How much faith in a dial Caliper?

[rockfish]

I do a lot of work that calls for very shallow counterbores with tight tolerances. Many of these counterbores are large enough to make using pin gages impractical and shallow enough to make bore gages impossible to use. In all the years I have worked in this trade, I have never worked for a place that had anything that would accurately check a feature like this, other than using dial calipers. Because of this, I have become very skilled at accurately checking inside diameters with a dial caliper. I use 12" Mitutoyo dial calipers exclusively and use ring masters to verify size and feel. People may argue about the accuracy of dial calipers all day long, but if I hand you a part that I've checked with my caliper and tell you what that size is, after setting my caliper with a ring gage, I would be willing to bet a lot that I'm within .0005 of that size, every time. Oddly, I have a 12" Mitutoyo Digimatic caliper that I cannot get a feel for and do not trust. I have 2 pairs of 12" Mitutoyo dials and I trust both of them.

I also find dial calipers handy when checking some plastic details, especially those with thin walls as I have a better, low pressure feel with them than I can obtain with a micrometer.



Frank


Home

[bryan_machine]

1. If you search around on PM, you can likely find the thread where a fellow points out that if you give a group of people a set of pins of unknown to them size, and ask them to measure with calipers (be it dial, vernier, or digital) they'll be hard pressed to come within 0.003".

2. You talk about 0.012" but your example was more like 0.0015" - that factore of 10 rather matters... But even at 0.012" a 0.003" device uses up 25% of the tolerance - gulp....

3. You might need to have a chat with the Young One that goes like this "Regardless of what the print says, we get paid for THIS. If we (company) do not get paid, YOU (young one) will be out of a job." And THIS of course is "to tolerance" where tolerance gets consumed by gage error, temp issues, calibration issues, and customers who can always send the next order somewhere else if the part is not what they want, even if it DOES match the print. (Which means the shop doesn't get paid and he gets to work at McDonald's again.)

4. Calipers for measuring shallow holes are probably better because you are (I presume) using the rod coming of the back, rather than either set of jaws. The rod has much better alignment.

[RBD3]

When I started machining there were only Vernier calipers. I was warned that they were better called "very near" and should only be used for rough reference or in situations where where mics couldn't be used. When I see young ones [and sadly, some older who should know better] checking O.D.'s with digi-cal's, it shows how little they know about machining and proper inspection procedure.
There does seem to be a reluctance to strive for excellence and perfection among those newer to the trade. "good enough" wasn't in the vocabulary in yesteryear...

[trevj]

Well, my immediate boss and I had a chat, followed with a chat between the caliper guy and my boss.
A lot of things got straightened out. It was a one way chat.

Reading through the comments here, I feel vindicated in my position in so far as the use of calipers goes.

To clarify the piece being measured. Aircraft landing gear parts. The dimension check sheet says that the minimum size between two faces is (numbers made up) 2.0875", max is 2.0995". That is where the 12 thou comes from. Not a tolerance, but a hard set of min and max dimensional limits. Said individual informed me that the first measurement he had taken, came in 2 thou over max, and that the part was out of limits, that he did not need to do any more work to it. I measured it with his dial calipers, and got a reading 3 thou less than he did, told him to use something that was accurate and reliable, and thus started the whole fiasco.

Cheers
Trev

[bryan_machine]

Huh? If it's 0.003 over max, surely it needs to be made at least 0.003 shorter? Regardless of what warped ruler you measure with?

[trevj]

Originally Posted by bryan_machine

Huh? If it's 0.003 over max, surely it needs to be made at least 0.003 shorter? Regardless of what warped ruler you measure with?

Uhhhh....Noooo.... So "Huh?" right back atcha. What needs to be made 3 thou shorter? The calipers?

His measurement put the part 2 thou outside it's allowable distance (measuring a separation between two faces, essentially an inspection to see if the part has been bent or warped) My measurement was 3 thou less than his, using the same measuring tool, and resulted in a dimension that was inside the specified dimension range per the Dimensional Inspection sheet, rather than one that would have been outside. That brought up the issues, and the argument, in which, he was adamant that the caliper WAS the right tool for the job.

Clear as mud? Measurement was an inside face dimension check. Over max is = to "broken part" Under minimum is as well.

Cheers
Trev

[RJT]

I'm surprized no one has commented on the quality of the caliper. A top quality (B&S, Etalon, Helios, older Starrett) vs a good quality (Mititoyo, new Starrett) vs a POS Chinese or Eastern Europeon will give widely different readings. Put the operator dependant feel on top of that and your juggling variables. A good vernier or dial caliper used correctly +/- .002 all day long. +/- .001 if you have a master to compare it to. My $.02 RJT

[machine1medic]

Accuracy is a matter of "practice" & "comprehention".

This fellow's refusal to accept the need to uses differing instruments,
is one example of bad practice.

Another group that pigion-holes all dial callipers into a
definition of "inadequecy", has poor comprehension.
Being aware of the repeatablillity of every single instrument
a machinist owns is an occupational responcibllity.

Calling all callipers less accute, is lazy thinking.
+/- .005 , gimme a brake.
m1m

[DMF_TomB]

Caliper accuracy opinion. My 30 years experience is I can easily measure to within 0.001" (if less than 6"), that is repeat 10 out of 10 times

1) I wipe jaws clean AND the part to measure each use.
2) I check zero each use.
3) I keep gib reasonably tight as I not like a loose jaw.
4) I buy a new caliper every year of professional heavy use.
5) I have verified and have never had a problem with a Harbor Freight caliper.
6) I rock caliper to get minimum OD reading.
7) I use a moderately high pressure, but obviously is I am measuring something soft of thin I ease up on pressure. Pressure should not be much higher than what a micrometer uses.
8) I make sure part is at room temperature. Very easy for hot part to cool off undersize.
9) If caliper is dropped once it is No longer good for precise work. Get over it and buy a new one.

Main Jaws are the most accurate. Inside jaws, depth rod and depth on left side can be off 0.002". Our calibration guys usually use a small abrasive stone is depth measurement reading is off.

Anybody who complains zero might be off.... Ah I probably check that hundredths of times each day. My preference is a digital caliper.

If I have something expensive of course I would also use a micrometer.

[The real Leigh]

I would add one item to DMF's excellent list (disregarding #5).

I always measure by pressing directly on the jaws rather than using the thumb adjustment. This pushes the jaws against the work as accurately as possible and removes all stress from the measuring system.

Rocking to assure proper seating is also essential, as he mentioned.

- Leigh

[hickstick_10]

I dont see what the big deal is, 0.012 is well within the accuracy resolution of a caliper, be it Size +- 0.006 or Limit tolerances, if hes near eather end of the tolerance scale any self respecting workman would check it with a micrometer anyways, just to be sure.

And besides, he shouldnt be aiming for eather limits of the tolerances anyways.

If he hasn't scrapped anything then his measuring may not be an issue.

[trevj]

It is an inspection of a part that is being checked while overhaul is in progress, and if it falls outside the two dimensions on the sheet, it is not repairable. The dimensions are in tenths of a thou. The last one or two tenths either way, are the most important tenths, eh!

That, in a nutshell, is what the issue was.

I like the step by step there DMF. I gotta say that accuracy and repeatability are certainly achievable, given care and technique, but in the case of the gent about whom I started this query, I am thinking that there was less of either than there should have been.

His measuring has become an issue, as he is supposed to be picking up some responsibility towards the end that he should be reasonably able to be trusted to work unsupervised, and be able to make those decisions in a manner that leaves no one doubting his work. That's been undermined a fair lot of late. Not so much from the one incident, but rather the argument and insistence that he was correct both in the choice of tools, and the measurement.


Cheers
Trev

[machine1medic]

faith & trust ??

Odd words to have in a mind-set that manages one's accountabillity,
for his / her performance and out-come.

It's been decades, I'll bet since I ever applied any measuring tool
to any part or machine-component, with-out involuntarilly
verify-ing the condition of said measuring tool.

I've even found myself abruptly stop an intense conversation,
because a depth-mike felt strange when I picked it up.
Nob was loose, maybe a 32nd of a turn.
OR
Noticed a tiny wiggle in a dial or needle, that I wasn't even really watching for it.
When you don't have any extra or practice parts..........
you.........."trust and faith" ??

m1m

[ducatidon2]

Leigh

thanx for the tip on using the jaws vice thumbwheel.
D_