PDA

View Full Version : Making a micrometer



Erik Brewster
04-16-2010, 01:39 PM
I've been thinking about making a micrometer. lately. This is purely a fun project. I don't have any delusions that my project will measure as accurately as my nearly new Mitutoyo digital mikes.

Has anyone done this? Any web sites / articles / books on the topic to inspire me?

Your Old Dog
04-16-2010, 01:56 PM
It's not all that bad of an idea. It would allow you to make all manner of measuring devices for your metal and wood working equipment. I bought a 1/4 40tpi tab and dye to do just that.

Tony Ennis
04-16-2010, 01:57 PM
There was a thread here recently about a clever way to make a shaft having two different threads. The threads work together to yield an effective TPI far in excess of anything most people can machine.

...of course, I don't remember the name of technique...

Weston Bye
04-16-2010, 01:58 PM
A 4-40 screw will give you .025" per turn. Add a barrel/knob with 25 equally spaced divisions and a pointer and you can measure to .001". Add a c-frame and anvil and you have a crude micrometer.

jugs
04-16-2010, 02:07 PM
There was a thread here recently about a clever way to make a shaft having two different threads. The threads work together to yield an effective TPI far in excess of anything most people can machine.

...of course, I don't remember the name of technique...

differential threads

philbur
04-16-2010, 02:18 PM
I think you will need to lathe cut the threads to produce a precision tool. Using a tap and die is a pretty rough and ready way of cutting threads.

Phil:)


It's not all that bad of an idea. It would allow you to make all manner of measuring devices for your metal and wood working equipment. I bought a 1/4 40tpi tab and dye to do just that.

Mcgyver
04-16-2010, 04:56 PM
I've been thinking about making a micrometer. lately. This is purely a fun project. I don't have any delusions that my project will measure as accurately as my nearly new Mitutoyo digital mikes.

Has anyone done this? Any web sites / articles / books on the topic to inspire me?

there was a series in Model Engineer, long time ago. 50's maybe? its was several parts and built a complete range of mics irrc, very well done article. the indexes are out there on the net then knowing what issues you could buy, borrow or get the issues from a library

Erik Brewster
04-16-2010, 05:05 PM
Sure enough. Here is a link that shows a few make a micrometer series over the years.

http://www.itech.net.au/modelengineer/ME_Response.asp

Spin Doctor
04-16-2010, 05:29 PM
If you make a bed stop for the lathe with a screw anddial you are in effect making a micrometer

Lew Hartswick
04-16-2010, 06:38 PM
I have a pix to show youall but that cursed Phto place has lost the
IMG generating mark. What did they do? Cursed programers wish
they'd learn to quit fixin it if it aint broke. :-(
...Lew...

J Tiers
04-16-2010, 08:30 PM
I made the one I use on my carriage stop.... it's actually reasonably accurate. I just wanted one I didn't have to baby.... It is 50 thou per rev


http://img.photobucket.com/albums/0803/jstanley/holdr2.jpg

Pherdie
04-16-2010, 09:39 PM
Here's another example. A shop built unit for setting tooling radius on a grinding system, 25 thou/turn @40tpi:
http://i657.photobucket.com/albums/uu291/pherdie/RadiusMicrometer.jpg

MotorradMike
04-16-2010, 09:42 PM
I'm all for making something to save $$ but it's really hard to beat what's available in the micrometer department at your local Canadian Tire Store.

Adjusts quickly and opens beers too.

http://farm5.static.flickr.com/4050/4526649075_d8daf7d3cc.jpg

darryl
04-16-2010, 09:53 PM
A while back I did some testing with threaded rod to see how accurately the threads were made. The best one I found was a piece of stainless 1/4-20, which of course gives 50 thou movement per turn. Over a distance, it had less compounded error than any of the other threaded rods I have- but I'm digressing already.

I have wanted to make something similar to a micrometer, and to make the measurement readout somewhat consistent with my other tools, I wanted to use 40 turns per inch, which gives a per-turn resolution of 25 thou. After farting around with the steady rest project, I used that to help in the cutting of some threads to test. Turns out these threads cut on the lathe are much more accurately laid out than any threaded rod, which I'm happy about of course and did somewhat expect anyway. So far I've turned about a 2 inch long section and compared it with a micrometer, and it tracks quite well. There was virtually no compounded error, which I have to attribute to two things- first the accuracy of the leadscrew on the lathe, and secondly the attention I paid to cleaning the lathe, the screw, the threading parts in general, etc, and lubing the ways.

So I'm happy that I'm able to turn a thread to the desired degree of accuracy- next is the material used to turn the threads on. The rod I used came out of a printer, and this particular one is about 5/16 diameter. Even though at 40 threads per inch the depth is quite shallow (less than 10 thou I think it was), and I was keeping the depth of cut very shallow, there will still be deflection of the rod during the cuts. Using the steady virtually eliminated this problem. I took my time- the first cut was 3 thou, next was 2 thou, and all the rest were 1 thou. After each cut, I ran in reverse and lightly held a file on the threads to burnish the burring back into the groove, then took a spring pass to further strip off the burring. I did this mostly for the sake of the bearings on the steady, but I think it also makes the job of cutting the threads easier, and I also believe it contributes to the accuracy of the result.

Somewhere down in the shop I now have a test piece of rod threaded to 40 tpi that is a proof of concept. Then I did another one to the absolute limit of what my lathe can do for tpi, which I think is 160 or so tpi. I haven't checked that one for compound error over distance, but looking at it with a magnifying glass it looks good. In particular, that was a test of the cutting tool to see if the point would hold up- in other words to see if the depth of the cut would be the same from one end of the thread to the other. Printer rods have so far given me the best results.

After that bit of experimenting, my attention turned to the nut that would engage the thread. I don't mean me- I did propose to it, but it turned me down :) sic- anyway I thought I would try using the thread itself to form the nut in some plastic. I took a piece of solid pvc, drilled a hole a tad smaller than the threaded rod OD, then bandsawed through the side of the hole and a bit past. This let me squeeze the plastic against the rod while turning it, running it back and forth a few times, and probably doing a combination of cutting and forming the threads in the plastic. With a screw to adjust the squeezing of the plastic I could do this, plus after the fact I could adjust the play.

So- I ended up with threads in the plastic piece, a method of removing the play, and as a bonus the threaded rod got polished up quite a bit. Pvc is pretty hard on HSS cutting tools, so in part I reasoned that it should be able to further debur and clean up the thread while the nut was being formed. That seems to have worked out also.

Of course, now I've gone on to some other projects, but I will get back to the 'custom micrometer' sometime soon.

loose nut
04-18-2010, 09:58 AM
Many years ago there was an article in Model Engineer magazine about a Royal Navy engineering student who had a final exam before being made up. He had to make a 1" perfectly square cube to .001" tolerance and a hole in a plate that it would just slide through. He was put in a room with the usual hand tools files, chisel, hacksaw etc. and a drill and threading gear. No mills, lathes or measuring equipment etc. Plus an assortment of stock. He had a fixed number of hours to complete the job.

At the end of his time he turned in a perfectly completed project. How did he do it. The first thing he made was a micrometer.

BobWarfield
04-18-2010, 11:44 AM
I'm all for making something to save $$ but it's really hard to beat what's available in the micrometer department at your local Canadian Tire Store.

Adjusts quickly and opens beers too.

http://farm5.static.flickr.com/4050/4526649075_d8daf7d3cc.jpg

John Stevenson has a digital one on a pipe wrench!

BW

Erik Brewster
04-19-2010, 04:32 PM
J Tiers and Pherdie, thanks for the pictures. Did either of you make an adjustable nut, like are in commercial micrometers?

Darryl, thanks for the story. It helps a lot to hear different ways to approach it.

Does anyone have an idea of how the holes for the anvil and spindle are made, so they are in line? It seems like it would be a good idea to get those in line very closely so there isn't a lot of fitting to get the faces parallel.

I think I'll have to try to knock out a simple one with the bandsaw and threaded rod to do a one better than a cardboard mock up. I'm sure it will help me make up my mind on how to progress.

Erik Brewster
04-19-2010, 04:35 PM
I'm all for making something to save $$ but it's really hard to beat what's available in the micrometer department at your local Canadian Tire Store.

Adjusts quickly and opens beers too.

http://farm5.static.flickr.com/4050/4526649075_d8daf7d3cc.jpg
I've got a set of those. They work really well. I found that when I got them, I stopped using them as wrenches, because there was no more guessing on nut sizes. I suppose I could have used calipers before, but never did until I had this type.

A.K. Boomer
04-19-2010, 07:05 PM
I have a pix to show youall but that cursed Phto place has lost the
IMG generating mark. What did they do? Cursed programers wish
they'd learn to quit fixin it if it aint broke. :-(
...Lew...


Be grateful you didn't go there and immediately have your computer gridlock, its been going on ever since P.B. and Ive just learned to live with it.