OldTools Archive

Recent Search Bios FAQ

264714 Micah Salb <msalb@l...> 2018‑01‑11 Re: When Good Squares Go Bad
Dear Tom:

There is wisdom in your words.  While I’m not sure that I will accept your
Great-Uncle Charley’s advice, some of the rest of what you wrote seems worth
taking in.  In fairness, though, it is not particularly advice that I need.  I
am not as persnickety in my work as perhaps I should be (depending on whom you
ask).  But when making a bookcase, for example, and a shelf is at 89 degrees
where it tenons into one side but 91 degrees where it tenons into the other
side, you’re going to have some problems that will not be easily fixed.

What I’ve been doing with my not-square squares is drawing a line once, flipping
the square, drawing another line, and splitting the difference.  It’s a pain.
And enough difference that not doing it would surely make a difference.


From: Thomas Conroy [mailto:booktoolcutter
Sent: Thursday, January 11, 2018 8:38 AM
To: Micah Salb ; oldtools@s...
Subject: Re: When Good Squares Go Bad

Micah Salb wrote:
"I am tired of squares that aren?t square.  I don?t understand how craftsmen of
yonder days did good work with squares that weren?t square!
"Are there reliable ways of squaring a square?"

You are probably demanding too much precision in your work. Thirty-five years
ago my binding teacher told me "The human eye is designed to forgive a great
deal around the general theme of squareness." If you look at 18th century books
the covers look just fine, but if you slap a protractor on them you will find
that plenty of those covers have corners that are five or ten degrees off.
Sometimes they are twenty degrees off, and you still don't notice it unless you
are looking for it. So chill out a bit. Take a stiff dose of whatever intoxicant
is still legal and suits your taste (My Great-Uncle Charley was a roofer. He
told me once that to do the work, you had to be drunk enough that the roof
looked flat to you; otherwise it wasn't safe. Fact.) If you can't see the
problem without precision measuring tools, then it doesn't matter.

Fit pieces to each other, not to a separate standard. Yeah, that was Maudsley's
great advance, the standard plane surface and so on. Fine, indeed necessary, for
machining metal. Not needed for wood, which is not a precision material. Get a
piece of wood planed to a crossection true to a hundredth of a degree and what
do you have? A piece of wood that is two degrees off after the next big

Precision is nice, I suppose, but too much precision is a killer. Years ago I
saw a TV show where they were interviewing one of the ex-Nazi scientists who
ended up in Russia. A commissar came up to him at a reception and held up a
little vial. "Look at our Soviet plutonium," he said, "it is 99.9 per cent pure;
so you can see how much better it is than American plutonium. They can only make
90% pure." The ex-Nazi told the commissar----well, he said he told the
commissar, and I'm sure he wanted to: "The American plutonium is better. It
costs a tenth as much, and it blows up just as well." You're not a machinist,
are you? Why would you need a run-out of under .0002 inches in 6 inches?

Its a disease that attacks some people. Karl Holtey made beautiful infill planes
for a long time. I've seen one, handled it even if I remember correctly, didn't
have the chutzpah to ask its owner if I could try it. But Holtey wanted more and
more and more precision, and couldn't get it with the infill shimmying around
all over the place, so he got rid of the wood for his most mature designs. Ended
up with planes a lot uglier than the infills, and no better for working wood. I
like shimmying; I wouldn't want a Holtey-designed exotic dancer.

Back in the '80s someone wrote in a letter to Fine Woodworking: " 'Pretty close'
is not good enough for machines because they don't know how to be anything but
exact. With 'pretty close' machines become confused and their bearings heat up.
But man has always known what to do with 'pretty close.' He hits it with a
hammer, or puts in some shims, or changes the design, or adds some molding.
Man's ability to respond to 'pretty close' is precisely what makes him human and
vastly superior to machines. The ability to adapt to 'pretty close' makes human
work so much more pleasing and sought after than that of the machine...After
all, if God had wanted man to be perfect, he would not have invented wood
filler." [Thomas P. Sullivan to FWW #26, Jan/Feb. 1981, p. 4.]

Oh, you wanted practical advice? With a 10X magnifier and a hard, sharp pencil
the line-and-flip test will show you if a square is precise to .0008 over six
inches, and maybe .0004.----about half of Starrett's guarantee on their working
squares.  If a square won't pass this, get rid of it. Never buy a used square to
use it (c*ll*ct*ng them is another matter). Buy new, with a guarantee of
precision (BSC guaranteed machinists' squares are common and inexpensive) and
test a new square when you get it---send it back if it doesn't pass. Trying to
true up a used square is an extreme sport in itself. I've done it (once) and I
might do it again some day, just for the challenge, but if you want a square to
use, buy a new accurate one.

Or make your own of wood. It is actually surprisingly easy to make one, and easy
to true it before it is assembled, if you follow Chris Schwarz's method as shown
to St. Roy:


I have two or three I made from rosewood offcuts, and I use them all the time,
for, of course, checking or for marking with a pencil. I keep steel squares for
use with a marking knife.

If precision really matters to you get a new Starrett or equivalent (the Swiss
company that bought Brown and Sharpe, or the Japanese one, or there is probably
someone in Germany) and keep it in a fitted case to check your working squares.
I believe the high-end firms will true a square that is out. Expensive,
probably, but if precision really matters to you....

I say I trued a steel square, by the way; well, this is maybe not quite right. I
worked away at one for six or eight years, getting it as good as I could and
having new problems crop up every time I corrected an old one. Finally I got fed
up and buried it in a drawer, still not satisfied with it. Five years later I
took it out to have another go, and I couldn't figure out what was wrong with
it. Kept trying to figure out the last problem for a while, but finally I gave
in and started to use it. So that's my last bit of advice to you: put an
inaccurate square away in a drawer for a few years. Maybe it will correct

Tom Conroy
Disclaimer: This email and any files transmitted with it contain confidential
information and are intended only for the individual named. If you are not the
intended recipient, you are notified that disclosing, copying, distributing, or
taking any action in reliance on the contents of this information is strictly
prohibited. If you have received this e-mail by mistake, please delete this
e-mail from your system and notify the sender immediately by e-mail. Thank you.

Recent Search Bios FAQ