Top quality Engineer's Square

[lurker]

I think too many of you are trying to crawl up yourselves a bit here. :lol:

The adge that you only get what you pay for does not always work.

ie a £100 square is not 10 times as accurate as a £10 one and even if it was then marking out is likely to negate that. And its just as likely to get damaged as a cheapo one.

I'm fortunate enough to have access to measuring equipment that fills a building and as BB says has to be temperature controlled. I bought a 10" cheap engineers square and got "the lads" to check it out for me. They say its as accurate as any other in the company's machine shops (most are M&W costing eyewatering sums and kept in there own boxes). I use this as a reference for my other squares and don't bash it about.

The big issue IMHO is the square is square enough, but importantly is stays that way (and you check it frequently).

My problem is (being a rubbish woodworker) is I can mark out accurately (thanks to my engineering apprenticeship) but can't cut to the line

 

[bugbear]

I have a Starrett combo square. Since Starrett is a (or was) a top name here in the states, I always assumed that it was SQUARE. Hah! Wrongo!

Assumign that tools from any manufacturer, including Starrett, Mitutoyo, Moore & Wright etc are "absolutely accurate" is a mistake.

All these manufacturers makeq a wide range of tools, at different prices, and for different requirements.

AFAIK they ALL make toolroom grade reference squares. These are guarenteed accurate to quite remarkable tolerances.

They can only be used to this degree of accuracy under highly controlled circumstances. They are so accurate that thermal expansion of the square or workpiece caused by a 1 degree change in temperature is greater than their tolerance, and (clearly) what would be a tiny "ding" on a roofing square is a disaster on a square of this tolerance, so handling and storage must be done with care.

Fortunately, less accurate squares are also made, requiring less care in use, and at a lower price.

It is (I hope) obvious that a combination square will never be as accurate as a fixed "engineer's square" since the bearing surfaces in a combo square are quite small, and subject to frictional wear.

In short (as I've said before) there's no such thing as "dead accurate"; just "close (r)tolerances"

BugBear

 

[bugbear]

Yes,
Tolerances are more useful than descriptive words. Good is a bit vague......

I have some remarkable straight edges from Dick fine tools where tolerance is quoted;

such as 500mm long tolerance 0.004 mm.

Now 0.1 mm is almost exactly 0.004 inches (4 thou")

So I think this means + or - 0.00016 inches or plus or minus less than a couple of ten thousandths of an inch!
At 21.60 euros this has to be the bargain of the century.

I'm sure someone wise will correct my maths if I have got this wrong?

That would be a remarkable bargain; I suspect the tolerance is probably 0.004mm per <some>; perhaps you could check, since I can't find the information on their web site.

BugBear

 

[engineer one]

this has certainly grown into an interesting and informative thread.

what is certain is that a "truly accurate" square cannot be used outside a clean room which is temperature controlled, and actually touching it may well throw it out. :?

so we have to accept a certain level of inaccuracy to be able to use the item in day to day life. obviously 89 or 91 degrees are no good, but where in the range of 90+- 0.3 degrees are we prepared to accept things?

as for bugbears comment about the bearing surface on a combination square in part i would suggest he is wrong. on most engineers squares i have (3, 6,9 and 12inch at least) the actual area of the stock which encloses the blade is quite small, being approximately 70% of the interface. also the way many people use their squares is going to cause major problems because the number of people who hold the square by the blade is amazing

certainly on my combination squares the bearing surface is quite a lot longer. obviously wear takes its toll, but you do need to check all things measuring on a regular basis.

my final thought is we all spend a considerable amount of money
and time using tapes which have moveable ends and we all accept that accuracy. :roll:

paul :wink:

 

[David C]

BB,

I also cannot find the + or - tolerance per ?? except on a European standards website where you have to pay to read the information ;-(

Perhaps an e mail to Doktor Rudolf will supply the answer ...

best wishes,
David

 

[bugbear]

certainly on my combination squares the bearing surface is quite a lot longer. obviously wear takes its toll, but you do need to check all things measuring on a regular basis.

In all (not that many...) combination square I've seen, the blade runs on small bearing surfaces at each end of the slot it sits in, not even the full length of the slot. There is no attachement to the "face" of the rule.

BugBear

 

[Racers]

Hi, Bugbear

Yes combination squares rules only ride on a small area at each end of the slot that is so you can adjust them by filing, I have done mine both the 90 and 45 deg angles are quite good now (temperature depending)and it didn't take long.


Pete

 

[Paul Kierstead]

Luckily for me, my shop is temperature controlled so I'll never have to worry about any project not being square.

 

[bugbear]

Hi, Bugbear

Yes combination squares rules only ride on a small area at each end of the slot that is so you can adjust them by filing, I have done mine both the 90 and 45 deg angles are quite good now (temperature depending)and it didn't take long.


Pete

Heh. I've done that. But only after I'd taken out the 1/32" bow that had been worn into the edges of the rule, and made the edges parallel (*) (**) !!

I guess it was a fairly well used combination square

BugBear

(*) actually filed one edge with reference to my surface plate
(**) actually filed the second edge in reference to the first edge, checking every 1/2" with a micrometer across the width of the rule

 

[Harbo]

The tolerances of Dick's tools are stated in their catalogue (hard copy).
The precision knife edge rules are Grade 00 to DIN 874 - the 500mm Tolerance is 0.004mm Cost 54.50 Euros plus VAT.
The 300mm one is tolerance 0.003mm and costs 26.50 Euros.

They also do knife edge squares to similar tolerances. Because of their thickness and bevel edge not much good for marking out but good for checking edges and other squares.

Rod

 

[David C]

Harbo,

I did quote that tolerance about ten posts back....

However as it is stated in the catalogue the meaning escapes me.

Are we talking +or- 0.004 mm overall or +or- 0.004mm per 100 mm? for the 500 mm precision straight edge.

David

 

[woodbloke]

Harbo,

I did quote that tolerance about ten posts back....

However as it is stated in the catalogue the meaning escapes me.

Are we talking +or- 0.004 mm overall or +or- 0.004mm per 100 mm? for the 500 mm precision straight edge.

David

.....we're talking about woodwork, and after a bottle of red vino, it's off to bed.....zzzzzzzzzzzzzzzzzzzzzzzzzzzzzz - Rob

 

[PaulO]

Live steam scale models are fascinating, and I seem to remember that they could not be made to work well because in a half scale model.......

David C

You need to scale according to Reynold's number.

 

[Harbo]

Sorry David C - I have been away for a few days and have been only quickly scanning the threads.

As the tolerances are quoted differently for each of the three rules - increasing with the length - I would read it 0.004mm per 500mm.

It also quotes that a square 0.5degree out of true, gives an error of 8.7mm per m.

Rod

 

[PaulO]

which is not too bad.

Well it is, actually - it's either accurate or it's not

Accuracy depends on the scale that you are using. An engineers surface plate prepared using the three surfaces and scraping/frosting is generally considered "flat", but the surface finish is actually quite poor. So over (very) small distances, it can't be considered "flat". In the same way an engineers square with a (small) bow across any of the dimensions may be square for certain dimensions, but not for others.

Sorry if I have missed similar comments on other pages of the thread, but I have short attention span, and can't read more than 1 page at a time :roll:

 

[David C]

Harbo,

No worries. That is what I assume though it is not stated absolutely explicitly.

Anyway it is the best most affordable testing st edge that I have ever owned and I am very pleased with it.

The small version is fantastic for testing chisel and blade backs to see how much work they will take to flatten or minutely hollow!

Also very pleased indeed with the lifetimes supply of plastic engineering shim stock that a kind reader from USA sent me.

David C

 

[newt]

Just a thought. If you have what you think is a straight edge (SE) could you check it as follows. Flat piece of MDF longer than the S E, drive a small panel pin at each end, place SE against both pins, mark the centre of the SE then draw a line against the SE for the full length. Now place the straight on the other side of the pins, line up the centre mark and again draw a line. If the SE is true the 2 lines should be parallel spaced apart by the thickness of the pins. I checked my 1 meter rule and it was concave on one side and not surprisingly convex on the other, I thought it was straight. The really sad thing is I was thinking about this when I woke up at 3 in the morning. I accept this is very very crude.

 

[Paul Kierstead]

I accept this is very very crude.

Crude? I think not. It is quite elegant and -- AFAIK -- is more or less (sans pins usually) the accepted way to check a straight edge that doesn't have to be to laboratory standards.

 

[David C]

Newt,

This is a good method, but not quite sure why you would assume a ruler to be straight? Generally they are not.

Now in the past workshops had wooden straight edges which they made and checked.

Three edges which agree with each other must be straight.

Sounds like a good planing exercise to me.

David Charlesworth

 

[Karl]

Hi Whiteant

Not sure if you've purchased your Incra yet, but I have just ordered a set of squares from Aminster. http://www.axminster.co.uk/product-Axminster-Three-Piece-Precision-Square-Set-649978.htm

In terms of squareness they seem to be more than upto the task, and you get 3 for pretty much half the price of the Incra. The longest isn't quite as long as the Incra though.

Mine should be arriving in the next day or two - I will let you know the outcome if you'd like.

Cheers

Karl