# How Square is Your Square: A dial indicator method

Page 2 of 5
• posted on January 4, 2013, 4:07 am

------------------------------------------------------------------ You can forget a log fireplace in SoCal.
The South Coast Air Quality District outlawed open log fire burning fire places years ago as an air pollution source problem.
Lew
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 2:32 am
snipped-for-privacy@garagewoodworks.com wrote in

of checking before and I do now. The Draw-A-Line and flip method will never tell you how far out of wack your square is
Pfui. Of course it will. The more the lines diverge, the more out of whack it is.

can measure the distance betwene the lines/knife marks with a caliper (good luck).
That'll work. Feeler gauges will probably work better.

answer that question so why bother checking in the first place?
Of course you can easily answer that question. Decide what percentage error is acceptable in your project, and you can easily calculate the permissible gap between the marks. Suppose you want accuracy to one part in 1000 (99.9%). The gap between the pencil lines at a distance of 8" -- double the error in the square, remember -- must be no more than 8 * 2/1000 = 0.016", which is easily measurable with feeler gauges, a dial caliper, or a mike.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 2:52 am
On Thursday, January 3, 2013 9:32:58 PM UTC-5, Doug Miller wrote:

"More out of wack" is not a quantity. I can only be used in relative terms. 'This' is more out of wack than 'this'.

Huh? Measure the distance between two lines with a feeler gauges?

So you decide before hand the accuracy you want to achieve and then choose the appropriate square? Do you have different squares for different jobs? Are they labeled as such?

A distance of 0.016" equates to an angle error of 0.5 degrees for your 8" square. Would you calibrate anything with that "Square"? I wouldn't.
What is your limit of detection for measuring gaps between lines? I hope it's less than 0.016", but I can understand if it's not.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 2:53 am
snipped-for-privacy@garagewoodworks.com wrote:

You mean between two points--or two parallel lines. I agree that feeler guages seem suitable.
Bill
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:28 am
snipped-for-privacy@garagewoodworks.com wrote in

Yes, and I was using it in relative terms. Where's the problem?

Sure, why not? You can easily see if the gauge overlaps the lines or fits between them. BTW, the width of the pencil lines isn't really relevant -- use a marking knife instead of a pencil.

Of course.

I have a Starrett square that never leaves the wood shop. I also have a Craftsman combination square, and a Stanley framing square, that I use for carpentry projects. It never crossed my mind to use the Starrett when I built a shed in the back yard a few years ago, or when my son and I built a deck last summer -- that degree of precision simply isn't needed, and there's no point in risking an expensive precision tool on a construction site.

No it does not.
First of all, 0.016" is double the error, so the actual error is 0.008" over 8" or 0.001. The inverse tangent of 0.001 is 0.0573 degrees (rounded to 3 significant digits). If you want to be really picky, the 8" distance is actually the hypotenuse of the triangle, so we should use inverse sine instead, but the result is the same to at least the seventh decimal place.

I wouldn't either -- *if* your number was right. It's not, though, and I'm not sure why you seem to have a problem with using a square that's accurate to one part in a thousand for woodworking.

I can estimate much finer gaps than that by eye. The thinnest feeler gauge I own is 0.001"; anything less than that is too fragile for wood shop use, and is pointless anyway. In practice, a 0.002" gauge is quite sufficient for woodworking.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:43 am
On Thursday, January 3, 2013 10:28:59 PM UTC-5, Doug Miller wrote:

That's not how the math is done.
If your gap between the two lines (drawn from the same point) is 0.016" then half the distance is 0.008" which is your "Opposite Side". Now I understand the confusion here.

And I once caught a fish ------This Big------.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:53 am
snipped-for-privacy@garagewoodworks.com wrote in

LMAO
distance is 0.008" which is your "Opposite Side". Now I understand the confusion here.

And tangent = opposite / adjacent = 0.008 / 8 = 0.001.
(As I noted previously, the 8" is actually the hypotenuse, not the adjacent side, but for angles this small, it doesn't matter.)
Anyway... arctan (0.001) IS NOT 0.5 degrees. You're off by an order of magnitude. [...]

If you *can't* tell the difference between a gap of sixteen thou and a gap of six thou without using precision equipment, then you need to visit an optometrist. Seriously.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:59 am
On Thursday, January 3, 2013 10:53:50 PM UTC-5, Doug Miller wrote:

See my other post. I misplaced a decimal. Happens sometimes.

I can discern a difference. But i wouldn't be able to quantitate the two gaps to the level of accuracy required. And neither can you.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 4:11 am
snipped-for-privacy@garagewoodworks.com wrote in

accuracy required. And neither can you.
Don't tell me what I can and can't do.
You don't own feeler gauges, so you don't know how small a gap is visible, or how accurately it can be estimated with a little practice.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 10:19 am
wrote in message

And I once caught a fish ------This Big------. I=========================================================================I don't doubt him a bit. I can and I'm not the only one.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 1:41 pm
On 1/4/2013 5:19 AM, CW wrote:

No one mentioned using the top of their old Craftsman table saw to check the accuracy of their squares. I frequently check my square by laying the square on the table and checking the square using the edge and the miter slot. or a corner of the table. However the slot is more accessible since the saw has wings on the iron top to extend the work surface.
For setup of saw and router table I use an aluminum triangle square. If I drop it so what, while the edge maybe nicked it will not be knocked out of square.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:45 am
On Thursday, January 3, 2013 10:28:59 PM UTC-5, Doug Miller wrote:

Crap, you're right. It is 0.057. I misread a decimal place. Apologies.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:56 am
snipped-for-privacy@garagewoodworks.com wrote in

Accepted. Thank you.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:06 am
On 1/3/2013 6:30 PM, snipped-for-privacy@garagewoodworks.com wrote:

small gaps between the two pencil lines (especially with a thick pencil lead). The most you might be able to detect is a 0.010" difference which equates to a minimum detection of 0.036 degrees with an 8" square.

perfectly flat, otherwise you will not get an accurate calculation of your square's angle error.

Then use a marking knife.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:16 am
On Thursday, January 3, 2013 10:06:49 PM UTC-5, tiredofspam wrote:

You're still left with not being able to quantitate the error. Are you following along?
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:45 am
snipped-for-privacy@garagewoodworks.com wrote in

_Of course_ you can "quantitate" [sic] the error. What makes you think it can't be measured?
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:47 am
On Thursday, January 3, 2013 10:45:21 PM UTC-5, Doug Miller wrote:

Measure thousands of an inch between two lines. Sure it can be measured. Accurately? No.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 3:55 am
snipped-for-privacy@garagewoodworks.com wrote in

What makes you think that? Do you own a set of feeler gauges?
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 4:02 am
On Thursday, January 3, 2013 10:55:23 PM UTC-5, Doug Miller wrote:

I have never needed feeler gauges. They work on a very basic principle and I feel confident making that claim.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on January 4, 2013, 4:08 am
snipped-for-privacy@garagewoodworks.com wrote in

making that claim.

I hope you'll pardon this small correction: you have never *perceived* the need for feeler gauges.
IMHO you should reevaluate that perception.