Sharpies are NOT accurate marking devices

[natenaaron]

I see it all the time.  Trace the shape with a sharpie.  Mark the length with the sharpie.  Then complain because your cut is inaacurate. 

Yes I'm venting.  I watched a grown up measure and mark something with a sharpie.  He measured twice because that is what he is supposed to do.  He cut and what do you know he was off.  In this case, too long.  Nothing a grinder couldn't fix.  He did it again and was too short.  Obviously his saw is messed up.  It couldn't be the fact that he was marking with a chisel tip marker.  He had to make multiple cuts the same length and every one was different in the end.  Wait, set up a stop at the length you need?  Nah  Takes too much time

Maybe it is my woodworking back ground but when I mark something it is with as thin a line as possible.  A scribe or sharpened soap stone.  Heck a nail or flat head screw drive in a pinch, and how the heck hard is it drop a brick or piece of steel as a stop. 

Grrrr.  Rant over.

[Glenn]

In defense of the wide chisel tip markers or sharpie markers, you have your choice of 3 lengths, near, middle, and far side of the mark. Pencils on wood are a bit more accurate as the pencil mark is a bit thinner, so do you cut the mark, or leave the mark? If the mark is a bit long, it is nothing a hammer to compress the end grain of the wood will not fix. (grin)

Soap stone wears down from the chisel point and leave a mark 1/32 inch wide on one end of the line (long line) and 1/8 inch wide on the other end of the same line,

A metal worker will use a awl or scribe and scratch a 1/64th inch wide line on a piece of metal and then proceed to hand cut the metal with a ox/ac torch. Was not pretty at all and the only thing you could say about accuracy was that he did cut the metal in two. No guide used, just hand held and weaving across the cut like a Saturday night drunk on the highway.

Do not get me started on measurements as rulers will lie to you. The are NOT accurate !!  That is unless you use one ruler to measure and then use the same ruler to transfer the distance to the material being cut. Metal tape measures have a sliding end, push for an inside measurement and pull for a outside measurement. The difference is suppose to be the thickness of the hook. Ever check that measurement to see if it is accurate?

Ever look at the still steel metal (or wooden) ruler and find that the numbers on the end of the ruler START about 1/8 of an inch or more from the actual end of the ruler. Again use the same ruler to measure and then to transfer the cut distance and there is no problem, just DO NOT have one fellow call out a distance and the other fellow with a different ruler measure to make the cut.

Long as your venting, when was the last time you tested your square to see if it WAS actually square?

One fellow working alone with one ruler is not a problem, it is two people or two rulers that makes a mess of things. Oh, by the way can you cut this 2x4 a half an inch longer for me please?

 

[Smoggy]

I watched a TV prog about the rebuilding of a Spitfire. Despite the fine tolerances on the machining of parts, the the wing spar pin did not fit! Apparently, the presenter was doing his piece to camera with pin in hand, it had warmed to body temp and a flush fit turned into an interference fit. If body heat can do that in short order to a goodly fist full of metal......what will it do to the steel rule you've been holding half the morning? One of the first things I learned about measurements as an aprentice was accuracy is temperature dependant......I later came to learn it's also relative. 6" to a gardener is about the length of their dibber, to a welder it's long enough to span the gap, to a foundryman, plus shrinkage, to an engineer it is pedanticly 6,000" at ambient temperature. to a motor cycle racer it may be life or death, So if you want a near enough measurement then marking it of with a sharpie is fine, if you want an accuracy......use a cooled sharpie

[Culver Creek Hunt Club]

For steel steel most used in surveyors tapes.   correction = desired lengthx.0000065(actual temp - 67F)

[Frosty]

This is starting to sound like the old adage, "close enough for government work. Measured with a mic, marked with a piece of chalk, cut with an axe.

If you use a steel rule, tape, divider, caliper, mic, etc. on steel at the SAME temp the COE will be the same within a few hundred thou. There are or were correction charts for dissimilar metals.

Frosty The Lucky.

[ThomasPowers]

Historically a smith in the early 1800s might work to a tolerance of "The thickness of a worn shilling" and parts would then be fitted. (based on the tolerance being reported in a letter complaining about the cylinders being produced for early steam engines)

I'm not a machinist but I know accuracy costs money; being excessively accurate can destroy the profitability of an item.

[Kozzy]

At one point, someone in our main shop had the bright idea of buying several tape measures that he found at bargain prices--that way everyone would have their own for quick measurements and verifications (not layout but double-checking).

In 6 feet they varied by about 1/4".  Took a few frustrating boingles before the problem was narrowed down.  Now we verify tapes against a standard before they hit the shop floor.

As to sharpies, Frosty got it right in that you have to use the thing right to provide a reasonably accurate cut line.  There are also sharpies with very fine points which give you a line no wider than a pencil mark.  In some cases where marks are really hard to see, I use a liquid-paper pen:  There is a style that is essentially a fine-point ball point pen for white ink which doesn't last long but sure shows up like the dickins.

Anyone needing more than about 1/32" accuracy should be scribing anyway.  You can probably get to 1/64" with an ink or pencil line but you reach the "might as well do it right anyway" point around that tolerance.

A scribing knife in versions for both wood and steel are excellent first knife projects:  Small enough to be cheap and not painful if things go awry but useful enough to be worth having and using.  I know most metal scribes are single-point tools but I prefer the knife end style that looks like the wood version with a bit different edge grind.

[Culver Creek Hunt Club]

3 hours ago, Frosty said:

If you use a steel rule, tape, divider, caliper, mic, etc. on steel at the SAME temp the COE will be the same within a few hundred thou.

100%. we never had to correct when measuring long steel items like pipeline or large structural when everything was the same temp. 

[Anachronist58]

Here is how I cope:

Silver Steak over wide Sharpie background.

Oh, and Slag, something from My trade:

"Tribal Knowledge Trumps ISO".

Robert Taylor

[Smoggy]

....and that Robet is the Engineers method, marking out blue applied to scribe into, easy to see and likely as acurate as you can mark fo machine to. Just don't forget to adjust for the backlash!!!!!!

[Charles R. Stevens]

Yep, sharpie and scribe if I need it real close, chalk/sharpie if I need it smithin' close. 

[Frank Turley]

I've made lots of scribers. Scribes sit in a chair and make figures on a ledger. A scratch awl is a scriber, probably if you're in a Southern state. For steel, I usually use car hood springs. Harden in oil; temper to a dark straw. The point is finished to 20 degrees included angle. Rattail on the other end and a twist in the middle. Blacksmiths twist; machinists knurl.

Scribers are great giveaways, especially for "Iron in the Hat."

This happened in a harpsichord factory. The old hand hollered to the new kid, "What's that measurement again?" The reply was, "Sixteen inches and uh, and uh, uh, two bumps past the big bump in the middle!"  

 

[Charles R. Stevens]

So my English isn't so good, I was educated in the late seventies early eighties...

[rockstar.esq]

I think there's a misconception lurking in this.  Precision and accuracy are not the same thing.  Precision is a measure of repeatability whereas accuracy is how close something is to a given standard.

If a group of arrows are tightly clustered, the archer is very precise.  If those arrows aren't centered on the bullseye, the archer isn't very accurate.  It's just as possible to be accurate without precision as it is to be precise without accuracy.

Many people assume that greater granularity in precision is always an aid to accuracy.  That is often not true because we interpret what we're doing by comparison.  A line that's visibly off is easier to spot and correct than a line that's so fine that perspective is lost.

Imagine looking at a round target and trying to center fine crosshairs on it.  While it's true that fine crosshairs aren't obscuring the target, they're also giving you a huge area to find the center of.

Now imagine looking at a round target and trying to center a front sight that's a ball.  The ball is sized to appear just the tiniest bit smaller than the largest diameter of your target.  It's pretty easy to see when the "big ball" isn't round because the sight isn't aligned.

That's how people make 1,000 yard hits with iron sights.

Getting back to the sharpie example.  If you use a thin straight edge like masking tape, you can intentionally ink both the work and the straight edge.  However crude or thick the resulting line is, one edge will be perfectly accurate.  That gives you a huge area of contrast for cutting, while providing the necessary accuracy.

 

[Anachronist58]

rockstar.esq, that's an ear full. I must consult my ancient dictionary.

Last night I was (virtually) under a microscope scraping titanium from the cylindrical lands of a carbide reamer.

The Sharpie was deployed to "proof" the "precision" of the material removal.

We call it "cutting paint", when we have to remove <.0001" of material at a time.

I request more time to consider your analogy, as my buffer is currently maxed out.

Robert

[JHCC]

On 9/8/2016 at 4:49 PM, Frank Turley said:

This happened in a harpsichord factory. The old hand hollered to the new kid, "What's that measurement again?" The reply was, "Sixteen inches and uh, and uh, uh, two bumps past the big bump in the middle!"  

 

When I was working in the violin shop, everything was done in metric, but you'd go with "a scant millimeter", "a millimeter", or "a fat millimeter", depending on whether you were looking at the near side, middle, or far side of the ruler mark.

[Frank Turley]

"Close enough for government work" initially meant the work was good, accurate. Later on, it came to mean the work was not so good.

[rockstar.esq]

Anachronist, On one hand, I can only marvel at anyone holding tolerances that small in machine operations.  On the other, I spent some time in the semi-conductor industry where they were maintaining manufacturing specifications down to the angstrom level. That's a tenth of a nanometer.

It's finer than frog hair when measurements have to factor the probability of whether the thing you're measuring is really there!

 

[Anachronist58]

rockstar.esq, hold that thought - I'm going out the door to my night job.

Robert