In article , "Bob S." wrote:
Larry,

Also read your other post below. To make the point clear, since it is not
irrelevant, the miter gauge needs to be 90° to the miter slot before any
alignment if it is the point of reference. As adjustments are being made,
(this being the key thought) the angle will change (angle of attack) and so
will the distance between the blade and reference point. He must reset
his reference point to the front of the blade and then push it forward to
check for any differences.

The angle doesn't matter, and neither does the distance of the reference
point from the front of the blade. All that matters is the *difference*
between the distance from the reference point to the front, and to the rear,
of the blade. Granted, that difference is easiest to measure when one or the
other of those distances is zero, but that is *not* required.

In any event, your point about resetting the reference point to the front of
the blade applies regardless of the angle of the miter gauge -- that is, it's
no less important with the miter gauge at 90 than with it at, say, 45.

This is exactly what you do with a TS Aligner. You pre-load the dial gauge
at the front tooth, zero the gauge and then rotate the blade and check the
same tooth at the rear. Meanwhile, the dial gauge is being held perfectly
aligned to the miter slot. And the reference point is a small point on the
dial indicator - not a flat edge.

All true. But the angle _does_not_matter_. The TS-Aligner would work just as
well (for aligning table saws, anyway), if the bar that holds the dial
indicator were mounted at a thirty-nine degree angle to the miter slot. The
path traveled by the reference point is _always_parallel_ to the miter slot,
no matter what angle the miter gauge is set at.

Consider a four-sided figure with its corners defined thus:
a) point end of a pencil clamped to miter fence and touching front of blade
b) dead-center of the miter gauge fence when pencil is touching front of blade
c) point end of same pencil when miter gauge is moved to rear of blade
d) dead-center of miter gauge fence in this latter position

These four points define a parallelogram (opposite sides are parallel). If the
miter guage is set at ninety degrees to the miter slot, that parallelogram
happens to be a rectangle. No matter the angle, the path of the pencil point
is dead parallel to the path of the miter guage and hence to the miter slot.
And thus aligning the blade to the path of the pencil guarantees that the
blade and miter slot are parallel, completely independent of the angle of the
miter gauge.

I agree with the statement that the miter gauge can be set at any angle and
the measurement can be taken - for that one adjustment. But... make an
adjustment to the blade in relation to the reference point and the distance
(more/less) will change. He must reset the reference point before making
the next check.

So? That's an artifact of having changed the angle of the miter slot relative
to the trunnion, and is completely unrelated to the angle of the miter gauge
relative to the miter slot. It makes no difference *what* the angle of the
miter gauge is, you *still* have to either reset the reference point to the
front of the blade, or measure the distance between them, before taking the
next measurement at the rear of the blade.

Now as he stated, he was using a pencil so it has a pointy
end and the angle is not measurable between the point and the blade.

The angle between the point and the blade is irrelevant. The *distance*
between them, fore and aft, is _all_that_matters_.

[snip]
So to sum this up:

1. Use a pointy object as the reference and reset the reference after each
adjustment

Bingo! *Regardless* of the angle set on the miter gauge.

2. If you use an edge (such as a cut piece of wood), it needs to be 90° to
the miter slot so as the adjustments approach being parallel, the 90° edge
of the reference point will (ideally) now be flat against the blade with no
discernable gap along the edge.

Checking for gaps fore and aft on a cut piece of wood is functionally
identical to checking the gaps at the front and back of the cut. This in turn
is functionally identical to using a "pointy object" as the reference, at two
places separated by a distance equal to the width of the cut piece -- and
unless that piece is the same width as the blade, it's guaranteed to be a less
accurate test than using a point reference at the front and rear of the blade.

And thus, if you use an edge such as a cut piece of wood, it's imperative that
the miter gauge be set at any angle _except_ 90 degrees, so that the cut piece
of wood presents a _corner_ as the reference surface instead of a flat face.

Which puts you right back to Number 1: use a point reference to start with.

3. Buy a TS Aligner and be done with it....

Works great for those who can afford it. IIRC, the OP is on a limited budget.

--
Regards,
Doug Miller (alphageek-at-milmac-dot-com)

For a copy of my TrollFilter for NewsProxy/Nfilter,
send email to autoresponder at filterinfo-at-milmac-dot-com