• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.
[4]

Metrology 101

[3]
[10] Like what you see?
Click here to donate to this forum and upgrade your account!

nobog

Active User
Active Member
Joined
Jun 10, 2013
Messages
79
Likes
16
#1
Measurement Basics 101 - or just a few tidbits I've picked up over the years.....

Or Metrology 101

Metrology is the science that deals with measurement (not to be confused with meteorology which you watch after the news). Metrologist: One who measures or deals with the science of measurement.

We all measure things; 2 x 4’s, curtain rods, cylinder bores and such. But what is the best way and how accurate do we need to be?

First a couple of definitions;

Accuracy:

  1. Conformity to fact.
  2. Precision; exactness.
  3. The ability of a measurement to match the actual value of the quantity being measured.

Precision:

  1. The ability of a measurement to be consistently reproduced.
  2. The number of significant digits to which a value has been reliably measured.

You can see from the above that if you are always accurate you are precise but you can be precise but not accurate. How’s that? If you take a measurement 5 times and the known true value is 1.000 but you get numbers like:

.990, .991, .989, .990 & .990 then you are very precise but not very accurate.

Resolution is the ability to discern one number from another. It’s impossible to read to the thousandths of an inch (.001) with a tape measure but easy with a micrometer. The micrometer has better resolution than the tape measure but the tape measure has a greater range.

Implied digits:

In engineering the number of digits to the right of the decimal implies, or indicates the accuracy of the dimension. A typical scenario might be something like this:

1.” = +/- .050”

1.0” = +/- .030”

1.00” = +/- .010”

1.000” = +/- .005”

Nothing is exact, there is always a tolerance. If you say “my cylinder bore measures 2.501” it might be 2.5012 or 2.5008 but it is not 2.501000000000. Of course theoretically it could be but without the proper (and expensive) measurement equipment the point is moot.

Calibration is to compare your equipment to a known standard. “Old Faithful” micrometer may fit your hand like a glove but if it hasn’t been calibrated its reading is suspect. You will need a Gage Block to check your micrometer. Most gage blocks are accurate to a few millionths (.000001) of their nominal dimension. These are available from your local industrial tool distributor. Calibration can apply to your radar gun also – or almost anything that takes a measurement is a candidate.

For checking shafts for out-of-roundness you will need a dial indicator. Many come with a magnetic stand or offer other ways of mounting. Additionally a “V Block set” is necessary to support the ends of what you want to check. None of this does any good with out a good starting surface and they have thought of that too. A Surface Plate is used as a base for most metrology work. Most surface plates are made from granite although some steel surface plates are also used.

Surface finish is generally measured in Ra (Roughness Average) and is usually expressed in microinches. A polished surface may be a 1 to 4 whereas a drilled or milled surface may be a 125. Smoother may not always be better. Surface finish can have a large impact on part performance, it all depends on the application and requirements. SF measurement is usually not a home measurement process, it is only mentioned as something that should be taken into account when specifying part geometry or for specific performance requirements. A good guide can be found in Machinery’s Handbook – the standard reference book.

One degree: One seems like a small number but it can be significant. The naked eye is actually very good at discerning small angles. If a 40’ long house is not square by one degree it is off 8”. If the Golden Gate Bridge was one degree off over its length it would be off by 73 feet! Your 1” long part that’s 1 degree out of square is shorter on one side that the other by .0175”. Moral; get a machinist square – or learn Trigonometry.

So, what should I buy?

The undoubted universal measuring tool is called the Dial Caliper, or these days the Electronic Caliper. In general the higher the cost the more durable with more features. Cheaper calipers, although fairly accurate, usually lack the durability and “feel” of the higher priced units. 6” calipers can cost anywhere from $20 to $200 – get one. I recently bought a 4” caliper for $12 – new, I wouldn’t use this on the Space Shuttle but as a back up it works just fine. Most people prefer the digital variety however the old stand-by analog dial type has been working superbly for decades. Avoid vernier calipers – buy one and you will find out why.

Expect accuracy of +/- .002” although manufacturers will claim better. The resolution on many electronic calipers is .0005 (the last digit is either a 0 or 5) but it is largely ignored – see the accuracy statement above. If you need accuracy of less than .001” then the caliper is the wrong tool anyway. For home use, avoid spending money on calipers with SPC output (used for data collection in industry) or the like. You should not be measuring cylinder bores or pistons with a caliper.

A 0-1” micrometer is also a standard in many tool boxes. Most can accurately measure to the nearest .0001”. For reference, a human hair measures about .003”. The price/quality scenario applies to micrometers much as it does for most things in life.

Of course the above just scratches the surface of the available tools and scenarios, get an ENCO (edit Feb 2017 - ENCO got absorbed into MSC Ind Supply - there is really no shortage of places online to find measurement instruments) catalog or the like and start shopping, the possibilities are endless.

You’ll be measuring like a pro with a few simple rules; have the proper measuring device for the job, keep things clean, avoid parallax errors (measuring jaws not flat or square with a surface), and calibrate. And… critical measurements can be temperature sensitive as materials expand when heated.

Jim
 
Last edited:

chips&more

Active User
Active Member
Joined
Mar 19, 2014
Messages
2,062
Likes
1,391
#2
Excellent reading for the home shop machinist…good job. But, if you think you have finally stopped the world from misusing/transposing the words “precision” and “accuracy”, think again. I tried for 40 years with no luck!
 

Tony Wells

Former Vice President
Staff member
Administrator
Joined
Jan 22, 2011
Messages
6,880
Likes
1,916
#3
Surface roughness, in its various forms of specification is measure with a profilometer. I have one. There are several ways to evaluate surface finish, and these instruments can give various readouts according to what parameters you wish to see.

Most machinists however, use a fingernail comparator to make a physical comparison to a sample, usually machined in small sections on a plate. Each method, such as turning, milling, grinding, etc, have their own characteristics, and these sample plates have several of each, typically what is seen on engineering drawings. 500µ, 250µ, 125µ, 63µ, 32µ, 16µ etc.
 

itsme_Bernie

Active User
H-M Supporter-Premium Member
Joined
Jan 11, 2013
Messages
1,752
Likes
76
#5
Jim, thank you so much for posting this. I am eating it up! I copied and pasted it into my personal notes (with your name there, I always do that)

... I am also waiting for the next installment! You have fans now, watch out!

Bernie
 

jpfabricator

Active User
H-M Supporter-Premium Member
Joined
Sep 19, 2012
Messages
3,042
Likes
698
#6
I enjoyed this read very much. I am already waiting for the next chapter.

Jake Parker
 

nobog

Active User
Active Member
Joined
Jun 10, 2013
Messages
79
Likes
16
#7
OK, here's some more tidbits:

Drill Bits Explained

So what the deal with drill bits? How come they are packaged in 13, 15, 29 & 115 pc sets? Who figured out all those sizes?
The twist drill was patented in 1863 by Stephen A. Morse and since then a standardized system has been utilized ensuring uniformity, and interchangeability.

The “standard” set (American):

The standard set of drill bits is the 29 pc set that goes from 1/16” to ½” with 1/64” steps (.0156”). This is the de facto drill bit set. Cheap people buy 15 pc sets from 1/16” to ½” with 1/32” steps (.0315”). Really cheap people buy 13 pc sets that only go up to ¼”.

Wire Gage sizes & Letter sizes:


Number drill bit gauge sizes are analogous to, but different from, American wire gauge. Number gauge is routinely used from size 80 (the smallest - .0135”) to size 1 (the largest -.228”) followed by letter gauge size A (the smallest - .234”) to size Z (the largest - 413”).

The above (fractional, wire, letter) all add up to 115 pcs.

Metric

Metric drill bit sizes define the diameter of the bit in terms of standard metric lengths. Standards organizations define sets of sizes that are conventionally manufactured and stocked.

What’s a Jobber-length drill bit?

Jobber-length drills are standard drills, as compared with screw machine drills (stub drills) which have a much shorter length of flute. As a result screw machines drills are much more rigid, but cannot drill holes as deep as jobber-length drills. Where this name (jobber-length) came from is a complete mystery.

High Speed Steel?

High speed steel does not mean to run the drill bit at high rpm, it is a form of tool steel where the bits are much more resistant to the effect of heat. They can be used to drill in metal, hardwood, and most other materials at greater cutting speeds than carbon steel bits and have largely replaced them in commercial applications. There are seemingly endless coatings applied to bits these days but for the most part if you buy just plain High Speed Steel bits you will be OK and of course they are cheaper.

What’s a Silver and Deming Drill?


Again, I have no idea how that name came about but the long and short of it is that they are drill bit sizes from .5” dia to 1.5” dia. Most have ½” shanks. They are not made of silver. Large bits can take huge bites of materials and may not be suitable for a hand drill. A drill press or mill is recommended for these larger sizes.

118 vs 135 point angle:


Point angle is measured as the included angle, in other words a 118 degree point angle is more pointed than a 135 degree point angle. 118 degree is the standard whereas 135 allows for less “walking” and is sometimes referred to as “self centering” – these are generally more expensive than the 118 deg type although both work fine for most shop work.
The most common mistake when drilling is no coolant – use cutting fluid! This extends the life of the bit, allows for deeper cuts, and produces a better quality finish in the hole.

So what should I buy?

One thing not to buy is the so-called Maintenance Grade bits. These are OK to use if they were the last set on earth but the quality in general is very poor. Jobber length drills are as common as salt on Minnesota roads and are considered the standard drill bit. Although harder to find, the screw machine length drill bits are excellent for all around use also. Their shorter length makes them more ridged. Unless the price is right I would avoid fancy coatings, these are mainly used in industrial applications and offer little benefit to the average home mechanic. If you do any thread tapping you will most likely need all three sets (fractional, wire, letter) of bits. 115 pc sets can range from $100 to $300, like most things you get what you pay for, plan on $150 - $200 for a set of reasonable quality bits. In any event make sure it says High Speed Steel.

What not to do with a drill; try to “ream” the hole bigger by cutting on the side. Drill bits are not designed to cut on the side of the flutes. This is what end-mills and reamers are for.

Lastly, get a drill chart. This takes most of the confusion out of all the sizes and what drill bit to use for which tap.

Jim
 

chuckorlando

Active Member
Active Member
Joined
Nov 24, 2013
Messages
1,639
Likes
14
#8
Good read. I would add that you should wipe down your anvils on mics and the like before using and ring the air out your gage blocks before using. Calipers get you close but mics get you there
 

nobog

Active User
Active Member
Joined
Jun 10, 2013
Messages
79
Likes
16
#9
For those unfamiliar with gage blocks and how they "stack together" (wringing):

Wringing makes the gage blocks "stick" to each other when you combine sizes.

http://www.starrett-webber.com/GB46.html


The first time I experienced this phenomena I was totally amazed, it just didn't seem possible.

JK
 

chips&more

Active User
Active Member
Joined
Mar 19, 2014
Messages
2,062
Likes
1,391
#10
Again, excellent reading. Would like to add to the family of drill bits, split point and cobalt. The stubby 135° split point cobalt drill bit style gets most of the use in my home machine shop!
 

nobog

Active User
Active Member
Joined
Jun 10, 2013
Messages
79
Likes
16
#13
Re: Total Indicated Runout.

Since I am generally a lazy man, it would probably be easiest just to read:

https://en.wikipedia.org/wiki/Total_indicator_reading

The highlight of the article would be:
The purpose of emphasizing the "total" in TIR was to duly maintain the distinction between per-side differences and both-sides-considered differences, which requires perennial conscious attention in lathe work.

Jim
 

4gsr

HM Chief Foreman
Staff member
H-M Supporter-Premium Member
Joined
Aug 8, 2011
Messages
4,454
Likes
2,744
#14
..........snip...........

What’s a Silver and Deming Drill?

Again, I have no idea how that name came about but the long and short of it is that they are drill bit sizes from .5” dia to 1.5” dia. Most have ½” shanks. They are not made of silver. Large bits can take huge bites of materials and may not be suitable for a hand drill. A drill press or mill is recommended for these larger sizes.

......snip......


Jim
I may not have the history totally correct, but "Silver & Deming" was a brand name of a Blacksmith Drill Press, that was hand operated back in its day, before the 1900's. They were generally post mounted and had a spindle that had a 1/2" hole in the spindle with a square head set screw that held drill bits in place. And of course, the drill bits always had a 1/2" shank on them!

Ken
 

Bill C.

Active User
Active Member
Joined
Mar 11, 2013
Messages
1,389
Likes
152
#15
Surface roughness, in its various forms of specification is measure with a profilometer. I have one. There are several ways to evaluate surface finish, and these instruments can give various readouts according to what parameters you wish to see.

Most machinists however, use a fingernail comparator to make a physical comparison to a sample, usually machined in small sections on a plate. Each method, such as turning, milling, grinding, etc, have their own characteristics, and these sample plates have several of each, typically what is seen on engineering drawings. 500µ, 250µ, 125µ, 63µ, 32µ, 16µ etc.
----------------------------------------
I haven't seen roughness scale in years. I usually did my best to have a fine surface finish on my final cuts. If it didn't look good to me and had stock I might pull out some fine grade sandpaper to cleanup scratches. That depended on the amount of time allow on the job. I was referring to non-grinding operations.
 

Bill C.

Active User
Active Member
Joined
Mar 11, 2013
Messages
1,389
Likes
152
#16
Excellent reading for the home shop machinist…good job. But, if you think you have finally stopped the world from misusing/transposing the words “precision” and “accuracy”, think again. I tried for 40 years with no luck!

A very good point about transposing the words. Expectation and reality are two more words that can get misused. What the customer whether it is yourself, family or stranger expects to see or pay for and what what they get can be two totally different realities. Just in the mind's eye I guess.
 

4gsr

HM Chief Foreman
Staff member
H-M Supporter-Premium Member
Joined
Aug 8, 2011
Messages
4,454
Likes
2,744
#17
Surface roughness, in its various forms of specification is measure with a profilometer. I have one. There are several ways to evaluate surface finish, and these instruments can give various readouts according to what parameters you wish to see.

Most machinists however, use a fingernail comparator to make a physical comparison to a sample, usually machined in small sections on a plate. Each method, such as turning, milling, grinding, etc, have their own characteristics, and these sample plates have several of each, typically what is seen on engineering drawings. 500µ, 250µ, 125µ, 63µ, 32µ, 16µ etc.
I've always told inspectors that my calibrated fingernail was better than their profilometer could check to any day! They didn't like that. I was the engineer and pretty much could over ride what their findings were!
 

Tony Wells

Former Vice President
Staff member
Administrator
Joined
Jan 22, 2011
Messages
6,880
Likes
1,916
#18
I think most people overlook what the main purposes of providing a finish specification are. It's one thing to say, "Everything I make, I like to put a mirror finish on." vs I just make it as fast as I can. Really, it's not much different in a non-hobby shop to size tolerance. To minimize the time (=maximize the $$), there is no point at all in taking additional cuts to bring a feature closer to nominal size if it is within allowed limits. I'll admit being guilty at one time of being the machinist on the floor who would do exactly that, until it was drummed into my head that you can make more money using the tolerances given. It can really have a large bearing on time invested if the parts are large simply to make another (unnecessary) pass to bring a part closer to nominal.

It's the same with finishes with an additional aspect. I have seen instances where the surface can literally be toosmooth. I have pushed on a few cases to have drawing changes to make the point officially stated rather than have a part rejected by a receiving inspector based on his internal knowledge of part requirements.
 

The Liberal Arts Garage

Active User
H-M Supporter-Premium Member
Joined
Sep 4, 2013
Messages
478
Likes
119
#19
Two notes: stub/screw machine drills are shorter to gain clearance in multi- tool
holders ; also good to gain working height in small mills/ drills.BLJHB. Also,
In serious work, check each drill across the lips to check for diametric wear.
 

TC0853

Active Member
Active Member
Joined
Jan 31, 2016
Messages
30
Likes
6
#20
Although I knew, or at least have known most of this stuff, it's always helpful to see it altogether in one place, and sometimes just to hear, or read it, in somebody else's words, helps clear things up. Don't know why, it just is. And, I have always wondered why "Silver & Deming". I own a complete set, so it's nice to clarify that, for the most part, it doesn't really mean anything. Thanks for taking the time to bring this information to us.
 

The Liberal Arts Garage

Active User
H-M Supporter-Premium Member
Joined
Sep 4, 2013
Messages
478
Likes
119
#21
All though it is implied, do not fail to clean all surfaces and tools ( Including
your own moving parts)........BLJHB.
 

The Liberal Arts Garage

Active User
H-M Supporter-Premium Member
Joined
Sep 4, 2013
Messages
478
Likes
119
#22
Jim, thank you so much for posting this. I am eating it up! I copied and pasted it into my personal notes (with your name there, I always do that)

... I am also waiting for the next installment! You have fans now, watch out!

Bernie
. Please don't forget "stiff". calipers-- practically free at yard
Sales etc. Learn a "nice touch". Clean, delicately file as needed, adjust tension
to smooth, not sticky, motion .practice on known standards. Sizes unlimited !
Old Fart advice from.........BLJHB.
 

T Bredehoft

Active User
H-M Supporter-Premium Member
Joined
Dec 27, 2014
Messages
2,161
Likes
1,509
#23
Measure drill bits to size... only on the cutting end. half way down the drill they are smaller, only by a few thousands, but smaller.
 

Wreck™Wreck

Active User
H-M Supporter-Premium Member
Joined
Sep 29, 2014
Messages
1,941
Likes
1,509
#24
Twist drills have back taper, also have never had an issue with vernier calipers except for the 40" Mitutoyo where I work which is entirely to heavy (-:
Screw machine drills are short because the machines only make short parts mostly.

Gauge blocks are rather expensive for home use, a nice alternative are on size steel dowel pins which are within .0005 of nominal diameter when bought from reputable manufacturers and annular bearings are good for larger diameters as they are good for less then .001 on the ID and OD
 

atunguyd

Active User
Active Member
Joined
Jun 9, 2014
Messages
171
Likes
95
#26
To me the accurate not precise is incorrect. The target should show holes in the centre but the holes should be excessively large or not round. The shots are accurate but the precision of exactly where they hit is rough.

Sent from my SM-N920C using Tapatalk
 

Tony Wells

Former Vice President
Staff member
Administrator
Joined
Jan 22, 2011
Messages
6,880
Likes
1,916
#27
Mathematically, the center point of the average circle (overall position of the represented circle) is not too far from bullseye, therefore is accurate, but the spread is all over the place, and therefore not precise. The holes themselves are another feature and not germane to the definition. I believe this same reference is used in an example in Wiki, FWIW

https://en.wikipedia.org/wiki/Accuracy_and_precision
 
[6]
[5] [7]