do you check the squares in your shop?

[David Fegen]

Five years ago I bought a brand new Starrett 4" double-square and it has been my daily go-to tool. Yesterday, I decided to put it up against my Starrett 6" combination-square and I found it is about 1/32" off over the 4" length. I have no idea how long it has been like this.

Looking at the ridges the ruler rides on, they look run down (pic attached). I guess I need to file them and re-calibrate the double-square to be true again (or just buy another one!).

This got me thinking...

Adjustable squares will never be as accurate over the long-term as fixed squares (given quality is matched). So ideally, I should really only be using an adjustable square when needed and otherwise be using a fixed try square (for checking an edge to a face, etc.)

I should be checking all the squares in my shop. Maybe every 6 months or every year, etc. I know the most commonly prescribed way is to take a piece of scrap, straighten an edge, draw/mark a line from the square, then flip it around - any error will be doubled. While it's not that much effort, it still takes time & energy. It seems much easier to just put the test square up to a square known to be true. But of course, a highly accurate square can be expensive.



My questions:

1) do you check the squares in your shop? if so, how often?

2) how many people have a high-precision square they keep in a box that is the arbitrator of truth. If so, what square do you trust for this (Chris Schwarz seems to indicate his Vesper try square is his arbitrator of truth) - Starrett, etc.?

[steven c newman]

Check only if and when I drop one...

[Richard Coers]

I always check them when new, but never against another square. Find a piece of wood with a very straight edge. Lay the square on the wood and mark at both ends of the leg of the square. Flip it over and line up the mark at the base of the square and then compare to the mark at the top. Any error is double with this test.

[Edward Weber]

I'm with the others, check when new or dropped. My usage won't knock them out of square.

[Jim Koepke]

Mine get checked often. Drawing two lines takes very little time.

In Euclid's Door by Walker & Toplin this method is shown to check shop made squares. It also shows tying a longer straight edge to the square's blade to extend the test so any discrepancy is measured over a much longer distance.

The use of my combination or double squares is limited. Many years ago there were about a half dozen combination squares in the shop. After testing them only a few remained. The remaining ones have posts for the blade to ride on. This makes for an ability to adjust them fairly easy. Go slow on the filing. A little bit can make a lot of difference.

jtk

[Edward Weber]

I have a few combination squares but I rarely, if ever, use them as a square. I use them as a marking gauge more than anything else.

A few decades ago, bought a 12" framing square that had so much varnish (clear coat) on it, it was little bit out of square. As we all know, small errors can add up quickly.
Empire did replace it, although I had already ruined enough wood to P___ me off, Ever since I always check a new or unknown tool.

[Derek Cohen]

I like the adjustability of my 4” and 6” Starrett double squares. I know that they are accurate because I check them periodically. Still, when it is a money shot, I only use my Chris Vesper machinist squares. I never check these as I know they will always be accurate, regardless of being dropped or worse.

I modified a carpenters steel square to set up the crosscut fence on the slider. The ultimate test is that the cuts are perfectly square, but in tuning the square I flipped-and-marked-and-tapped-and filed until I was satisfied with its accuracy, and then keep it safe between uses.

Perfect accuracy may not be as important in woodworking as in metalworking, but it does not take much extra effort to have a tool set up to remove all doubts.

Regards from Perth

Derek

[Dave Zellers]

I bought a Woodpeckers 12" square when it was on sale maybe 3 years ago. I specifically wanted to check my 40 years accumulation of various squares including 2 aluminum framing squares from my old carpenter days.

I couldn't find one that I felt was out more than an acceptable amount (I'm crazy fussy) and more importantly, both my aluminum framing squares from 40+ years ago, which would each include a few drops, were spot on.

Having spent 80 bucks on my new 'reference' square, I was both disappointed and happy at the same time.

I've built an awful lot of dead on square cabinets over the years without that 'reference' square, so I'm not sure why I bothered.

[Andrew Hughes]

I have a Starrett no 61 for my reference square. I use it to check and adjust the squares that sit around the shop.
I also use it the set my table saw blade square then it goes back into the cabinet.
I thought everyone has a reference square.

[glenn bradley]

Yes, regularly.

[David Carroll]

I have a fair number of squares, I don't really collect them, but I have accumulated a bunch over the years. I'm surprised that the Starrett would be so far out. I wonder if there is a knick or burr on the guide rail kicking one end up. I would imagine the wear from moving the rule back and fourth would be distributed more or less evenly.

I have an old Lufkin 4-inch machinists square that I keep in it's case and use it only to check, or set other squares, sort of the final word. But I don't do it very often. If I suspect one is out of square I will generally just knife a line across a board (that has a straight edge) and flip the square this way and that.

DC

[Mel Fulks]

Some of those squares have an adjustment that is sometimes used to mark a line not 90 degrees.

[Derek Cohen]

Still my favourite square ...


https://www.youtube.com/watch?v=dB0MBGiX8TQ&ab_channel=ChrisWong

Regards from Perth

Derek

[Edward Weber]

You owe me 2 minutes

[Derek Cohen]

But it was fun :)

Regards from Perth

Derek

[Mark Hennebury]

Checking squares against your "reference" square doesn't tell you if your squares are actually square. It only tells you if they are a complimentary angle to the reference square.

You can check squares using three squares and a flat surface like a jointer table.
Check squares back to back
Mark the squares 1. 2. 3.
Using 1. as the reference, check 2. and 3. against it.
If 2 and 3 fit against 1. it means that 2 and 3 are the same angle. You don't know what angle at this point. but you know it is a complimentary angle to 1. you don't know what angle 1. is either.
Now check 2. and 3. against each other, if they fit then all squares are 90 degrees.

502910

[Eric Brown]

Did you know you can also check your center square also? Use a square board (not a round) and use just one corner. Mark a line and flip. Same as regular square. An interesting note, most center squares will put the edge of the rule exactly on the center. The best way to use this then is with a knife to mark with. Using a pencil the line will be marked to one side. Flipping the square will end up with two parallel lines. The only center square I know of that can be offset was made by Brown & Sharpe. Patented on June 29, 1937 (#2,085,461) it has adjustable screws. If interested I can post pictures.

[Edward Weber]

But it was fun :)

Regards from Perth

Derek

You and I may have a different version of fun

[steven c newman]

Sometimes...one just has to trust a square. My biggest thing is....I use the same square throughout the Project....I do not change squares. To do so, is to take a chance that the next square will be off just a blonde hair from the first square...but, I rather not take that chance..

[Haitham Jaber]

I tweak my Starrett squares with a needle file.
Few Starrett come dead on. I don't know why. I own several of them. I started with a reliable try square that doesn't slide but can be disassembled.

1. I plane the edge of a piece of lumber as a reference
2. Use it as a reference to scribe a line with a thin marking knife and reverse the square to see if it's dead on
3. I estimate the offset of the two lines in case there is
4. I see if the angle is bigger or smaller than 90 degrees
5. I disassemble the square
6. Put the handle in a little jeweller's vice or the like
7. Wear diopter 5 optivisors
8. Pick a little hand shaped needle file (a flat file I mean) with the thickness approx 20thou thinner than the slot in which the ruler slides in
9. The ruler when tightened is pushed against two little shoulders that I will call here risers, 1/3 of the thickness of the slot, interrupted in the middle by the clamping bolt's hole, and rising by 30thou approx over the bottom of the slot
10. Whether I have to adjust the angle of the square in one or the other direction I focus on filing more one riser than the other.
11. The trick is not to file just one riser but to give strokes that slide over the full length of the slot, applying slight more pressure on one finger of your hand than the other so that you will maintain a continuous line throughout the length of the risers
12. 2 or 3 passes 1/2 inches long, very accurate, taking care not to touch the sides of the slot (put some tape if you want) and then recheck scribing two
opposite lines as before
13. continue slowly by trial and error and in half an hour or more it's done
14. being hardened steel you can be very accurate and proceed by unnoticeable amounts
15. use a cut 2 file
16. to make the things faster, before finally checking with the piece of wood, I constantly check the angle with a Helios Preisser square (precision grade 0) to check for square.
These are squares without a base that I use just as a reference, not meant to be on the bench when working hard

Enjoy

[Mike Cutler]

Mine get checked often.
There are videos on how to tune a combination square. should anyone need to tune one.
My table saw blade gets checked with drafting triangles. I know, low tech, but those plastic drafting triangles have accuracies that are a factor of machine baudrate and samples. They're more than accurate enough for checking tablesaw blades perpendicular to the table surface, or a jointer fence.
My shaper and table saw sleds get set with one of Brian Lamb's, large triangles, and 6' Starrett straight edge. Two butt ends laid on a table the way they came off the saw, will be a straight edge when butted together. If not, the sled isn't at 90 degrees to the cutter of blade
At the end of the day though, the wood will tell you if your cuts are square, just make the cut ends work against each other to expose the error.

[Jim Koepke]

I know, low tech, but those plastic drafting triangles have accuracies that are a factor of machine baudrate and samples.

Back in "the day" high school students made accurate drafting triangles.

502951


the wood will tell you if your cuts are square, just make the cut ends work against each other to expose the error.

Checking can be done quickly with a line up.

502950

The third one from the right seems to have a little lean.

Most of the time two or three squares are used on my bench. The are checked for agreement.

Squares to a woodworker are like watches to a train crew, if they don't agree, we don't go until they all agree with the master.

jtk

[David Fegen]

Wow, thanks everyone so much for the comments & input! It's great to get a variety of responses and see how everyone does things a little differently - then I can pick and choose things that I think will work for me. The community is definitely an amazing resource - thanks everybody!

[Malcolm Schweizer]

My 6” Starrett stays in my apron and is used many times a day. I check it regularly. Every Friday is shop cleanup and tool tuneup, and it gets checked with the scribing two lines method.

[Jim Koepke]

Wow, thanks everyone so much for the comments & input! It's great to get a variety of responses and see how everyone does things a little differently - then I can pick and choose things that I think will work for me. The community is definitely an amazing resource - thanks everybody!

One point to consider is most try squares have a brass strip on the inside of the stock. The blade should be square to to the inside of the stock. If the stock is well made and not damaged the unprotected back may also be square, though it is not as easy to check for square.

jtk

[Curt Harms]

Mine get checked often.
There are videos on how to tune a combination square. should anyone need to tune one.
My table saw blade gets checked with drafting triangles. I know, low tech, but those plastic drafting triangles have accuracies that are a factor of machine baudrate and samples. They're more than accurate enough for checking tablesaw blades perpendicular to the table surface, or a jointer fence.
My shaper and table saw sleds get set with one of Brian Lamb's, large triangles, and 6' Starrett straight edge. Two butt ends laid on a table the way they came off the saw, will be a straight edge when butted together. If not, the sled isn't at 90 degrees to the cutter of blade
At the end of the day though, the wood will tell you if your cuts are square, just make the cut ends work against each other to expose the error.

I too use plastic drafting triangles as a standard. They're obviously not suitable for daily use but seem to work well to check daily users. I use a triangle from Staedtler (sp?) who makes better drafting tools I think. Like any other tool I'd be skeptical about no-name Asian devices.