaches parallel to the slot, the cut becomes straight. The path the wood takes needs to be aligned with the blade.

I believe that would be a 5" radius! (Unless OP has a 20" saw)

Bob

[joe milana]

xxxxxxxxxx

[Bill Huber]

Imagine the blade 90 degrees to the miter slot. You will get a 10" radius on the cut. As it approaches parallel to the slot, the cut becomes straight. The path the wood takes needs to be aligned with the blade.

I am not saying you are wrong, I just can't see it. If the slot is not parallel to the blade the wood would go into the back of the blade as you made the cut or away from the blade as you made the cut.

If you are just cutting on the front teeth of the blade and those teeth are at a 90 degree angle to the table the face to end should still be at 90 degrees. The angle of the cut along the face will be off but the face to the end will still be at 90 degrees, how will that change, I just can't see it yet.

[John Coloccia]

I am not saying you are wrong, I just can't see it. If the slot is not parallel to the blade the wood would go into the back of the blade as you made the cut or away from the blade as you made the cut.

If you are just cutting on the front teeth of the blade and those teeth are at a 90 degree angle to the table the face to end should still be at 90 degrees. The angle of the cut along the face will be off but the face to the end will still be at 90 degrees, how will that change, I just can't see it yet.

The teeth don't move straight. They move in an arc (5" radius, of course, not 10" like I originally said...thanks Bob :) ). The profile left on the piece of wood that is cut is the 2d projection of the blade on a plane perpendicular to the direction the wood is moving. So as an example, put a piece of cardboard as a movie screen at the back of the table saw. Shine a light on the blade, and you get a straight, vertical line on the cardboard. That's the profile it will cut. Now twist the trunion so that the blade is crooked....say, it's heeled in at 10 degrees. It won't be a straight line anymore. It will be a gentle arc. At the extreme of 90 degrees, you will get a 5" radius circle, and that's what will be cut into the edge of the board.

Over a short distance, like 3/4" for example, a small misalignment won't look like an arc. Over a short distance, it will simply look as though the blade is tilted a bit.

[Bob Wingard]

If the blade is not parallel to the slot, what you can get is a mini version of a cove cut ... while it is conceivable, I would think the misalignment would be pretty obvious without need of any precision indicators.

[phil harold]

If you are just cutting on the front teeth of the blade and those teeth are at a 90 degree angle to the table the face to end should still be at 90 degrees. The angle of the cut along the face will be off but the face to the end will still be at 90 degrees, how will that change, I just can't see it yet.
If the front of the blade is closer to the slot then the middle
The bottom of the board will be shorter than the top

[Andrew Pitonyak]

Imagine the blade 90 degrees to the miter slot. You will get a 10" radius on the cut. As it approaches parallel to the slot, the cut becomes straight. The path the wood takes needs to be aligned with the blade.

I don't have time to do the match now, but, at 90 degree cut with a 10" blade on 1.5" material would leave a different of about 0.23" from top to bottom.

I did not draw the picture or take the time to make sure that the difference is truely related to the sin of the angle, but, if so, then the sin(x) * .23 = 1/32. Solve for x. So x = arcsin(1/32/.23) = roughly 8 degrees. Again, I did not draw the pictures and verify the geometry, and it is certainly something to check, but I am more inclined to agree with Bill Huber that it is probably related to the saw top.

I am curious, what happens if you cut on the other side of the blade?

Also, if there is something with the table top, you should be able to see that when you simply set the piece of wood against the blade while the blade is not spinning. If you do not see the gap, then it may be something as simple as blade deflection.... but you saw this with multiple cuts.

[glenn bradley]

I'm in the camp that suspects material travel to blade to not be perpendicular. Getting a bit of a cove cut as a result.

[Michael Panis]

I was sold on the idea of the blade's not being aligned with the mitre slot, but measured it last night and it is aligned to within 0.001" so that does not appear to be the problem.
I did notice two things, though..
a) I had tightened the arbor nut with a wrench (I know...I know...hand-tighten only). Loosening it and tightening it by hand seemed to decrease the problem. I believe that what was happening was that the excessive force put a slight dish in the blade. This means that when the arbor was exactly parallel to the table top, the dish caused the top of the blade to be closer to the left of the blade. Since you align the surface of the blade to the top, to get the dished surface as close as possible to the edge of the engineering square, I had to tilt the top of the blade to the right. The teeth, however, only travel around the circumference of the blade. By aligning the dished surface to the table top, I had misaligned the teeth such that the teeth at the top of the blade were closer to the right while those near the table stop were closer the workpiece.

Furthermore, when I put a new blade on the saw, I didn't re-align it, so it produced the exact same bevel error!

b) Even though I aligned the blade to the tabletop with the zero clearance insert removed, I believe there may have been spots on the ZCI that were slightly higher, almost imperceptibly, than the table. This might have added to the problem, although to a smaller degree.

I still have a little more alignment and testing to do...

Thanks for all the suggestions...

---Mike

[Jay Maiers]

It is a crosscut using an Incra mitre guage. Good point about the mitre slot alignment. I've checked it before, but not recently. Thanks!
(And I did make sure the Zero clearance insert is not interfering)

Does this mean that it only happens when you are using the mitre gauge, and not when you use the fence?

Does it happen with both miter slots, or just one of them?

Is it possible that the motor torque is pullng the saw out of square while it's running?

Have you checked to make sure the Incra is really at 90 degrees? (I don't think this one matters, but one never really knows...)

[Alan Schwabacher]

Have you made check suggested earlier in the thread, to put the two cut parts back together on a flat surface and check to see if they meet? This will distinguish between a tilted but flat cut from a tilted blade, (if the pieces meet top to bottom) and a wedge shaped cut (pieces meet at top but not on bottom) that would result from a misaligned movement of stock. Note that stock movement can be misaligned either if the blade is out of parallel with the miter slot, or if the stock shifts on the miter gauge as you cut. Sandpaper on the miter gauge fence prevents stock shifting.

[Myk Rian]

Not only should the blade be parallel with the miter slot, so should the fence.

[Steve Baumgartner]

Despite the theoretical appeal of coving due to misalignement to miter slot, I don't think that is a credible explanation. Think about it: the depth of the cove is equal or less than the difference between the distance between the blade and miter slot where the teeth pass through the table surface vs at the center of the blade (equal if the blade height is exactly the thickness of the wood). A 1/32 cove would mean at least a 1/32 skew, which would mean a 1/16 difference between front and back of blade!! If his saw was that far out, there would be other problems such as burning and massive chipping at the back side of the cut.

Something else, such as the suspected cupping due to overtightening is going on.

While thinking about this, the following strange idea crossed my mind: the coving is away from the blade at the bottom on both the miter gauge side (cut by the front of the blade) and the offcut side (cut by the back of the blade). The angles are the same instead of complementary! That means if you do the traditional cut-an-flip test, it will look like you are cutting perfectly at 90 degrees despite the cove!
Steve

[glenn bradley]

I was sold on the idea of the blade's not being aligned with the mitre slot, but measured it last night and it is aligned to within 0.001" so that does not appear to be the problem.

Oh darn. That was my vest guess ;-)


a) I had tightened the arbor nut with a wrench (I know...I know...hand-tighten only). Loosening it and tightening it by hand seemed to decrease the problem. I believe that what was happening was that the excessive force put a slight dish in the blade. This means that when the arbor was exactly parallel to the table top, the dish caused the top of the blade to be closer to the left of the blade. Since you align the surface of the blade to the top, to get the dished surface as close as possible to the edge of the engineering square, I had to tilt the top of the blade to the right. The teeth, however, only travel around the circumference of the blade. By aligning the dished surface to the table top, I had misaligned the teeth such that the teeth at the top of the blade were closer to the right while those near the table stop were closer the workpiece.

This is a little confusing as the blade washer should have and opposing profile to the arbor flange. This disallows blade plate distortion. Is this not the original washer or arbor perhaps? Good luck with your pursuit and please let us know what you find ;-)

[phil harold]

This is a little confusing as the blade washer should have and opposing profile to the arbor flange. This disallows blade plate distortion. Is this not the original washer or arbor perhaps? Good luck with your pursuit and please let us know what you find ;-)

Since you checked all the other things during a tune up, we can now take those out of the equation

Glen is right on washers
On my jet I have 4" Blade Stabilizer on each side of the blade,
then the stock blade washers

Like these:

http://www.toolstoday.com/p-5322-saw-blade-stabilizers.aspx?variantids=7485,0&affiliateid=10054&utm_source=AmazonProductAds&gdftrk=gdfV22404_a_7c1444_a_7c6337_a_7cSTF_d_4

and with all threads, we like to see pictures
so show us you saw, zero insert, and a pretty piece of wood

[Bill Huber]

I agree, if you can cove the blade by turning the arbor nut you have a problem with that but I am not saying that will fix the other problem but it should be corrected.

[Chris Parks]

Check the square? Also so to get a longer reference on the saw blade, crank it down very low, remove the ZCI and use the square upside down so it lays against the blade under the table. You then see the blade of the square over the whole of the saw blade.