edge finder sensitivity

The recent thread on edge finders got me wondering.

I'm of the 'Spin it at 1000 or so, bring it slowly to the edge and watch for the "jump"" school. I have two edge finders, .500 and .200 diameter. All other things being equal, which is more accurate in finding an edge? The jump on the .200 is harder to see, but it almost seems more consistent.

Reply to
BillM
Loading thread data ...

I like the 0.200 better. Just personal preference.

Karl

Reply to
Karl Townsend

I favor the .200" size but not for accuracy (which I have not compared between it and the .500)

It's just easier, in my mind, to add .100" to the 'step off' dimension, rather than .250".

Having said that, I used the .500" one for about 40 years before I even saw a .200" one.

Lewis.

*****
Reply to
limeylew

The screws in my mill are 8 tpi . Easier for me to use the .500 , cuz it's zackly two turns of the wheel to center .

Reply to
Terry Coombs

I like the .200" dia. edge finder because my feed screws are .200" per turn and 1/2 turn gets me to the spindle center line :-)).

Here is a question: Has somebody here actually TESTED an edgefinder to determine accuracy and repeatability?

Never got to check it myself because the results were usually workable, but it would be nice to know.

Wolfgang

Reply to
wfhabicher

I've had a mill for like a month now . I'm happy to be on the piece I wanna machine ...

Reply to
Terry Coombs

I have, but my results would not be helpful to others because particular edgefinders can vary quite a lot.

A well-made edgefinder, e.g. Brown & Sharpe, is quite good. It gets even better if carefully lapped with fine compound, then carefully cleaned and lubricated for use.

I trust my best edgefinder to half a thou, though I very rarely need that kind of accuracy.

Reply to
Don Foreman

...

Another important factor is what you're mounting the finder in. We use a dedicated end mill holder and never remove the finder. Put the finder in a drill chuck and you haven't got much accuracy.

I'm "kick school". My son proved to me that this puts the 0 point 1/2 thou too far over. "flick it and center school" is more accurate if you've got young eyes and a serious touch for machine accuracy.

Karl

Reply to
Karl Townsend

I checked how repeatable mine was the other day because of this thread. It repeated to about half a thousandth turning at 600 RPM on a filed surface and I could get it within a thousandth by feeling for a step when it was stationary, which is how I use the center finder. When I want accuracy I surface-grind the block before milling and drilling. A surface-ground edge seems to behave differently with an edge finder, maybe it's the friction or lack of it?

I haven't made anything recently that could be used to check the accuracy of the edge finder. Maybe a carefully bored hole near an edge would do.

Reply to
Jim Wilkins

========= Anyone have accuracy numbers for the cheapscrew method of using a dowel pin and feeler gage or slip of paper?

Unka' George [George McDuffee]

------------------------------------------- He that will not apply new remedies, must expect new evils: for Time is the greatest innovator: and if Time, of course, alter things to the worse, and wisdom and counsel shall not alter them to the better, what shall be the end?

Francis Bacon (1561-1626), English philosopher, essayist, statesman. Essays, "Of Innovations" (1597-1625).

Reply to
F. George McDuffee

Unka George, (we're probably a similar age?)

I think that by using a dowel pin in a collet the accuracy of picking up an edge depends very much on the run-out(TIR)of the collet/pin combination.

The kick-off type edge locator is much less affected by run-out.

As others have mentioned the surface finish used to locate off is important; The better the surface finish the smaller the locating error, up to the point where the inherent process error governs.

The point about lapping the two sliding surfaces on the locator is also a good idea. If one were really anal one would check the flatness of these surfaces with an optical flat and a monochromatic light :-)).

As to whether the kick-off method or 'dead-nuts' concentricity gives better locating accuracy off an edge... I don't know for sure; but here is an observation based on physical principles:

1) Two objects cannot occupy the same space at the same time. 2) A 1" diameter pin will not fit into a 1" diameter hole without force.

Based on this I would venture an opinion that the kick-off method is more accurate; it is certainly more precise ie. it has greater repeatability.

.0005" accuracy is, in all likelihood sufficient for most work. I run at 1000 RPM or so when edge finding, and also place a drop of oil on the pick-up edge to "improve" things :-)).

Wolfgang

Reply to
wfhabicher

70 years

========== But of importance only if the edge is flat and parallel to the axis within 0.0005 inches.

Unka' George [George McDuffee]

------------------------------------------- He that will not apply new remedies, must expect new evils: for Time is the greatest innovator: and if Time, of course, alter things to the worse, and wisdom and counsel shall not alter them to the better, what shall be the end?

Francis Bacon (1561-1626), English philosopher, essayist, statesman. Essays, "Of Innovations" (1597-1625).

Reply to
F. George McDuffee

Actually, it doesn't matter how true the body of an edge finder is running as the action is only with the tip against the part that you are trying to locate.

Try it...mount an edge finder in a drill chuck with a 1/32" shim under one jaw, then locate the same edge without the shim...same location.

Mike

Reply to
The Davenport's

(...)

I dunno, but as a data point, teenut appears to say one can locate to within a tenth using oiled tissue paper and a mounted cutter.

formatting link
See entry "Wed, 05 Jan 2000 23:07:11 GMT"

--Winston

Reply to
Winston

George, (I'll be 63 this fall)

I read teenut's article you referred to; in fact I read the whole list.

When you asked about the accuracy of the dowel pin-in-collet plus cigarette paper method of edge finding, and if you wanted to locate the spindle centre line over the edge, then my posting above is correct in that the collet/spindle/pin run out limits the accuracy of edge location.

However, after reading teenut's old post it occurred to me that you may be talking about locating a work piece edge with the edge of a milling cutter. This is a horse of a different colour because here we want to determine the EFFECTIVE diameter of the cutter because the location of any edge machined, from the located edge, is determined by that EFFECTIVE cutter diameter which includes its wobble and run-out. The paper-stuck-on-work edge is very effective for this purpose.

I have seen people use feeler gauge stock to do this but I wouldn't do that to a sharp cutter. Brass shim stock is what I have used for this purpose because its thickness is easily added to any dimension. With cigarette paper it is hard to tell how much to allow for; I suppose for most purposes the .001" thickness of the paper is ignored.

Wolfgang

Reply to
wfhabicher

============= Use your micrometer to verify the paper thickness -- seems to be very consistent within a packet.

You appear to be correct in that the centerfinder used in the kick-out mode will compensate for some collet eccetricity, and does allow for checking under power/rotation, while a dowel pin and feeler gage will not.

FWIW - I use an electronic edge finder like this one.

formatting link
have date on these?

Wish I had a Schmidt if only to put in my tool box and look at.

formatting link
Also what has the groups experence been with the "audible" edge finder with a flat ground on the tip to produce a clicking noice? Less accurate? More accurate? than the standard edge finder? for examples see
formatting link
2/3s of way down)
formatting link

Unka' George [George McDuffee]

------------------------------------------- He that will not apply new remedies, must expect new evils: for Time is the greatest innovator: and if Time, of course, alter things to the worse, and wisdom and counsel shall not alter them to the better, what shall be the end?

Francis Bacon (1561-1626), English philosopher, essayist, statesman. Essays, "Of Innovations" (1597-1625).

Reply to
F. George McDuffee

Unka George:

This obviously isn't my normal sandbox, but in cruising through here this thread happened to catch my eye. Here are the relevant portions of a reply I made to an edge finder thread in amc.

=============================================================== "Newsgroups: alt.machines.cnc From: BottleBob Date: Thu, 07 Apr 2005 01:48:38 GMT Subject: Re: just how good are electronic edge finders?

Paul:

The solid stem electric edge finder is junk IMO. They run out even when new and all you have to do is overtravel a little too much and you tweak them. Electric edge finders are used with the spindle not turning. I also have a ball tip electric edge finder that I use from time to time but here too, you have to touch off your part, record the axis dimension then turn the edge finder 180 degrees and touch off again and subtract one dimension from the other and divide by 2 and move that amount to get the correct edge reading because they run out a little also. The thing about the ball type is it's spring loaded so if you overtravel a little you're not going to do it much damage. ===============================================================

formatting link
I bought a couple of those "audible" edge finders that make a clicking noise when they "kick-out". They are not very loud in a noisy shop environment. They have a small flat ground on the edge finder. I tried them a few times but the spinning flat has a tendency to dig into soft materials, so I don't bother to use them anymore.

The following bears on how the accuracy of an edge finder is influenced by the concentricity of the edge finder body. From an amc thread in 2003:

================================================================== Yesterday I made a $5.00 bet with another machinist, here is what it entails. Picture a 1/2" shank edge finder (.200 tip) in an ER 32 collet holder. Find the edge of a part to the nearest .0001, zero the axis. Take out the holder and put a narrow .010 brass shim between ONE side of the edge finder and the collet to simulate some severe runout. Re edge-find the same edge of the same part as before. Look to see how far off the axis reading is from before. I said it will probably be within .001 of the prior reading. We didn't have time to physically test this yesterday, but probably will Monday. Will I be $5.00 richer, or should I go try to dig up 500 pennies to pay off my gambling debt?

OK, here's what happened with the bet at work today. I took a 1/2" shank edge finder and put it in an ER-32 collet holder (made sure everything was clean with no mico chips that I could see), scaled the edge finder 1" from the top of the collet (I wanted to minimize any possible exaggeration if the collet taper or holder taper were off a little bit. Cleaned the spindle. Indicated a 1-2-3 block in the vise and cleaned out the spindle taper and installed the holder and edge finder. Called over my betting opponent and a witness that held the money, just a formality. Brought the Z axis down to -9.880 (recorded that figure so I could come to the same setting later), edge found the

1-2-3 block and zero'd the readout. Took the holder out of the machine and turned it 180 degrees to see if there was a difference, it was off about a .0001, no biggie. Took the tool holder out and went and made up a .015 think brass shim. The shim was about .400 wide and a little longer that the collet, I pre-bent it in a V-block with a transfer punch. Put the shim and the edge finder in the collet, not an easy fit, scaled the edge finder 1" from the collet. Tightened it up and went to the machine with my entourage. Brought the Z down to -9.880 and fed the edge finder over in .010 increments, then .001 increments, then .0001 increments until it kicked out. It ended up being .0001 different than the original edge find. Much cheering, dancing, and commotion was seen and heard coming from MY corner! My betting buddy was pretty convinced but he wanted to try it in a drill chuck (just in case I had rigged the collet to tilt the precise amount the get my reading), we put the shim on one of the three chuck jaws, inserted the edge finder, tightened it, and went back to the machine. To make a long story short it was out .0002 from the original edge find. He was convinced, and took his loss gracefully. A fine time was had by one and all. LOL BTW, I plan to use my windfall profits to buy donuts for the shop tomorrow morning. ===================================================================
Reply to
BottleBob

Thanks for the insight and some real world experience and numbers

-- much better than opinions.

FWIW -- I have always had the spindle turning [c. 500 rpm] when using the electronic edge finder. Also very careful sneaking up to a contact. Light is not very bright though when it makes contact.

Unka' George [George McDuffee]

------------------------------------------- He that will not apply new remedies, must expect new evils: for Time is the greatest innovator: and if Time, of course, alter things to the worse, and wisdom and counsel shall not alter them to the better, what shall be the end?

Francis Bacon (1561-1626), English philosopher, essayist, statesman. Essays, "Of Innovations" (1597-1625).

Reply to
F. George McDuffee

Unka George:

If the electric edge finder is spinning, AND if it runs out a little, it will probably light up when the high point contacts a conducting surface. But you really won't know if that high point is .0001", .001", or even .010" off center. That's why I would contact your part with the edge finder stopped (then zero out your axis), back off your part, turn the edge finder 180 degrees and contact your part again to see if there is a difference in the two positions of the edge finder. Half that distance difference should be your off-center runout

Reply to
BottleBob

I don't want to sound nit-picky about this but to be really certain about the effect of excentricity of the edge locator, the 180 degree turn-around test described above would need to be repeated at 90 degrees from the first tests. Unless, of course, you were really sure that at the first trial you indeed were at the high point of any excentricity.

Wolfgang

Reply to
wfhabicher

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.