Rotary Laser centre finder

[Michael GilliganParticipant @michaelgilligan61133]

Posted by MICHAEL WILLIAMS on 25/01/2014 23:34:14:

 

Problem is though no matter how good a device is you are still trying to line up on a centre pop or such like making super accuracy a pointless exercise .

Far better to apply ingenuity to making laser devices which can locate tooling relative to absolute datums on machine or work .

What about laser equivalent of toolmakers buttons ?

.

For those thinking along these lines, the Ickey Ball is [literally] "a very good place to start".

It should work rather well in conjunction with the aforementioned Rotary device.

MichaelG.

Edited By Michael Gilligan on 25/01/2014 23:52:21

[John Stevenson 1Participant @johnstevenson1]

Not bad £106 to £138 a pop, think I'll order a dozen to play with

 

OK joking, its the link that's important

 

Found those crossed line laser pics.

 

 

And lined up at 100mm above the work.

 

 

Bombs away

 

Edited By John Stevenson on 26/01/2014 00:24:13

[Michael GilliganParticipant @michaelgilligan61133]

Posted by John Stevenson on 26/01/2014 00:23:46:

Not bad £106 to £138 a pop, think I'll order a dozen to play with

OK joking, its the link that's important

.

Yes, the price is unfortunate … Mine was amongst the goodies in a nice wooden Toolmaker's cabinet that I bought years ago. … Can't afford another though.

Question 1: How much should they cost ?

… given that the asterisked dimensions are good to a tenth of a thou'

Question 2: Does anyone have a bright idea how to make something along the same lines, that's good enough to be useful, and cheap enough to buy or make ?

MichaelG.

.

P.S. The DamBusters rig looks excellent !!

MichaelG.

[John McNamaraParticipant @johnmcnamara74883]

Hi All

I think Michael Williams has sparked an Idea with toolmakers buttons.

He got me thinking and I sketched up using AutoCAD how a 12mm wide by 15mm high button might show up with the cone of the laser beam directed upon it.

For the test I have displaced the button .025mm to the right therefore it is off the axis of the machine by that amount

I have angled the rotating beam 10 degrees from vertical.

As the image below shows the line of the intersecting laser beam does slope as would be expected.

However not a lot. Thinking on this It made it clear that what was needed was that the beam and should strike the button at a smaller angle. This would amplify any spindle alignment error. An angled button would be better but that would be very difficult to place on the work, I therefore drew another button the same size but with a chamfer on it leaving the base parallel. This would make it easy to place the button in the normal way. while providing a surface for the laser to strike.

As the following illustration shows the angle of the striking is now much greater as expected The cone displays the error very graphically over 5mm out of level. (note the displacement was not changed at .025mm) There would be little problem in correcting the .025mm error to maybe a quarter of that by levelling the cone.

AutoCAD was even kind enough to make the edge of the cone fuzzy as it would show to a lesser extent with the laser. Reviewing Dan Gelbart's video confirms this the image was quite clear when he applied the laser to the side of a bar to centre drill it.

He mentions 50 micrometre accuracy for his tool that's a fraction less than .002 inches. I guess he believes that is the minimum accuracy for all measurements possible with this tool? Cone on Cone I believe we can do a considerably better.

Marking out to .001 inches is possible on a good day but we must add to that the error left by the spotting drill starting off axis… And that was on a good day.

Toolmakers buttons….
Ok they are a bit of a pain to set up but they can locate a hole a lot more accurately than a centre pop.

There are a couple more images in my album

Regards
john

Edited By John McNamara on 26/01/2014 06:07:23

[John McNamaraParticipant @johnmcnamara74883]

Clarification of the following sentence in the above post, it is not Clear.

Marking out to .001 inches is possible on a good day but we must add to that the error left by the spotting drill starting off axis… And that was on a good day.

Should read

Marking out (with a centre punch) to within .001 inches accuracy is possible on a good day but we must add to that the error left by the spotting drill starting off axis… And that was on a good day.

The laser pointer can do a lot better than marking out… particularly when focussed on a toolmakers button with a chamfer.

Regards
john

[Michael GilliganParticipant @michaelgilligan61133]

John,

Thanks for the AutoCAD demonstration of concept.

… Very useful diagrams.

MichaelG.

.

Edit: If it's not too much trouble; would you be so kind as to try an Ickey Ball in that sketch.  I feel sure that this would be a superb way to set an "origin" datum point.

Edited By Michael Gilligan on 26/01/2014 09:26:07

[Michael GilliganParticipant @michaelgilligan61133]

I've just found this brief autobiographical note by our hero.

MichaelG

.

Edit: … the forum editing software won't let me insert the link. It keeps insisting that I must "Paste from Word"

Edit: … the software seems to be paranoid, but I have found a way

http://spie.org/

x42147

sorry, you will need to concatenate those lines, then add [.][x][m][l]

… with the brackets removed

 

Edited By Michael Gilligan on 26/01/2014 10:11:29

and many more times !!

Edited By Michael Gilligan on 26/01/2014 10:39:01

[Ian PParticipant @ianp]

Michael

Has your last post about not being able to insert a link been deleted?

I have had 7 attempts to post a link, all have failed even when I follow the instructions.

Ian

[John McNamaraParticipant @johnmcnamara74883]

Hi Michael

The Ball target as requested….

The cone works better, you need the length of the flat surface of the target cone to show the error. The small degree of difference between the beam angle and the cone angle is what does it. My example uses one degree, less than one degree might even be better.

it would also be a lot easier to make on the lathe. Just bore the hole, turn and lap the cylinder to exact size then and chamfer to form the cone and carefully part off. With care it should be pretty accurate, Not hardened but good enough for most work.

Regards
john

[Ian PParticipant @ianp]

Michael

Oh! I see its reappeared, together with all your edits.

I was trying to give the same link, it seems the forum software has been changed, but Katy has not informed us.

Ian P

[Michael GilliganParticipant @michaelgilligan61133]

John,

Many thanks for the prompt, and cost-saving (!), illustration.

MichaelG.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Ian Phillips on 26/01/2014 10:34:00:

Michael

Oh! I see its reappeared, together with all your edits.

I was trying to give the same link, it seems the forum software has been changed, but Katy has not informed us.

Ian P

.

Ian,

It looks like the forum software is probably blocking direct links to anything ending in "exemel".

MichaelG.

[Michael HornerParticipant @michaelhorner54327]

Hi

If you have to drill a hole in the job, couldn't you use a ball bearing?

You'd probably want to to stain it black otherwise you might get bejazzeld!

Cheers Michael.

[John McNamaraParticipant @johnmcnamara74883]

Hi Michael H

For referencing a workpiece A ball is not as good as a cone for centring with a Laser as shown above. However If you drilled and reamed a master hole in a piece that was to be removed and replaced in the mill several times you could place the cone target in the hole and reference the machine to that point each time. Or screw the reference cone to the workpiece for the duration of the job. making it absolutely rigid,

Various referencing targets could be made with different size spigots for different size holes,

Toolmakers buttons (with a chamfer for the laser not shown in the link below) just need a flat base to be screwed onto the work. **LINK**

Regards
john

[Ady1Participant @ady1]

Family braincell moment:

If the laser is slightly adjustable, ie with a grubscrew then the drill can spot a point and the laser can be adjusted to the spot point

Then the laser is spot on for the cutting point of that particular drill/mill

Occasionally I come across a drill which "isn't quite right" but I don't have a replacement

It may be the drill itself or the way the chuck holds it

Edited By Ady1 on 26/01/2014 11:52:36

[John McNamaraParticipant @johnmcnamara74883]

Hi Ady1

I think you have missed the beauty of this design it is inherently accurate working on first principals. The point rotates around the axis of the spindle bearings, All you need to do to change the circle or spot size is raise or lower the quill. The only adjustment needed is to align the laser so it crosses the axis to get the smallest spot size, but even if you don't do that it will still show a perfectly centred circle.

Regards
john

Edited By John McNamara on 26/01/2014 12:04:17

[Michael GilliganParticipant @michaelgilligan61133]

Posted by John McNamara on 26/01/2014 10:33:29:

Hi Michael

The Ball target as requested….

The cone works better, you need the length of the flat surface of the target cone to show the error.

.

John,

Sorry to labour this, but I've just realised that you have the light shining on, or near, the equator … I wonder if it would give more useful results if incident somewhere near the "Tropic of Cancer". [?]

MichaelG.

[John Stevenson 1Participant @johnstevenson1]

"Tropic of Cancer". [?]

Where's that ? I'm not into astrology

[Billy MillsParticipant @billymills]

Just over two years ago in **LINK** the idea of a slant mounted laser was discussed, ChrisS then made one and it went on show at AP on the SMEE stand. Sir John's Dambuster pictures were then posted as a "been there done it way before you" but it is not the conical beam idea but two projected lines.

Thanks John McNamara for posting the link, the video series looks very good- something interesting to watch and learn from.

Billy.

[John Stevenson 1Participant @johnstevenson1]

So we have to thank Sir Billy for the patent rights to this design.

Just read the linked post and to be honest I had forgotten about it [ or more to the point – not quite understood where Sir Billy was coming from ]

Must admit that my Dambusters centring device has been removed from the router and the 5v supply used to supply a decent LED and I'm back to the old fag paper method [ can't teach an old dog new tricks ! ]

However I can see a use for Sir Billys method on the mill picking up certain features, so where do I send the royalties ?

[John McNamaraParticipant @johnmcnamara74883]

Hi Michael G

Alas placing the beam on the tropics does not help make it more accurate.

The beam strikes the ball at too steep an angle. It is confirming the beam is close (within .025 inches) but for greater accuracy you need a cone. Dan Gelbart quotes 50 micron typical accuracy. But he does not mention a cone.

The cone is the best way with a one degree or less difference to the beam. Its a bit like; actually the same as the shadows getting longer near the sunset…

There must be a geometer in this forum who can express it mathematically.

Regards
John

[Michael GilliganParticipant @michaelgilligan61133]

Posted by John Stevenson on 26/01/2014 14:22:44:

"Tropic of Cancer". [?]

Where's that ? I'm not into astrology

.

Tropic of Cancer

[Michael GilliganParticipant @michaelgilligan61133]

Posted by John McNamara on 26/01/2014 15:28:01:

Hi Michael G

Alas placing the beam on the tropics does not help make it more accurate.

.

Thanks again, John

What I was hoping was that the circle of light would become emphatically ovoid when the ball was displaced from centre. … But, thinking further; that would only be obvious when viewed from above.

MichaelG.

[Thor 🇳🇴Participant @thor]

Thanks for the photo/explanation of the Mk 2 version Michael (Cox). I have a laser pointer approximately 12mm outer dia. I gues a new project is coming up.

Thor

Edited By Thor on 26/01/2014 16:10:14

[Boiler BriParticipant @boilerbri]

Good thread, just bought me a lazer from the lanes in meadow hall. Now to make the metalwork.

Excellent concept. Some very useful links as we'll..

Bri

[Author]

Posts