Thread: Engineers Square - reasonable quality

Originally Posted by ian

Barri, in answer to your original question, as commented above, precision and accuracy costs real dollars.

Dial indicator $30. The "better" square I ordered $69. Vernier calipers for the 5 cut method about $30. I think our dollar is the same as yours Ian. These aren't big dollars and these devices have many uses in a workshop. If my garage blew up these would be high on the list of replacements

The comment above that "accuracy cost money" but so does mistakes.

Interesting photo but wouldn't you need a perfectly aligned mitre gauge or sled to make this?

BTW If you said to me that my simple question in the OP would lead to 4 pages I wouldn't have believed you.

Originally Posted by barri

Interesting photo but wouldn't you need a perfectly aligned mitre gauge or sled to make this?

nope

in essence, once you strip away the ornamentation, what you have is three sticks that together create an exact 90 degrees, even if they are just nailed together.
once the "sticks" are joined together, the outside edges are trued to an exact 90 degrees, using the "flip and check" method.

As an aside, a 3-4-5 triangle creates a 90 degree angle to a higher degree of precision than is possible to measure with any micrometer or dial indicator.

Ian, I was a maths teacher in a former life so 3-4-5 is Pythagoras's theorem and you're right if a triangle has 3-4-5 sides then you must have a 90 deg angle. The builder who built my house used this to check the squareness of the concrete floor. I was impressed. This is also similar to checking if a box (I make quite a few) is square where you measure both diagonals. Anyway my square arrived today but I'm too busy to unpack it but I will report back. BTW even with 3-4-5 you still need a good measuring device and avoid parallax to check those sides and these sides need to be long enough eg 30cm-40cm-50cm to get real accuracy compared with 3cm-4cm-5cm. We'll have to agree to disagree then that 3-4-5 isn't as accurate as the 5 cut or dial indicator methods.

Barri

I am not criticising the idea nor saying it doesn't work.

Mr Ng's system works perfectly - every time.

It costs virtually nothing to do.

The method you have is novel, clever, smart and accurate. Go for it. Buy a $400 engineers square, smile and move on.



EDIT - Just saw you got your square - enjoy.

Originally Posted by barri

I've used the 5 cut method many times and it does job well but also relies on vernier calipers (which could be dropped, bitten by an insect or left out in the sun)

The vernier's accuracy is irrelevant as in the end all that you need are two readings that are the same - their actual value makes no difference.

Originally Posted by Bohdan

The vernier's accuracy is irrelevant as in the end all that you need are two readings that are the same - their actual value makes no difference.

Not true. Once you have made the 5th cut. You need to measure the far end and the near end accurately eg vernier calipers and subtract. This answer is divided by 4 and that gives you the error per mm or inch depending on what scales you are using. That error times the length of you fence pivot point to adjustment point and that answer decides how much you need to move your fence, generally using a feeler guage. So those two initial readings need to be accurate. See Ng's video for a better explanation

And you then repeat the test and measurement to make sure that you have done the correction correctly, the two ends are the same size and you are perfectly square. You are totally independent of any external measuring device accuracy be it length or squareness.

I actually left out one step. After dividing by 4 you then divide by the length of the 5th cut. That then is the error per mm.

Also not true. The final step is still measuring the near and far of the 5th cut with vernier calipers again and subtracting finding the error as I mentioned above. When the error is below .005" (some say less) then that's considered good enough. That's the aim. Its almost impossible for the two ends to be exactly the same but close is good. Don't take my word for it but watch the Wood Whisperer's ( https://www.youtube.com/watch?v=uE9f4bp_wm8&t=911s ) and William Ngs ( https://www.youtube.com/watch?v=UbG-n--LFgQ&t=1773s ) video which shows exactly how to do this. They use vernier calipers the whole way including the final check. You have to do that to work out the error

Originally Posted by barri

Ian, I was a maths teacher in a former life so 3-4-5 is Pythagoras's theorem and you're right if a triangle has 3-4-5 sides then you must have a 90 deg angle. The builder who built my house used this to check the squareness of the concrete floor. I was impressed. This is also similar to checking if a box (I make quite a few) is square where you measure both diagonals. Anyway my square arrived today but I'm too busy to unpack it but I will report back. BTW even with 3-4-5 you still need a good measuring device and avoid parallax to check those sides and these sides need to be long enough eg 30cm-40cm-50cm to get real accuracy compared with 3cm-4cm-5cm. We'll have to agree to disagree then that 3-4-5 isn't as accurate as the 5 cut or dial indicator methods.

all you really need for your 3-4-5 dimensions, is a straight edge (or snapped string line) and a pair of dividers.

But this is not about any criticism of your method, merely to point out that with some basic hand tools, and a little skill, you can make a large wooden square that is square enough to set your fence at 90.00 degrees

Well said Ian. I think some people think the 5 cut and dial indicator method and overkill and would prefer to do what you suggest or even use a cheap square. Simple equipment can get you there. Not denying that. I just want people to be aware that there are methods out there that are super accurate if you want that. If one person tries one of these precise methods and likes it then I'm happy I mentioned it. I guess my maths background leads me to the more accurate way.

OK I wasn't game to post this just in case we got another 4 pages but I did promise to report back. I also promise not to mention the ..... cut .... or the dial ..... methods and only talk about what this thread was meant to be about.
IMG_20170205_162958.jpg
Unpacked this today and are very pleased with it. Did I few measurements and found it is pretty well spot on. It is chunky, solid and well made. The edges as I wanted were dead straight and smooth. It arrived in a nice padded box with a coating of an oily substance to protect it. I might leave it in the box for protection and bring it out when I need a little more accuracy.
In terms of value, well for me it wasn't much more than the "better" squares found in hardware stores. Very happy with it.

Originally Posted by barri

OK I wasn't game to post this just in case we got another 4 pages ...

I don't have a problem with this thread running to 4 pages (or more), for me a thread like this just enhances the sense of community

The best thing I bought for setting up tools etc is one of these: Shinwa gauge squares : Tools from Japan, Japanese woodworking tools direct from Japan.

Utterly, utterly brilliant and extremely accurate.

Originally Posted by ian

As an aside, a 3-4-5 triangle creates a 90 degree angle to a higher degree of precision than is possible to measure with any micrometer or dial indicator.

An under-measurement of 1mm in 300mm x 400mm x 500mm

-> 300 x 400 x 499 would give you a "right" angle = 89.8 deg

300 x 400 x 501 => 90.2 deg

It's still prone to possible error.

Cheers,
Paul