After a while away from woodworking I finally have a good shed, am buying my tools back and starting some projects.
The square issue is not new for me but I just realized that only 10% is my incompetence while the other 90% comes from the fact that my squares suck. I have four metal squares: two casted in a single piece – a big carpentry one and one of those small triangles; an “L” two piece one and a combination square. All of them seems not be reliable enough for a perfect square. I notice it clearly when I cut a piece of wood using any of the squares and check the square after the cut. It seems as they were all off something like a half or a full degree, IDK, since all of them looks bad I don’t have a reference to be completely sure.
My question is how do you guys deal with this and if there is a way of doing square check without using an actual square?
Or the quest for the perfect square is just useless?
Replies
Place your square on the edge of something straight and draw or score a line, then flip the square & draw/ score another line close to the first one. If the lines are not parallel your square is out. This makes it easier to see because it doubles the error.
Also:
https://www.finewoodworking.com/project-guides/hand-tools/truing-combination-square
There are probably a zillion youtubes for truing up a square.
You can also put 2 squares back to back on a flat surface. Once you have a known true square, use it to check everything else.
The inside edges of the square should be used for marking your square line.
I use a Woodpecker Square as standard. All of the economical squares that were checked have respectable inside edges. The top edges are not square to the side guide.
The top edge would be the edge you are using as a saw guide.
Pythagorean theorem works for basic square check. Might just introduce more error depending what you use to measure and what you are making
There is a better mathematical way to create a right angle than the Pythagorean Theorem. You need a straight edge (on a board) or a straight line (on a piece of paper) and a divider/compass. From a point on the edge or line, use your compass to mark points the same distance from the original point. Then set your compass to a larger distance, and from the two outside points, mark arcs that intersect. Draw a line from the original point to the spot where the two new arcs intersect. That line will be perpendicular to the original line/edge. If you use sharp dividers, the accuracy will be determined by the width of your scratched line/arc. If you use a compass with a pencil in it, it will be cruder as the pencil line will be wider.
If you are trying to use one of your squares to guide a saw, you need to clamp the square in place; hand holding it is not adequate. Then check the accuracy of your saw. Determine if the square aligns with the cut; if not, your technique is to blame.
I use PEC double squares, Starrett Combo squares, Veritas precision squares, and others. I have spent the time in the past to assure that they all agree, aligning or discarding those that do not.
There are certain tools in our arsenal that either work correctly or are completely useless (think power jointer or shoulder plane) . The silly old adage of using a single tape measure for the whole project overcoming the fact that it is irregular over its length is an example of the pursuit of fantasy. If the tape is off by 1/64" at 18" and 1/16" at 84", and dead-on at 78", there is no way parts cut from that tape's measurements all magically align ;-))
Quality rules are not terribly expensive. Some name brands are. I would pick some, assure that they agree, and bury the rest of the measuring devices in your shop in a deep dark hole. The frustration of trusting something that is lying to you cannot be over-rated. The waste in material, time and electricity will buy you a quality set of squares and rules.
If you did want an accurate square, then a pack of engineer's squares is a good investment.
These are very cheap, made to (from a woodworking perspective) ridiculous accuracy and come in a range of sizes - I find the smallest very useful.
Otherwise there are some excellent answers above for checking a square.
Often it is not the square itself, but the techniques applied that are the issue - blades not totally square to tables for instance, mitre gauges not square to blades, and not always measuring from the same reference face are very common causes of non-squareness.
+1 on engineering or machinist squares. I suggest you buy a good 12” Starrett combo square.
Inaccurate squares aren’t easy to fix. If I can’t I toss them.
Framing squares can be horribly inaccurate. The best one I have is - you guessed it - a Starrett.
Good idea to make you’re own large square with 2 and 3 foot arms.
And if you find your carpenter's square out of square strike the square on the outside corner against a hard surface to reduce the angle or use a center punch at the inside corner to increase the angle.