[Metrology] How to develop a feel for measuring round rods with a micrometer?

WobblyHand

WobblyHand

H-M Supporter - Diamond Member
H-M Lifetime Diamond Member
Joined
Dec 18, 2019
Messages
6,451
From the title, you can tell I'm having issues with accurately and precisely measuring the diameter of a turned rod. I'm trying to measure and correct the headstock alignment of my lathe. Annoyingly, it seems it is easier for me to measure the diameter by the chuck than further towards the tailstock. For some reason, the tailstock side measurements seem to be (for lack of a better description) bimodal. The readings are clustered around two different values, maybe 0.0005" apart.

Now I know that most of the answer is practice. This morning I made so many measurements that my hand was cramping. But it didn't seem I was making much progress. I have had one consistent run of 10 measurements with a standard deviation of 0.00023". The measurement set of 10 I made previous to it had a standard deviation of 0.002!

Can anyone share some tips and wisdom how to get the right feel to know whether one is truly on diameter? I try to lightly rock the mic to see if I'm really on diameter, but it always seems that my consistent measurements are a little tight on pull away. I'm concerned that the 'good' measurements are consistent but not necessarily accurate. It would seem this is a fundamental skill to master.

It's kind of frustrating to think you have it right, lock the mike and bring it back to the part and find out it is quite (0.001") loose, or on the other hand feel like you might be slightly flattening the part.

Measuring off lathe doesn't seem to be a problem for me. Just on lathe. If a kind soul could offer some insight, I'd greatly appreciate it.
 
What type of mic handle do you have? do you set the friction by your feel, or a friction thimble that ratchets?
 
Close your micrometer on zero.
If your mic has a ratchet it should zero after 1 or two clicks. Friction thimble it should zero when the thimble slips with a slow smooth breakaway.
Look very closely how it closes and the exact readings you get. Work on your technique until you can get the mic to read zero exactly every time.
Now repeat that feel and technique when taking a measurement.
 
I do have a ratcheting friction thimble. Use it most of the time, except when I feel like its lying to me.
 
Asm109 said:
Close your micrometer on zero.
If your mic has a ratchet it should zero after 1 or two clicks. Friction thimble it should zero when the thimble slips with a slow smooth breakaway.
Look very closely how it closes and the exact readings you get. Work on your technique until you can get the mic to read zero exactly every time.
Now repeat that feel and technique when taking a measurement.
Click to expand...
This is a 1-2" mic with a ratchet thimble. I can 'zero' it to a 1" standard, (that comes with the mic) but the feel is not the same as say using a 0-1" mic on a flat or zero. I'm measuring a 1.96" diameter aluminum rod in my lathe. Using a 0-1 mic is easy. The 1-2 is easy on flat stuff. I'm having problems being on peak diameter. Measuring flat stuff I'm plus minus a tenth at most. The round stuff seems to be the problem, at larger sizes.
 
Let me put it another way. If one is on peak diameter, sliding the caliper off peak either direction should feel the same, correct? Should be initially snug and then fall off (become loose)? Dunno, maybe I'm doing it all wrong, preconceived notions, etc. What is the right way to measure the diameter on lathe? Can't even believe I'm writing this... Seems I'm missing some tiny detail or technique that makes everything work!
 
OK, found a couple of things.

1) I wasn't locking the mic after setting the ratchet. A couple of times when removing the mic, I saw it rotate! This could account for the second distribution of readings. Once I locked the mic every time, the readings became more consistent. Can't say they are correct yet, but at least the consistency is improving.
2) Oops, sometimes I keyed in the wrong value in the calculator. Umm, yeah, that's going to mess things up. Going to record every reading on paper, then key it into the calculator. It will take a little longer, but is more audit-able.

Back to measuring stuff.
 
I think I understand what you're talking about and it might help to reflect on how a mic works. There are mics with friction and ratchet thimbles and some old ones with neither. You have the ratcheting type and those are intended to apply a consistent amount of pressure on the screw after making two clicks - not one, not three, just two. For friction thimbles you stop as soon as the thimble breaks free. In both cases, lock the spindle before removing it from the work.

Note that I didn't say to slide the mic or try to move the mic across the work to get the magic "feel" like we see the Youtube guys do. That technique is something they've developed over many years but for us mortals, using the design of the mic will usually be more consistent.

As for knowing when the high spots on either side of the work are contacting the spindle and anvil, look. I know that sounds cheesy but look and when the contact point of the work is in the center of both the spindle and anvil, gently rock the mic along the axis of the work to be sure it is seated evenly across the width of the spindle and anvil, then activate the thimble and then lock it. Over time, this becomes a subconscious thing and you won't even notice you're doing it. What you now have is the work contacting the spindle and anvil at a single point, the high point. You know it is so because you can see it and you've rocked the mic along the axis so it is fully seated. You've ratcheted twice so the right amount of pressure is applied to the spindle and you've locked it so the setting cannot change.

Now read it.
 
When working to tight tolerances, EVERYTHING matters. For instance, by just holding on to the micrometer with your hands while running a set of tests, it will grow, or warp, or both, and maybe more while you are doing the tests. To do fussy work, we must be truly fussy about everything that can change repeatability. I gave one example, there are MANY more subtle things that can change the measured results when everything else was done consistently from test to test. Beyond that, we hobbyists also have a penchant for trying to be far more fussy than what is needed for the project at hand (I plead guilty...). The first thing to determine is the needed accuracy for the measurement, and work to that standard as best as we can with the resources we have on hand.
 
You must log in or register to reply here.
Back
Top