Read this if you made an A+ in geometry class

[Bill Bryant]

I'm starting this new thread on the topic of my table saw's blade alignment problem now that I've done everything the Delta service people told me to do along with reading several Sawmill Creek posts very carefully and lying awake thinking this thing through from every angle I can conceive.

1. This is a Delta contractor's saw.

2. The trunnion rods are now solidly in the same plane.

3. The trunnion brackets snugly hold the trunnions with no play.

4. With the saw set at exactly 90 degrees, I put a dot on the blade about 3/16 from a tooth and used PALS to align the blade so that this dot is inside .001" when measured as far forward as possible and as far to the rear as possible.

5. I took these measurements using a modified A-Line-It tool in the right miter slot. The modification is a small perpendicular bar that screws on to business end of the dial indicator and has the test point screwed into the other end so that I can take measurements right at the plane of the table with the ZCI removed.

6. After being totally certain that the 90-degree saw blade was exactly parallel to the miter slot (inside .001") I lowered the saw blade and took new readings of the dot on the blade. The result is clear. Lowering (NOT TILTING) the blade makes it move out of the original plane and move both closer to the fence (.010") and slightly heeled in toward the fence.

7. When cutting a ZCI I end up with a slot that is slightly but visibly curved--the middle of the slot bows toward the fence, the front of the slot bows away from the fence.

8. It seems to me that the only thing that could be causing this problem is a non-parallel relationship between the axis of the blade arbor and the axis of the height adjustment pivot. If this is indeed the case, it is a machining error that can't be fixed, and the issue becomes one of looking up manufacturing tolerances and perhaps going through the hassle of trying to get the saw replaced. I've already discussed all of this with Delta service, and I've been told that the problem has to be a trunnion alignment issue. I've questioned that with them and been told it isn't possible that it could be anything else.



I need the best geometry experts on the forum to think this through with me and tell me if I'm missing something here.

Much thanks in advance to the top brains on this forum.

Bill

[Cliff Rohrabacher]

When you tuned the table and blade up did you take care to mark the spot on the blade where the indicator would register and then rotate the blade to make that same spot register to the indicator when you checked at the other side of the blade?

All blades have some error in them. Failure to rotate the blade allows that error to become part of your set up deceiving you into thinking you have a good tune up when in fact you don't. Then when you spin up the blade the error manifests in a blade arbor that is not 90-Deg to the miter slot.

[Chuck Wintle]

First what is a ZCI? And after you lowered the blade how were you able to measure the dot on the blade? Is it possible that something needs to be shimmed or adjusted?

[Bill Bryant]

I measured runout at the arbor earlier and it's too small to register--under .0005 at least.

The blade is a Ridge Carbide TS2000. I measured runout off this blade on the arbor and it's under .001 all the way around at 3/16 from the teeth. I can leave a dial indicator point on the blade as I rotate the blade with the motor pully and the indicator wiggles less than .001 all the way around.

A ZCI is a zero clearance insert.

After lowering the blade I measured the dot at 12 o'clock using the same dial indicator set up, this time in the middle instead of at the back or in front.

When a blade is raised and lowered, it should stay in the same plane. Clearly mine does not.

[Jim Chilenski]

Bill,

If I understand what you have stated then I would have to agree that the blade is moving out of alinement with the 'Y' axis as its center changes its position on the 'Z' axis. Sort of like this.

Blade at lowest height | (When looked straight down upon from above the table)

Blade at highest position / (When looked straight down upon from above the table)

If this is what is happening, then "Yes, it has to be a machining error".
But how to prove this? The only thing that I can think to do is to get a perfectly flat, if there really is such a thing, 1/4" steel plate that is 10" x 6" with a 5/8" arbor hole drilled close to one of the 10" sides. Mount this plate on the arbor and using your dial indicator measure the movement of the plate as you raise and lower the height of the plate.

This may prove the error in the height adjustment.

Jim

[David Epperson]

I'm thinking that the fence end of the pivot is higher (closer to the table top) than the other end, and not so much but also a bit out of square with the fence/miter slot. Now this does pose the issue as to whether or not the trunnion is parallel with the blade arbor and also whether or not that misalignment can be fixed or not. But not being able to personally examine a saw exactly like yours I can't tell. I'd bet that there is a tweak that could correct it though.

[Chuck Wintle]

I would insist Delta do something to correct the problem. They should either replace the saw or replace the defective parts. Which model of saw do you have?

[Bill Bryant]

I would insist Delta do something to correct the problem. They should either replace the saw or replace the defective parts. Which model of saw do you have?

Delta 36-982

[Lee Schierer]

7. When cutting a ZCI I end up with a slot that is slightly but visibly curved--the middle of the slot bows toward the fence, the front of the slot bows away from the fence.

Bill it seems to me that even though your blade is parallel to the miter slot the blade is not 90 degrees to the table. Check to see if the blade is tilted to the left. I normally use a drafting triangle to check to insure the blade is 90 degrees to the table before starting any other alignment. Most saws have stops that will get you close to 90, but a little sawdust will affect their accuracy, so I always check.

[Bill Bryant]

Bill it seems to me that even though your blade is parallel to the miter slot the blade is not 90 degrees to the table.

I set the 90 degrees with a Starrett square. I took 20 minutes doing this, using a light source behind the blade and confirming my setting by measuring on both sides of the blade. The stop is backed way out--you can go to 93 degrees if you like.

[Richard Dragin]

I just went and looked at my Delta contractor saw. It seems to do the same thing if I am understanding your procedure and duplicating it. I have an A-line-it and a Wixey digital gauge so I know I am at 90. I notice that when I turn the wheel the blade moves out more than when it is static.

So I guess the question becomes, besides the ZCI, how does the saw cut? Does this .010 differance effect the use of your fence measurment and if you make two cuts without moving the fence but at differant blade heights how far off do they differ?

Now my saw is 20 years old and I understand you wanting your new saw to be perfect but is it really off? The measurement you are describing is new to me as most folks are happy to be lined up with the miter slot at 90 and 45 and don't look beyond that.

[Bill Bryant]

most folks are happy to be lined up with the miter slot at 90 and 45 and don't look beyond that.

I want to be able to change my mind about saw blade height during a cutting run and not have some of the parts come out .01 wider than others. I also want to the blade not to go out of parallel with the fence as I lower it. What good does it do to set up the blade within .001 of parallel up at 3" if I do most of my cutting with it down around 1" or 2" where the blade is no longer parallel and burns the wood?

[Chuck Wintle]

Bill,
Just looking at the exploded parts diagram for your saw and was wondering if the weight of the motor could be affecting the blade arbor somehow?

[Ken Fitzgerald]

Bill,

The curve in the ZCI tells you that your blade or the trunnion is NOT lowering in a perfectly perpendicular plane with respect to the table top or the blade is bent.

Raise the blade to a height where the gullets are just clear of the table top and you can measure at that surface just below the gullet. Manually rotate the blade and check the run out on the blade. If you don't see a remarkable runout on the blade, then I'd bet your trunnion isn't lowering perpendicular to the surface of the table top.

[Mike Marcade]

Is it possible the axis of your blade arbor is not perpendicular to the axis of the blade raising mechanism. Then when you set your blade square to the table the raising mechanism would be a little bit off from straight up and down.

In particular I am talking about the axis of the bearing recesses in part #207 in your exploded parts diagram.

Maybe you could try angling the blade slightly like 1 or 2 degrees and see if the distance the blade travels (the .015" you talked about) changes?