Measuring Tool Errors

[Tom Bender]

Today I checked a few measuring tools for consistency. Mostly things are good but two aluminum rulers 36" and 48" are pretty far off; 1/16" over 32". Knowing that makes it manageable. My basement shop is always between 65 and 70 degrees so temperature is not a factor, just manufacturing quality is.

[Doug Dawson]

Today I checked a few measuring tools for consistency. Mostly things are good but two aluminum rulers 36" and 48" are pretty far off; 1/16" over 32". Knowing that makes it manageable. My basement shop is always between 65 and 70 degrees so temperature is not a factor, just manufacturing quality is.

How do you check if a ruler is straight? Answer: you have TWO of them. If they register off each other in both directions, they're straight.

[Andrew Seemann]

I have one of those aluminum yard sticks that was so far off, I scribbled over the markings with a Sharpie and wrote "Straight Edge Only" on the back. Given that it only cost about $5 and a Starrett 36" rule is a couple hundred bucks, I guess you get what you pay for. I mostly just use those aluminum rules for straight edges and making arcs though, almost never for measuring. The tick marks on them are too big.

Actually, I use 16' Stanley power-lock tape measures for anything over 24", and they seem to work just fine until they get really worn out. I've got about a half dozen I use in the shop. They are all within about 1/64 of each other over 36 inches, plenty accurate for anything made of wood.

[Richard Coers]

If those are big box "yardsticks", I'd only trust them on drywall for which they are made.

[Mark Hennebury]

How do you check if a ruler is straight? Answer: you have TWO of them. If they register off each other in both directions, they're straight.
Could explain what you mean in more detail, please.

[Doug Dawson]

Could explain what you mean in more detail, please.

If you just register the edges, (i.e. can't see light through a gap when you hold them edge-on against a light,) all that proves is that they might be identically distorted, as might happen with two units from the same flawed production run. If you then flip one of them end for end, registering the edges proves that they must also be _symmetrically_ distorted w.r.t. the halfway point down the ends of the units, which is even less likely and you should run out and buy a lottery ticket. Alternatively you could check against the opposite edge of one of the units, if both edges of that unit were supposed to be reference straight, which is not always the case.

Of course the best choice is to have a NIST-traceable straightedge to check everything against. Starrett sells a good one (36") for less than US$300. I do a lot of machine work so that's what I use. Not everyone likes to clear the metrological fog like that.

[glenn bradley]

I spent a reasonable amount of time early on getting measuring tools that matched. It is not all that hard but, you do have to be willing to return an item of two or, carry a "my shop certified" measuring device to the store where you plan to buy another. I have 3 R to L tapes, a couple sets of rules (hooked and not), combo squares, double squares etc. that all match.

The wive's tale of using the same tape for the whole project is flawed if you think about it for a moment. A tape that is off 1/16" at 30" and 1/8" at 76" will not assure a good build. It only assures a variable error factor depending on length of part being measured :D:D:D

[Andrew Seemann]

The wive's tale of using the same tape for the whole project is flawed if you think about it for a moment. A tape that is off 1/16" at 30" and 1/8" at 76" will not assure a good build. It only assures a variable error factor depending on length of part being measured :D:D:D

I've never had a problem using multiple tapes measures on the same project; that is kind of the whole point of standardized measuring units and devices. I have had a lot of problems finding where I just left my tape measure though, which is why I have so many:)