Hi Tim,

The last guy I discussed this with was also pleased to tell me that the
panel stock he used for the test cuts was "free" to him too. You have
to admit, somewhere along the way the wood cost something. And,
keeping it for test cuts doesn't exactly make it free. At the very
least, it could be made into something if it wasn't being reserved for
test cuts.

I used to think that people who advocate trial and error "test cut"
methods just have an aversion to using precise measurement instruments
(dial indicators, calipers, etc.). However, lately I've seen more and
more of them use these instruments to assist them in the process. So,
I just have to assume that people who advocate trial and error just
plain like the method. After all, they could use the instruments to
adjust the machine directly instead of using them to measure test cuts.
I've heard people describe it as a "skill" so perhaps there is some
sense of pride in finally arriving at a correct machine setting after a
challenging sequence of getting it wrong over and over.

Personally, I hate trial and error. I have a goal in mind (to produce
a finely crafted item of some sort) and I don't want fussy machine
adjustments to get in the way. I want the machine to do exactly what I
tell it to do on the first try - not after a dozen or so test cuts.
And I don't like a shop cluttered with peices being saved for test cuts
or the cutoffs from test cuts. If trial and error is a "skill", then
proficiency must be measured in ones ability to achieve the greatest
accuracy with the least number of test cuts. One could not do better
than to obtain the right setting without any test cuts. That's what I
do but I've been told by test cut enthusiasts that I "cheat". That's
OK by me, my goal is to produce a finely crafted item, not a pile of
test cuts.

This particular procedure (squaring up the fence on a sliding table by
trial and error) is most offensive to me because five test cuts must be
made before any results can be evaluated and any error correcting
adjustment can be made. Then another five test cuts must be made to
evaluate the adjustment to see if it was correct. So, unless you're
astoundingly lucky and get the right setting on the very first try,
you're in it for at least ten test cuts with a panel which (as you
said) should be near the capacity of your sliding table (48" for me!).
I'd call lugging around a half sheet of panel stock on the sliding
table an ordeal which I would definitely be reluctant to perform. And
a half sheet of anyting is hardly a "scrap".

Finally, at my age, there's no hope in resolving a scribed line to
within 0.010" by eye. The 100ths divisions on my scale just look grey
to me anymore. It would involve a magnifier, or perhaps I would use it
only for crude adjustments. If the whole machine were outfitted with
precise adjustments including vernier scales then the topic would be
moot. Nobody would think that there was any virtue in doing test cuts.
But, most woodworking machines have very crude adjustments. I've got
a square and a dial indicator and that's all that I need to do the job
right - with no test cuts.

Thanks,

Ed Bennett

http://www.ts-aligner.com


wrote:
Hi Ed,

I guess the costs depend on your situation. I use an off-cut that is
laying around anyway, and most of it remains when I'm done, so my
material cost is zero. I don't have an 18" precision square, however,
and that would be significant cash for a tool that I would only ever
use to square this sliding table fence.

I agree that the initial squaring can take more time than with a big
precision square, but we're talking the order of a few minutes. Once
the fence is squared, scribe a line on the sliding table so it's easy
to return it to the square position without recalibration.

As for periodic checking of squareness, you just need to grab a scrap
panel, make five cuts on it, break the resulting strip in half and
compare the width of the ends between your fingers. On the other hand,
getting out a square and dial indicator, doing the check and putting
them away takes time, too. If one process takes longer than the other,
I expect it wouldn't be by much. I'd never think to describe making
five cuts on a tablesaw as an "ordeal", but perhaps some woodworkers
do.

I do agree that using the precision square to align the fence every
time you change the fence will give you better squareness accuracy than
using the rotating stop or a scribed line. If I can resolve a scribed
line to 0.010" by eye, and that scribe mark is about 24" away from the
fence pivot point, my squareness would be within about 0.0075" in 18"
compared to your 0.001" in 18". There may be situations in woodworking
where this extra accuracy would be helpful, but I'm at a loss to think
of any at the moment.

As always, your milage may vary.

Cheers,

Tim