Edge Finder Question

[tornitore45]

good response, thank you all.
I zeroed my dials on the rear left corner so the part was all under positive dial readings.
I made a dual cylinder block Ray Hasbrouck No7.
has a bout 32 holes and came out pretty accurate.

I always zero the dial on the first snap, then try several time and fine adjust until it snap as close as zero as I can tell.

I will experiment some more.
One thing I do not like about the duals is that they are difficult to insert in the collet, the moving part goes in and then you have to align the body perfect. Then when I pull them out of the collet I do not want to pull the parts apart and have to push it out from the back.

[GlennW]

So hold on a minute here!!

I was just looking in a book, "Machine Shop Practice" in the chapter "precision hole location-the jig borer" and guess what? It details the use of the cylindrical .2 edge finder. It states that the proper method is to intentionally knock the contact cylinder off axis with the spindle rotating, move toward the edge until contact is made and slowly advance the table UNTIL THE WOBBLE OF THE CYLINDRICAL CONTACE CEASES, NO FURTHER, THE TABLE MOVEMENT IS STOPPED AT THIS POINT. At this point you will be exactly .1" from the true edge. This is for the cylindrical style, not the cone point style.

I then went to the Starrett website and looked up their edge finders and low and behold, it states to do the same unless I'm not understanding it!

http://www.starrett.com/pages/1512_t_an ... dge_finder
(the collet adapter at the bottom of the page is interresting!)

Pretty much the opposite of what everyone seems to do!

Now I gotta do the test all over again to see what's up with that as I was nearly perfect when it kicked

Here is a video for anyone that may not be familiar with them.

http://www.youtube.com/watch?v=f0od-cp_9dg

[seal killer]

Glenn--

That is news to me, as well. (Ok. Most stuff in this business is news to me.)

I had never seen an electronic edge finder in use. Thanks for that video and the info on the mechanical devices.

--Bill

[mechanicalmagic]

I have always gone to the snap.

So, I just measured 3 of my .200" edge finders, two were .0001" large, one on size. I'm staying with my old method.
I think, given the real world in my shop, there is no way a .20000" diameter is going to run .10000" from the edge of a part, without some friction, causing the snap.

Maybe today is the time I learn a better way?

Dave J.

[Harold_V]

UNTIL THE WOBBLE OF THE CYLINDRICAL CONTACE CEASES, NO FURTHER

Makes a great story, but the human eye can't discern movement under about .003", so it would not be reliable.

Harold

[GlennW]

Makes a great story, but the human eye can't discern movement under about .003", so it would not be reliable.
Harold

Agreed

Another theory vs practice deal.

I believe that it may be the perfect position in theory, but in practice would be nearly impossible to determine when perfect alignment was achieved.

It would also require a perfectly concentric spindle.

[tornitore45]

Glen, I looked at the movie you posted, but my take is a little different.

Clearly by the time the finder kicks you are past the edge everso slightly.
Judging when/where the wobble chnges from an invisible amount to zero is a matter of judgement and can be wrong.

What I got from the fideo is that if you back up after it has kiched and it kicks back to centered, then you end the motion when it finally centers.
Assuming the snap back is as accurate as the snap out, backing out from offset to center is a positiveindication of the edge.

We are splitting hairs here, my machining accuracy is not that good or critical but from now on I will do the reverse snap.

[GlennW]

Possibly so, but by backing out slightly, you reversed the handwheel/dial so you now have no reference as to where the correct zero location should be!

[tornitore45]

Why did not think of that? (backlash)

Because sitting in front of the 'puter theorizing is a far cry from being in the shop, with oily hands on the dials.

Is pretty settled then: Advace from wobble to offset, STOP, zero the dial, retry.

[larry_g]

I am purely talking out my butt here as I have not been in the shop for over a year now but I seem to remember that sometimes I would grab an edgefinder that was a 2X diameter of the lead on the screw. Find the edge and then get out of the way and do one crank on the handwheel to center. Like I say just dusting off poor memory but a factor in choosing what edgefinder if you have to rely on the dials of the machine.

lg
no neat sig line

[Harold_V]

I am purely talking out my butt here as I have not been in the shop for over a year now but I seem to remember that sometimes I would grab an edgefinder that was a 2X diameter of the lead on the screw. Find the edge and then get out of the way and do one crank on the handwheel to center. Like I say just dusting off poor memory but a factor in choosing what edgefinder if you have to rely on the dials of the machine.

lg
no neat sig line

The same method applies to any size edge finder. Knowing the radius involved, the dial setting is made such that when the dial is turned to "0", you are on the edge of the part.

I have used the same ½" edge finder since 1957. When I use a mill with .200" lead screw, I set the dial @ 150. It is then advanced 1¼ turn to "0", which is the edge of the part.

It's generally a good idea to not mess with a dial once it has been coordinated with the part. Otherwise, you are never sure that you haven't introduced error. By using the method you proposed, the veracity of the setting can be assured.

Harold

[10 Wheeler Rob]

Thanks for the good discussion.

I usually find the edge, reset the dial, and then find it a second time to check to see if the dial is set right.

Rob