Decent try square

[Cheshirechappie]

Too many moving parts to take that seriously. If the surface of the table is fractionally bowed, for example........

True - except that in a toolroom, all measuring equipment and reference surfaces are checked, and their inaccuracy is KNOWN. That's why metrology equipment is usually supplied against one of three grades - Workshop, Inspection and Calibration. There is a higher grade, which is comparison against the national reference standard, but that hardly needs to concern us! Obviously, there tends to be a difference in the time needed to make a measuring instrument that meets each higher grade of precision, with a consequent increase in cost - hence the price Andy Kev noted for a certified M&W square - someone had to check each square against calibration equipment and detrmine that the inaccuracy was within that specified for the grade of square, and fill in the certificate that goes with the individual square.

All very interesting in passing - but it's not really woodworking, is it!

 

[Cheshirechappie]

Another point with woodworking squares is that they do tend to be vulnerable to knocks. I once bought a new 12" Rabone Chesterman try-square by mail order, and it arrived packed in only a padded envelope. It did cross my mind that it could very easily have been knocked about in the postal system, but it proved to be OK, thankfully!

Always worth checking second-hand try squares, of course. First check the stock for any dings and burrs, which could throw the register of the reference face out a bit, and correct as necessary (fine file). Do the same for the blade, and check that it's straight, too - marking knives can take slivers off blade edges. Then check that both stock and blade are each the same width throughout, and correct if necessary. Finally, check for square on both inside and outside. Making all these checks should ensure that a square can be used with confidence on both the inside and outside faces, and on 'opposite' faces (standing on it's base and referencing against the inside of the blade, for example.

Goes without saying that it's worth checking new squares, too.

 

[Woodmonkey]

It sure is. I bought a crown square in my local tool shop a few months back, only about a tenner. But luckily I remembered to bring an engineer's square with me and had to go through 6 or 7 before i found a decent one - they were in a pile in a bucket!

 

[Farmer Giles]

The bill for the roof on the extension came in a lot less than estimated so I've treated myself to a forged starrett combination square with combined metric/imperial 300mm rule.

Amazon still haven't picked up the 450mm Bahco combination square and i was sick of tripping over the parcel in the office and Amazon's return system seems to be useless so for the sake of 12 quid I've cancelled the return and will get the files out and see if i can get it somewhere near square.

I'll use the starrett for furniture and other work requiring tighter tolerances and if the bahco can be fettled i'll use it for general joinery and building work.

 

[DennisCA]

I trust my triangle single piece squares the most, aluminium and plastic versions, one piece design is the trick IMO. It's not just good enough to be square when you get home, but it has to remain square through all the trials and tribulations in the workshop.

 

[MarkDennehy]

Yeah, but that's a lot of material between the reference faces Dennis - which means that when the temperature changes, those reference faces are being pulled or pushed by the expansion or contraction of the material. Which is why engineering squares didn't all switch over to being triangles a few hundred years ago.

Also, you drop one of them, they get knocked about just as readily as an L-shaped square; it's just that you'd have to regrind the faces to get back to square instead of centerpunching the joint or adjusting it some other way.

They're great for construction and framing where the additional strength is more important (not to mention the ability of them to lay flat on a 12x2 while the base hooks onto the edge or other such quick layout setups) than fine accuracy, but you're talking about an environment where you're squaring rough-sawn lumber and dragging a wide-nibbed pencil across the square to get your layout line. Being out by 0.1mm over the 30cm length of the square is immaterial when the surface you're measuring varies by 20-30 times more than that and the pencil line is ten times that or more.

 

[woodbrains]

The bill for the roof on the extension came in a lot less than estimated so I've treated myself to a forged starrett combination square with combined metric/imperial 300mm rule.

Amazon still haven't picked up the 450mm Bahco combination square and i was sick of tripping over the parcel in the office and Amazon's return system seems to be useless so for the sake of 12 quid I've cancelled the return and will get the files out and see if i can get it somewhere near square.

I'll use the starrett for furniture and other work requiring tighter tolerances and if the bahco can be fettled i'll use it for general joinery and building work.

Hello,

Starrett is an excellent choice of square, I think they are peerless.

However, I would have got a rule graduated only in the units I use. Dual reading rules are a PITA because the measurements don't scan across the the face of the rule and this is really annoying in use. Also Starrett dual rule isn't 12 inches long either, so you can't read 'downwards' so the imperial side is less useful. By the same token, if it was 12 inches, then the metric side would be compromised. If, like me, you use both metric and imperial (SAE) then it is better to have 2 rules, which I have in my Starrett combo square set.

Mike.

 

[Phil Pascoe]

https://www.youtube.com/watch?v=T7JqXbm-Nwo
is an interesting hour, although half their stuff is Chinese now.

 

[D_W]

https://www.youtube.com/watch?v=T7JqXbm-Nwo
is an interesting hour, although half their stuff is Chinese now.

Yes, and I'm sure that fraction will continue to increase.

One has to be careful to ensure that they're getting the american stuff if paying big coin.

 

[bugbear]

One has to be careful to ensure that they're getting the american stuff if paying big coin.

Pretty sure that if iPhone's can be made in China, they'll manage to make metrology to spec too.

BugBear

 

[D_W]

One has to be careful to ensure that they're getting the american stuff if paying big coin.

Pretty sure that if iPhone's can be made in China, they'll manage to make metrology to spec too.

BugBear

I'm not sure Iphones and certified squares have a lot in common.

but it's not like fine things can't be made in china. Just a matter of Starrett probably not making their fine things there. We have a step forward in perception to make before that occurs - that being that even if Starrett could make a certified square there for half or a third of what they can make it in the US, they're not going to submarine their high end stuff by doing that.

...yet.

As far as fine things go, I'm under the impression that the copy shops in China can pretty much duplicate a Rolex ceramic bezel daytona for about $700 - including the movement. I can't imagine that it's harder to make nice squares - but as above, somewhat unrelated.

 

[DennisCA]

Yeah, but that's a lot of material between the reference faces Dennis - which means that when the temperature changes, those reference faces are being pulled or pushed by the expansion or contraction of the material. Which is why engineering squares didn't all switch over to being triangles a few hundred years ago.

Also, you drop one of them, they get knocked about just as readily as an L-shaped square; it's just that you'd have to regrind the faces to get back to square instead of centerpunching the joint or adjusting it some other way.

They're great for construction and framing where the additional strength is more important (not to mention the ability of them to lay flat on a 12x2 while the base hooks onto the edge or other such quick layout setups) than fine accuracy, but you're talking about an environment where you're squaring rough-sawn lumber and dragging a wide-nibbed pencil across the square to get your layout line. Being out by 0.1mm over the 30cm length of the square is immaterial when the surface you're measuring varies by 20-30 times more than that and the pencil line is ten times that or more.

I would like to see how much thermal expansion and contraction affects accuracy. And as for dropping, I have dropped mine several times, I would have to throw it in the floor with all my strength to injure it.

Even so there are one piece machinist squares, I have those too. One piece design is what I value.

 

[bugbear]

I would like to see how much thermal expansion and contraction affects accuracy.

For linear accuracy (e.g. a rule) steel has a linear coefficient of thermal expansion of around 12 micrometres per metre per K (it varies with the exact alloying materials and heat treatment).

This means that for a metre rule calibrated at 20°C, it will expand by 120 micrometres or 0.12 of a millimetre 30°C (or a yard rule would expand by 4 thou).

But it can be much worse than this for a straight edge or square, where the linear expansion of PART of the item can cause a much larger bend in the whole item (consider the behaviour of a bi-metallic strip).

BugBear

 

[MarkDennehy]

Also remember that because it's a solid triangle, there's more mass expanding behind the reference face as you approach the corner, so thermal expansion won't just push the edge out of true, it curves it.

Again, for the work it's intended for, this really doesn't matter, it's going to be negligible compared to the other inaccuracies the work imposes.

 

[JWD]

If the OP is still searching, have a look at the M&W engineers squares on workshop heaven - i've bought the 4" one and haven't had the opportunity to use it yet but they're very well reviewed by the likes of Waters and Acland furniture school so cant be that bad.

 

[MarkDennehy]

I love how the day after this thread starts up, I find my favourite 4" engineers square is no longer square
And with the kit I have there's no way to reset it.


*sigh*

I guess I'll have to buy myself a M&W replacement for the solstice so

 

[D_W]

Do you have a small mill file to draw file the blade to square and then to parallel?

 

[custard]

Here's the thing about squares. When a square is certified "square" what is it that's actually being certified?

Here's a diagram that illustrates what I mean,

Is it edge A to B that's square within a specified tolerance? Or C to D? Or A to D? Or B to C? Or indeed all of them?

I forget the specifics of the DIN and BS grading, but it's certainly not all of them, in fact I'm pretty sure it's just one of them. In other words there's no obligation for a certified square to have either the blade or the stock parrallel. Taking this further, plenty of steel rulers don't have parallel sides either, even though the scales might be acceptably accurate. Which is why the earlier suggestion of lining up a ruler against a square held on the counter edge in a shop, and then flipping the square to the other side, won't actually ensure a successful purchase.

And this is only part of it. Imagine you've got two squares on a known flat surface, a bit like this,

You'd expect, with a quality square at least, that you'd get a gap free alignment where the two blades meet. But what about if you take one of those squares and rotate it through 90 degrees (so the edge of one blade is touching the face of the other blade). Are you still guaranteed a gap free alignment? No you aren't, certification doesn't preclude having a blade that's bent like a banana.

For all these reasons, and plenty more besides, that's why most serious makers I've met just take a deep breath, dig deep, and buy Starrett, Mitutoyo, or M&W combi squares. Because with brands like that you won't be caught out by the small print!

 

[MarkDennehy]

Do you have a small mill file to draw file the blade to square and then to parallel?

Yes. Shush! You're ruining the plan!

 

[MarkDennehy]

most serious makers I've met just take a deep breath, dig deep, and buy Starrett, Mitutoyo, or M&W combi squares

Whoa, stall the digger.
Combi squares? Not just ordinary square squares? I mean, I see the utility of them for having a depth stop/half-inch square for checking mortices or for making fast 45 degree lines, but would you really depend on a combi for square over an engineers square?