How much faith in a dial Caliper?

[trevj]

Sooooo....Near end of day today, got into an argument with one of the young fellas at work.

He carries his dial Caliper about with him and trusts it like it's the holy grail itself.

I figure they are great tools for checking stock size and sheet metal thicknesses.

Short of it was, when I told him that if he expected my signature on his work, he had better start measuring things with a micrometer. We routinely are asked to measure distances defined in half thou dimensions. He believes that a dial Caliper is as accurate as needed, because, in the case this afternoon, there is a 12 thou difference between the min and max dimensions, since that is way larger than the one thou he expects of his dial caliper, that I am being unreasonable. Eg: (numbers pulled outta my...) min allowable dimension, 2.1320, max 2.1335. Gotta be between, not over or under. I figure that last half thou may be the difference between a serviceable part, and one we don't have anymore. I want to be sure. I'm stuck training this one, sometimes supervising. Can't fire him.Not allowed to hit him with sticks.

<sigh> If he spent half the effort he spends arguing, on improving the work he does, he'd be getting somewhere.

Me, I'll happily trust a dial Caliper to a plus or minus 5 thou measurement. He thinks I should take it on faith that he can measure accurately to a thou with one. This after I used his caliper and got a 3 thou difference from the measurement he had told me he got (He said the dimension was over by two thou, I measured it as one thou under, and asked him to measure it with an accurate tool). And so the argument began... That and a endless circle around "you were asked to make a part this size" vs "the page had plus or minus .005 on all dimensions written on it" back to "yes, but you were instructed to make it that size on the work order, exactly, or as close as.." back to "but the drawing..." Christ I hate arguing over stuff I have to sign for.

What's a dial caliper reasonably expected to be able to do? I'll be checking with our calibration folks to see what they inspect them to meet. Mitutoyo's docs claim one thou for a dial caliper, one tenth for a micrometer.

Interested in what the expectations are from the tool. Sadly can't take you up on any suggestions for other than patience, with regards to the other one.

Cheers
Trev

[Finegrain]

I trust that my calipers are off by a couple thousandths .

Seriously, I have 2 4"ers that I use all the time. One of them is within 1 or 2 thou on OD/ID/depth, the other one is good on OD, but 3-4 thou off on ID and depth. Good enough for ballpark, then out comes something better.

Regards.

Finegrain

[hesstool]

There's no problem with him measuring a .012" tolerance. But this guy is clearly an idiot if he trusts them for anything tighter than +/-.005".

There's also the scenario I experienced 2 weeks ago when the dial skips a tooth and you don't realize it. You measure with the calipers and think you're .005" over. Then pull out the mic to work it in, and suddenly realize you've just scrapped a $900 piece of Inconel 625 because it's undersized by .02".

[Richard Rogers]

Dial calipers? I'm constantly resetting them to zero. I've had to trust them a few times for something that nothing else could really measure, but no, I would not say they are the best all-around instrument. Sometimes I measure with them, then mic them to verify what I'm getting. Plus or minus .002 is typical. I will say I've done more accurate work with them than that,but it makes one very nervous trusting them with something that hours of work are invested in.

Richard

[trevj]

Originally Posted by hesstool

There's no problem with him measuring a .012" tolerance. But this guy is clearly an idiot if he trusts them for anything tighter than +/-.005".

There's also the scenario I experienced 2 weeks ago when the dial skips a tooth and you don't realize it. You measure with the calipers and think you're .005" over. Then pull out the mic to work it in, and suddenly realize you've just scrapped a $900 piece of Inconel 625 because it's undersized by .02".

The root of the problem is that it is not a .012" tolerance, but a min and max size, expressed in half thou figures, that if the part is outside of, either way, it is unfit for use and requires a rebuild or it may be scrap.

The Inconel! indeed! Owwwwwwww.....

I read one gent's description of his method, and it involved carefully cross checking and re cross checking his dimensional readings against a known entity (in his case, gage pins) continuously while working. Seemed a reasonable means of grubbing out a better solution to the problem.

Cheers
Trev

[hesstool]

So would saying a .012" tolerance zone be more appropriate? If so, still plenty of real estate for calipers.

[67Cuda]

If it's a .012 spread, dial calipers it is, unless it's questionable as to being over or under the limit, then use a tool that can measure it more definitively.

I trust my dial Etalons to within .002.

Dials can jump a tooth, but digitals can be set to zero anywhere. I'll take a quality dial over a digital any day.


Tom

[Holescreek]

We routinely are asked to measure distances defined in half thou dimensions.

Quote him the 10:1 accuracy rule.

[The Energy Rebel]

In skilled hands (emphasis) one can usually hold .001 - .002 with calipers.
But experience teaches you to always doubt and reach for a doublecheck.
He doesn't have that experience and you're having a hard time teaching it to him.
When running tenths, we routinely have ring gauges, plug gauges, mics, but never rely on a set of calipers.
Even if you get it to pass thru the gauge because it warmed up in your hot little hand, the customer's source inspector can always find the time to send a box of rejects back, gift wrapped with his or her name on it.
That's probably all it will take for this greenhorn to learn the reason why.......hopefully.

[tailstock]

Sounds to me like this guy is not very confident reading a set of mics. and uses any excuse not to. Time he had a brain retooling or back to machine shop 101

[Greg Quenneville]

There are some good basic metrology books around that illustrate the problem with calipers-the Abbe error to name one. Dials, as already mentioned, add another level of mechanical uncertainty. I have two dial calipers, bought when I was unsure of micrometer reading and too ignorant of the basic principles. They are used for fabrication work now, and checking rough stock. Better than a tape measure, but only by one notch.

The question I have is this: Why the hell is this guy arguing with his instructor?
In my world I expect rapid adoption of lessons I teach, and do likewise when it's my turn to learn something. Perhaps the basic teacher student relationship needs to be better defined.

[The real Leigh]

If you need to measure a feature to a tolerance, set the caliper to the nominal dimension using a gage block stack. Repeatability should be within .001" or so, depending on the quality and condition of the caliper.

You have to decide whether that's sufficiently accurate for the particular task or not.

I use them without checking calibration for things like stock diameter or thickness, as an earlier poster mentioned. If I need to measure something more closely, I check the caliper at that dimension using gage blocks or pin gages or similar standards.

- Leigh

[Cyclotronguy]

I've been doing this so long I cannot remember a time when I did not know the smell of cutting oil.

Dial calipers.... stock check, screw identification, scribing stock, keeping drawings from floating away..... etc.

I don't even trust my mic's unless they've been certified and even then I like to check them against a shop grade standard every now and then..... just to make sure.

Cyclotronguy

[Conrad Hoffman]

I do lathe work to +/- .001" all the time using dial calipers (Starrett 120s), but I'd never consider calling something "good" until it was checked by mics that were checked by blocks or pins or mic standards. I make small parts. With anything over an inch and a half, I'd be far less trusting of the calipers, more because of user error than their inherent accuracy. With expensive materials or a piece with a lot of work going into it, mics would always be the rule.

There was in interesting phrase in the old movie Ronin. "If there's any doubt, then there is no doubt." It concerned the potential for an ambush, but it applies to a lot of things. If there's any doubt, then there's no doubt you have to check it. Any doubt about one tool and you check it with another. Then you check them with standards. If you still have doubts, you check by null comparison to a wrung together combination of blocks at the same dimension as the part.

CH

[anteekfreek]

Originally Posted by The Energy Rebel

In skilled hands (emphasis) one can usually hold .001 - .002 with calipers.

Absolutely.

It is VERY easy to measure something incorrectly with them.

Every type of measurement instrument has its place in a shop. You go from simple steel rules, to calipers, to micrometers, and right on up depending on the kind of accuracy needed for a job.

The type of feature being measured weighs heavily on which measurement instrument must be used, nearly regardless of the tolerance (to a point).

I use digital calipers every single day, and yet I have a mega-bucks DCC-CMM as part of my arsenal. I use the proper tool for each measurement task. The CMM is not a "measure anything" tool either.

[specfab]

As previously noted, it is possible to get fairly accurate results within 0.001-.002" depending on skill and familiarity. I would have no issue with measuring to a min/max gate of .0015" with my Starrett 120 calipers, well used by me frequently, if it was only required to get it inside that spread, HOWEVER, I would cross-check with mic if I were doing work that required certifying the accuracy. I would not know for certain what the size actually is from the caliper reading, but for a caliper that is checked frequently, used by one person with skill and experience (+ "feel"), getting to that level is possible. That being said, your "student" needs to read a bit about metrology philosophy and take your instruction seriously.

[spmaster]

Was the .012" tolerance zone you quoted a typo? Your trainee wouldn't last five minutes with me. In my apprenticeship I learned it my mentor's way first. Gradually I was allowed to spread my wings. Before that time I did it his way. There were no arguments. When it comes time for you to inspect his first piece and sign off on it make sure that the inspection equipment being used is suitable to you. If it is not, don't sign off. If he refuses to adapt and learn your methods and practices, it is time for a meeting of the minds between you, your boss and the trainee. Any tolerance zone smaller that .01" in my shop is checked with an instrument with at least .0005" resolution. Period. Greater than that a caliper is acceptable as long as the feature is running at nominal. Anything straying towards top or bottom you go to a finer measuring instrument for verification before offsetting. You may not be able to fire this clown, but you can certainly make every minute of his day uncomfortable until he decides to get with the program or find another place of employment that is more suited to coddling him.

[Old Bill]

That is why they call them "Very Nears". They get very near but not close enough. Mine seem to err most when checking an inside dem. I have learned to swear at them, not by them.Like their owner, they are old and worn.
Old Bill

[Laurentian]

My best operator always gets a smaller reading than I using 6" Mit. digi calipers. He always seems to be pressing hard, and this has been going on for over a decade, and he's aware of it We use the guage blocs and mics on parts all at room temperature et where we want to be. In our shop there is an understanding that its easier to remove stock than to add it, so with the little batches we run typical to the precision engraving and toolroom work we do. So we run everything at the upper quoted tolerance. I rather be grinding off a thou or two here durinmg asembly or inspection prior shipment. We have had alot of saves just on that principle alone.
Hugh

[A_Pmech]

If your shop as a metrology department send him over there for a talk with the QC guys. They'll set him straight in short order and probably throw his calipers in the trash.