What's new
What's new

Checking micrometer accuracy and recommended brands

  • Thread starter Admin5
  • Start date
  • Replies 5
  • Views 37,633
A

Admin5

Guest
Bob Ranck wrote:

Don, You, Fitch and the others raise a lot of very valid points. But now I must weigh in on this one.

My credentials: I am tool & Gage Supervisor in a LARGE machine shop.

I supervise Calibration and maintain over 5300 pieces of measuring and test equipment, just in this one plant. We have nearly 450 comopany-owned OD micrometers alone in this machine shop - Brown&Sharpes, Lufkins, Scherr-Tumico, Mitutoyo, Starrett, and Craftsman primarily, and I won't even begin to hazard a guess at the value (the largest ones go to 70 inches, $1400.00 per copy) but we also have a lot of the individual operators who have brought their own personal mics into the system, too. We calibrate EVERY ONE of these on a monthly schedule.

It is my exerience that , being first of all in good repair, being cleaned and recalibrated monthly, and having been of good manufacture to begin with, there is little PRACTICAL difference in the equipment. Added to this, there are many Chinese and Japanese "knock-offs" in our system that, with the same care, will produce equal result. When the ganefs were stealing all my new 0-1" Brown&Sharpe and Starretts, I bought a dozen of the $12.95 0-1's from somebody, Westport, MRO, Penn Tool, I'm not sure. They are a Chinese copy of the Mit's with the BIG scales on the sides, read to tenths, also. And they ARE just as reasonable and just as accurate and repeatable and reliable as the aforementioned high-dollar herd.

But the single differentiating factor is (you guessed it) THE USER! We have some real craftsmen, genuine *machinists*, who will make accurate measurements with any and all mics. Then, there are the "second-tier" individuals who somehow, always manage to just get the part within spec - barely- unless the mic has STARRETT on it. I don't know, just some sort of ego problem, a confidence factor, who knows? Personally, I find an esthetic appeal to each different one.

Anyway, that's why GR&R is divided into GAGE REPEATABILITY (the equipment factor) and GAGE REPRODUCIBILITY ( the operator factor).

We actually calibrate over 1,000 mics, (ID, OD, & depth) every month, not to mention about 150 lbs. of calipers, 80 lbs of small snap-gages, etc, ad infitum, nauseamque.

Carla wrote-

there's a simple, practical test you can do, that will tell you a bit about the accuracy of a micrometer, under field conditions......get a group of workers together, hand the first one a gage block, on which you've put a bit of masking tape over the size marking.......

(provide clean rags to handle the block and mike to minimise error from the heat of the people's hands).....

ask him/her to read the dimension of the block with the mike, then write down the figure without showing the figure to anyone else.....

get each person in turn to do the same......then compare the figures they'd written.....

you'll find the result interesting.....it points out a "fact of life" in machine and inspection work......accuracy of meaurement, in reference to known standards (the gage block, for example) will depend on operator skill, as well as the quality level of the instrument used to take the measurement....

that said......buying anything but the best quality of basic measuring tooling, like mikes, for example, is simply foolish....

there's no free lunch.......depending on a cheap mike could cost you a job worth many times the value of the difference between the cheap tool and a best quality one......or maybe you'll get lucky......just as with any other gamble.....

if you choose to depend on "getting lucky".....well, thats your choice......

i've always liked the feel of the "starrett no.226" mikes......the lighter "tumico" mikes are excellent in the larger sizes...it takes awhile to get the "feel" of those, but once you've become accustomed to the lighter frame, they're nice to use......carboloy tips are well worth the little more they cost......using the "tenths" vernier is a matter of personal preference....you can eyeball the tenths of divisions of a standard mike with a bit of practise....taking the tenths vernier as "gospel" infers a 50 millionths accuracy.....and you can't trust an outside mike to this level......there are people who will say their feel is that good......take that one "cum grano salis"..........

the swiss "tesa" mikes are excellent also...

i've tried to use the nip "mitutoyo" mikes.....they are nearly alright for open tolerance work, but i just don't like the feel of them.......since you can get used ones of these for almost nothing sometimes, it could be worthwhile to have a set of nip mikes for open tolerance rough scaly work, to preserve your good ones.

chinese????......polish?????.....why bother?????

what's the value of a workpiece relative to the cost savings in buying cheap tools????
cheers
carla

Forrest Addy replies-

Whenever I had close tolerence work to do, I ususally calibrated my feel with the particular mike to a known reference usually a stack of Weber gage blocks.

Temperature control is very important when precision measurements are involved. Sunlight will kill your best efforts. I've verbally blistered apprentices and toolroom attendents for careless handling of large mikes when I needed to use them immediately. It's natural to carry a large mike by cradling the inside of the frame in the hand. Also naturally, this injects a large heat input to the inside of the frame expanding the head away from the anvil meaning it will take an hour or more for the mike to reach thermal equalibrium. if you need to carry around a large mike hook a finger in the frame next to the spindle. Large mikes are those over 4" and especially those over 12".

The coefficient of thermal expansion is 6.5 millionths of an inch per inch per degree F. A mike warmed to 80 degrees by body temperature will measure a 6.0000" dia part at 68 degrees as 6.00047". 0.0005" is all the tolerence you have on a 150 mm ball bearing fit.

When I have very close tolerence work, I protect the mikes and the work from heat transfer by using shop towels as insulators.

All mechanical measurements world wide are referenced to standard laboratory conditions of 68 degree F (20 degrees C). If the measuring instruments are steel and the work is steel, it's ok if both are the same temperature. If one is bronze or aluminum ... get out the thermometer and calculator to determine the correction.

The Asain import mikes I've seen were workable but pretty poor compared to Starret and B&S mikes in my inventory.

As for the Polish mikes, I bought one from Enco for a tryout. I sent it off to a buddy in a cal lab. He ran it through the wringer and it came out smelling like - clean laundery. The gaging faces were square with the axis and parallel to each other, the spindle thread was within millionths of a tru pitch cylinder, the frames wer stiff and strong, the sleeve and barrel adjustments were proof against accidental change.

I thought so well of them, I bought a set of 1 - 12" and three sets of 0 - 4' as gifts.

Generally speaking, a measuring insrument should be 4 times as accurate as the work its intended for. I read a mike to "tenths" all the time but I'm constantly aware of how easy it is to "wish" a mike reading where I want it. Therefore I never wathc the mike when I take a reading. I know people who same they can "estimate" to 20 millionths with a good mike but I call them danm fools or liars.

A tenths one way or the other on a good mike is about as good as you can hope for in a good environment on favorable work. If you want better than that, better use dial comparative gaging equipment and have the setting standard nearby.

Charles J Hinckley P.E. replies-

Forest
I, like you, always have a good laugh when I hear people claim they can measure millionths whith a hand held mike, they just don't have the slightest idea of what they are talking about. I used to work for American Bosch designing diesel fuel injection pumps. We routinely worked with millionths. The clearance between the plunger and barrel on a jerk pump was 90 to 160 millionths if I remember correctly. These were not measured for production assembly. They were honed for the final machining step and were selectively sorted using air gauges and master plugs or ring gauges and then selectively assembled and checked again with an air gauge for a final assembly check. All done in an air-conditioned room. If we did want to do an actual measurement to millionths the part had to be put into a temperature controlled clean room for at least 36 hours to temperature stabilize and then be measured on a fancy coordinate measuring machine that was mounted on a 30 ' thick concrete pad (that's not a typo, 30 feet!) So anyone who claims they can measure millionths with a mike , well...... Since I left Bosch I have been designing pumps for industrial and naval use at Warren Pumps and our tolerences are a little less stringent!
 
I would like to thank all those members who have helped us, Eliya Software ,with new ideas for our newer features. We would also like to thank those who participated in the beta testing of our new product, READI ERP.
 
Carla said:
i've tried to use the nip "mitutoyo" mikes.....they are nearly alright for open tolerance work, but i just don't like the feel of them.......since you can get used ones of these for almost nothing sometimes, it could be worthwhile to have a set of nip mikes


Seriously? This shit made it into a "Best of" thread? (Stumbled upon this trying to learn something new. Today must be "Hate an Asian Day" or something.)
 
The coefficient of thermal expansion is 6.5 millionths of an inch per inch per degree F. A mike warmed to 80 degrees by body temperature will measure a 6.0000" dia part at 68 degrees as 6.00047". 0.0005" is all the tolerence you have on a 150 mm ball bearing fit.

Since thermal expansion of the frame increases the gap between anvil and spindle, wouldn't a 6.0000" part measure at 5.99953" since the spindle has to move .00047" further to make up for the expansion?
 
i usually check micrometers and other gages against gage blocks and note gage block error sizes for example if 4.0000 is -.0002 then it is actually 3.9998". if i got anything more than 0.0002" error i would double check and if confirmed send gage to be calibrated. i have seen gages not calibrated correctly although it is rare for calibration mistakes to occur.
.
i have no problem measuring to 0.0003" repeatability and usually get 0.0001" repeatability. of course as a apprentice i learned that milled surface are rarely perfectly flat. we check large parts on a cnc mill with a 0.0001" dial indicator. lap or the difference between the center and edges of a end mill or face mill pass we keep under 0.0003" and when inserts dull it is normal and expected for lap differences to go over 0.0003 to past 0.0005"
...... i would laugh myself silly if somebody even thought for a minute they could measure a part to within 0.0001" as the lap difference alone you could easily measure 0.0003" differences depending on where you put the micrometer.
........ when measuring round objects both od and id the larger sizes are more difficult to measure. we use indicating bore gages with a method to help hold square and or in line with axis. the larger sizes you can easily get a 0.0002" variation in readings. i am not convinced that expensive gages could easily and automatically get better repeatability. measuring something you wiped with a rag and then wipe lint from rag with your hand while coolant is dripping will always be more difficult than a guy in a climate controlled clean room sitting at a table. like who in their right mind would be surprised they might get different readings of 0.0002" or more ?
...... we check large cnc mill parts for straightness and flatness and perpendicularity. our cnc mill not even moving and the axises can be moving back and for easily up to 0.0005". i have seen waves milled into flat surfaces of 0.0003" as the servos were adjusting every few seconds. these waves can and are measured with a 0.0001" dial indicator after cutting. we often have to stop machine and put in handle mode and move a axis back and forth til it calms down and stabilizes within a .0001". this is seen on the servo screen which shows servo movement. i often take another pass of 0.001" as due to hardness variation in metal and or part deflection occurred and the 0.0005 flatness tolerance was exceeded
....... what i am saying is when somebody seems surprised that different machinist might measure 0.0002" or more differences my answer would be of course they could. there are many factors involved and it is normal and expected to get 0.0002" difference in measurements for the reasons i have just mentioned.
 
Reading that made me think that all the micrometers Carla has known and used don't have a ratchet thimble. Every time user influence affects measurement pressure results will vary. Micrometer manufacturers use the thimble ratchet to calibrate - touchy, feely just doesn't give uniform results.

If hand temperature affects the micrometer measurement result you're either holding it wrong or waaaaaaay tooooooooooo long ;)

Carla wrote-

there's a simple, practical test you can do, that will tell you a bit about the accuracy of a micrometer, under field conditions......get a group of workers together, hand the first one a gage block, on which you've put a bit of masking tape over the size marking.......

(provide clean rags to handle the block and mike to minimise error from the heat of the people's hands).....

ask him/her to read the dimension of the block with the mike, then write down the figure without showing the figure to anyone else.....

get each person in turn to do the same......then compare the figures they'd written.....

you'll find the result interesting.....it points out a "fact of life" in machine and inspection work......accuracy of meaurement, in reference to known standards (the gage block, for example) will depend on operator skill, as well as the quality level of the instrument used to take the measurement....

that said......buying anything but the best quality of basic measuring tooling, like mikes, for example, is simply foolish....

there's no free lunch.......depending on a cheap mike could cost you a job worth many times the value of the difference between the cheap tool and a best quality one......or maybe you'll get lucky......just as with any other gamble.....

if you choose to depend on "getting lucky".....well, thats your choice......

i've always liked the feel of the "starrett no.226" mikes......the lighter "tumico" mikes are excellent in the larger sizes...it takes awhile to get the "feel" of those, but once you've become accustomed to the lighter frame, they're nice to use......carboloy tips are well worth the little more they cost......using the "tenths" vernier is a matter of personal preference....you can eyeball the tenths of divisions of a standard mike with a bit of practise....taking the tenths vernier as "gospel" infers a 50 millionths accuracy.....and you can't trust an outside mike to this level......there are people who will say their feel is that good......take that one "cum grano salis"..........

the swiss "tesa" mikes are excellent also...

i've tried to use the nip "mitutoyo" mikes.....they are nearly alright for open tolerance work, but i just don't like the feel of them.......since you can get used ones of these for almost nothing sometimes, it could be worthwhile to have a set of nip mikes for open tolerance rough scaly work, to preserve your good ones.

chinese????......polish?????.....why bother?????

what's the value of a workpiece relative to the cost savings in buying cheap tools????
cheers
carla
 








 
Back
Top