Inside Micrometer Question

[T Long]

Hello,

I have a very basic question that I need an answer to.

I recently purchased a Starrett inside micrometer. It includes the various length extensions to measure from 2"-6".

When I insert the various sized rods, the mic does not read accurately.

Are the extentions supposed to be set against a ring or outside micrometer?

Also, the extensions have a machined flat area with a wrench provided that fits this flat area.

What is this wrench for?

Thanks in advance for answering this very basic question for a rookie.

[SGW]

How much of an error are you talking about? A couple thou? You should be able to adjust the rods with that wrench; the part of the adjusting rods it fits is a threaded nut, so you can adjust the legnth.

They would want to be set against a flat reference (like an outside micrometer), not a ring.

You might try giving Starrett a call and seeing if they have a "How to Adjust an Inside Micrometer" pamphlet.

[JCHannum]

How do you know it does not read accurately?
What model number is it? Different ones zero differently.
If you have a known dimension, they can be adjusted to that.

[T Long]

Thanks for the replies!

It has been off by as much as .015"-.020" depending on which extension rod is being used.

I have compared the accuracy by measuring it with an outside micrometer after installing the extensioin rod and "zeroing" the micrometer head.

The model that I'm using is the Starrett #124.

It certainly appears that the wrench provided is for the extension rods. However, the effort required to loosen the "adjuster tip" on the extension rod seems excessive. I have not tried to apply excessive pressure to loosen this.

Thanks again!

[T Long]

OK Guy's. I have figured this out.

I'm a little embarrased to say that it's a simple adjustment at the tip of the extension rod.

Thanks for your help!

Regards,

Trevor

[Spin Doctor]

When I was a young apprentice one of the first major tool purchases I made after the basics was a set of Starrett tubular inside mikes. I had it beat into my head that you never went by the reading on the inside micrometer. You always read the measurement off of an outside micrometer

[Forrest Addy]

Spin is right. A properly calibrated inside mike can be right on the money but you inevitably warm it when handing it making the ID mike expand a small but significant amount. It's better used as a transfer instrument.

If you transfer the reading to an outside mike you not only eliminate the expansion error but you reference the inside dimension to the very mike used to meassure the male part for which the bore is to be fitted.

Here's a factoid all machinists should keep in mind: most everything expands as its temperature increases. Bare skin handling of measuring tools used to take close tolerence readings leads to measurement errors. The international temperature standard for metrology (the study and practice of mechanical measurement) is 68 degrees F.

Iron and steel expands at 6.5 millionths of an inch per inch. Brass and bronze at 7.8 to 9.0 depending on alloy and aluminum at 9.6 to 11.2 - also depending.

If the steel 3" to 4" mike is 15 degrees warmer than the steel work, how much error is present in a 3.5000" reading and in which direction do you apply the correction? It aint much but it's enough to seriously affect a bearingrace fit.

When I ran the big machine tools I used to keep a little thermometer handy. Using it, I could compensate for temperature variation bringing 5 ft diameter machined features on aluminum castings still warm from machining to smack in the middle of the tolerence when restored to room temperature.

I remember some memorable arguements with staff inspectors who insisted on writing up my work for discrrepancy after they measured it on the machine applying no temperature correction.

The part immediately went to locked segregated storage. Days later, come time for the lessons learned meeting the measuring tools were re-applied to the deficient part and there it was, magically healed right in the middle of the tolerence. While the power that were puzzled over this, I explained about physics of thermal expansion as applied to accurate measurement.

No fault of the inspectors. They weren't machinists and as pipefitters or sheet metal worked had no experience with temperature affecting measurement.

[Thrud]

Forrest
The thermometer is an excellent Idea. It can be made far simpler by using one of the new no-contact IR digital thermometers. The are aften far more accurate as they can instantly read the surface temp without waiting for thermal transfer to take place. Even UAP/NAPA sells them for cheap now.

[darryl]

I've become aware of temperature-related errors while making parts, some of which had to be confined to quarters (the scrap bin). I understand the reasoning behind it all, but am puzzled as to why manufacturers don't use an NPO type of metal for calipers and micrometers. These metal alloys have no net expansion or contraction within reasonable temperature limits, and would seem ideal for measuring instruments. I read about this some time ago, but I don't remember the name given to the alloy.

[Oso]

"NPO", eh, I think you have just been "outed" as an electronics type maybe?

NPO (or COG) is the designation for (ceramic) capacitors which have a tempco in the low parts per million.

I bet those metals don't have good properties for lots of applications. "Invar" comes to mind, but it isn't an all-purpose material, as I recall.

[Forrest Addy]

Measuring tools having zero coefficients of expansion have been proposed but the metallurgy of Invar 36 for example is a bit expensive and it's about as durable as mild steel.

Supposedly Invar can be heat-treated but I've never seen discussion or claims of Invar's improved post heat treatment mechanical properties nor dimensional stability in the as heat-treated condition.

Invar's coeffecient of thermal expansion while low is not zero.

Here's a link that lists its properties.

http://www.edfagan.com/htmDocs/inv.htm

There's an advantage to using steel measuring instruments in a shop where the materials being worked are primarily steel and that is their coefficients of expansion are similar. No temperature corrections need to be applied so long as the temperature of the part and instrument are the same.

So while zero expansion might be a cool idea for measuring but the truth is maaybe not. Parts having tolerences close enough to include temperature correction as part of the manufacture and inspecion will still require the calculation of temperature effect if inspected in a non-controled environment. Thus there would be little practical advantage to purchase zero (?) temperature reactive measuring apparatus.

The ceramic gage blocks mentioned have their thermal expansion characteristics listed for the purpose of thermal expansion calculations. The principle advantage of ceramic gaging equipment is its surface durability not its low rate of thermal expansion.

[Thrud]

Forrest
Mitutoyo's Cera blocks are nearly the same expansion rate as steel (Cera is slightly lower). As you say they are very hard and also chemically inert. They wring better than Starrett/Webber Chrome Carbide. I really like them. They hold lab grade tolerances (for the most part) even with grade "B" blocks - a credit to Mitutoyo's manufacturing abilities.

[docsteve66]

The below is a condensation of years of debate on the metal expansion problem. I have not verified, so think and evaluate yourselves.
1. For outside mikes, the body expands out, the spindle also gets longer and the two movements counter act each outher. So for you need to correct for temp. The differnent materials expand at different rates. So to be correct you need tables of materials and error of instruments at temperatures.

2. for small mikes, the error is worse becasue the mike tends to be self correcting (the body/spindle ratios are self correcting but the metal being measured is not).

3. for larger mikes, it becomes more important that the error in the mike be known. The body is expanding with temp and has no off setting spindle expansion.

4. a series of holes will differ from each outher as the parent metal heats due to the heat of maching increasing as more holes are generated. The spacing will also vary, same reason.

5. After the part is made, let it sit overnite to "normalize", even hit it with hammer blows, and then remeasure. a careful miking will reveal a lot of the past history (which holes were first made, how dull/sharp tools are, how tools dulled in use).

6. Thanks to "tolerances" it usualy makes no difference any hoooo. But I bet RR could give us all a good lesson in how engines distort with heat. I know one out board manufacturer specifies loose (generous) clearances for consumer market, and tight for the knowledgable. the main differnece being thatthe "knowledgable" warm the engine before using the throttleand theconsumerdoes exactly the opposite.
Steve

[darryl]

Oso, right on the electronics designation, I'm a 'solderhead'. I didn't expect metals to be classified using that same term, but I don't know any better. I knew you guys would understand what I meant. Of course, that stands for 'negative positive zero', used mainly ( or only ) for capacitors. The above discussions are interesting in that many more factors are involved than just the instrument being accurate over a temperature range, the durability and repeatability of the instrument, are but two. Fitting parts to close toleraances is certainly an intriguing part of this hobby.