cnc machining question

[Laz]

CNC machining question ( guess this is the right place for this question )

can anyone describe the fastest way possible to locate a part using a
dial-indicator mounted in a toolholder in the spindle of a vmc ? the part is
without tooling holes, the indicator is fairly simple, in that I would not
know what the offset is

the part might be a 3d form so I might not be able to use a mic to establish
its size

Thanx in advance

Laz

[Cliff]

Post by Laz
CNC machining question ( guess this is the right place for this question )
can anyone describe the fastest way possible to locate a part using a
dial-indicator mounted in a toolholder in the spindle of a vmc ? the part is
without tooling holes, the indicator is fairly simple, in that I would not
know what the offset is
the part might be a 3d form so I might not be able to use a mic to establish
its size
Thanx in advance

You sort of need some sort of feature to indicate
or use an edge finder on or something .....
What & where is it?

BTW, IF this is not the first operation on the part sometimes
you can add features for locating in earlier operations to locate
on/with later .... even if they later get machined off.

[Cliff]

Post by Cliff

Post by Laz
CNC machining question ( guess this is the right place for this question )
can anyone describe the fastest way possible to locate a part using a
dial-indicator mounted in a toolholder in the spindle of a vmc ? the part is
without tooling holes, the indicator is fairly simple, in that I would not
know what the offset is
the part might be a 3d form so I might not be able to use a mic to establish
its size
Thanx in advance

You sort of need some sort of feature to indicate
or use an edge finder on or something .....
What & where is it?
BTW, IF this is not the first operation on the part sometimes
you can add features for locating in earlier operations to locate
on/with later .... even if they later get machined off.

You can also sometimes indicate a known location on the table
and then the datum on the part ...... your axis positions can
then be subtracted to tell you how far that second feature
is from the known location, assuming that your spindle was
locked and your indicator readings the same. Or you can adjust
for the indicator reading a bit too but it's not quite as accurate as
you can get some "trig errors" due to the mechanism of some
indicators & how you have things positioned/set-up.

You also have part/axis alignment issues with only one indication
position. You usually need at least two (general case).

HTH

[Jan Nielsen]

Post by Laz
can anyone describe the fastest way possible to locate a part using a
dial-indicator mounted in a toolholder in the spindle of a vmc ? the part is
without tooling holes, the indicator is fairly simple, in that I would not
know what the offset is

Since you don't know the offset, you could start finding it by touching
off on a part of known size.

Position the point of the indicator as close to the center of the spindle as
possible. Touch the part on one side, zero the read-out and zero the axis.
Rotate the indicator 180°, move to the opposite side, and repeat the process.
Subtract the part thickness from the axis read-out, divide the result by two,
and you have the offset.

Same procedure if you just want to find the center of a part, only here you
have to perform the show in both the X and Y axis.

Post by Laz
the part might be a 3d form so I might not be able to use a mic to establish
its size

Some trig may be involved for error compensation when touching an angled
surface. The larger the diameter of your indicator point the larger the error.

[Laz]

Thank you Jan and Cliff;

I was thinking pretty much along the same lines, but often those with
experience know the tricks of the trade. The reason I was asking is that I
recently interviewed for a cnc setup/operator/programmer position where the
mgr said they don't use edgefinders to locate their parts- instead used
indicators- he did not elaborate beyond that and I figured that if I got the
job I'd learn ( and I was hoping to do just that ), unfortunately....

Further, edgefinders get damaged or subject to magnetism etc. and it might
be even faster to locate my part at the same time I'm doing my set up. I'm
going to try locating with my dial indicator 1st and compare with the
edgefinder and see if I can get some good numbers.

Thanx again

Laz

Post by Jan Nielsen

Post by Laz
can anyone describe the fastest way possible to locate a part using a
dial-indicator mounted in a toolholder in the spindle of a vmc ? the part is
without tooling holes, the indicator is fairly simple, in that I would not
know what the offset is

Since you don't know the offset, you could start finding it by touching
off on a part of known size.
Position the point of the indicator as close to the center of the spindle as
possible. Touch the part on one side, zero the read-out and zero the axis.
Rotate the indicator 180°, move to the opposite side, and repeat the process.
Subtract the part thickness from the axis read-out, divide the result by two,
and you have the offset.
Same procedure if you just want to find the center of a part, only here you
have to perform the show in both the X and Y axis.

Post by Laz
the part might be a 3d form so I might not be able to use a mic to establish
its size

Some trig may be involved for error compensation when touching an angled
surface. The larger the diameter of your indicator point the larger the error.
--
- JN -

[BottleBob]

Post by Laz
Thank you Jan and Cliff;
The reason I was asking is that I
recently interviewed for a cnc setup/operator/programmer position where the
mgr said they don't use edgefinders to locate their parts- instead used
indicators- he did not elaborate beyond that and I figured that if I got the
job I'd learn ( and I was hoping to do just that ), unfortunately....

Las:

Here are the relevant parts of a post I made in 2001.

===========================================================
Edge finders: I sometimes see experienced machinists wasting
time
creeping up on an edge finder as if making them kick out was damaging to
them. Just blast in with the advance button/knob set to .01, when it
kicks, back off - change to the .001 setting blast in again, when it
kicks, back off - change to the .0001 setting blast in till it kicks,
back off - THEN creep in.

Indicating Vises/angle plates/whatever: I see people setting
their
indicator to zero on one end of the vise and then cranking to the other
end - make an adjustment - then crank all the way back - make another
adjustment, etc. etc. What I do is snug one bolt and start cranking on
X and as soon as the indicator moves start making adjustments WHILE I'm
cranking. When you get the hang of it, it should take no more than one
pass to align it - then tighten it - and then a final check pass.

Indicator sweeping for center: I've seen people sweep one side
of a
block then the other - then back - and forth - and back - etc. I set my
indicator sweeper a little larger than the block on both ends - move the
block/table against the indicator and zero the indicator - then zero the
axis. Bring the indicator down on the other side of the block - move
the block/table against the indicator till it reads zero - check what
your axis readout says - divide it by two and move over that amount.
Then reset your indicator to read zero against one side of the block -
and then bring it up and down on the other side. You should be pretty
close, make any fine adjustments necessary. It takes longer to write
about it than it does to actually do it. <g>
If you want zero to be on one edge of the block rather than
center,
just measure the block and move half that amount and reset your part
zero point.
=======================================================

Here's the original thread name if it might be helpful:

Local: Sat,Jan 6 2001 5:57 pm
Subject: Re: Setup Tips

--
BottleBob
http://home.earthlink.net/~bottlbob

[PrecisionMachinisT]

Post by BottleBob

Post by Laz
Thank you Jan and Cliff;
The reason I was asking is that I
recently interviewed for a cnc setup/operator/programmer position where the
mgr said they don't use edgefinders to locate their parts- instead used
indicators- he did not elaborate beyond that and I figured that if I got the
job I'd learn ( and I was hoping to do just that ), unfortunately....

Here are the relevant parts of a post I made in 2001.
===========================================================
Edge finders: I sometimes see experienced machinists wasting
time
creeping up on an edge finder as if making them kick out was damaging to
them. Just blast in with the advance button/knob set to .01, when it
kicks, back off - change to the .001 setting blast in again, when it
kicks, back off - change to the .0001 setting blast in till it kicks,
back off - THEN creep in.
Indicating Vises/angle plates/whatever: I see people setting
their
indicator to zero on one end of the vise and then cranking to the other
end - make an adjustment - then crank all the way back - make another
adjustment, etc. etc. What I do is snug one bolt and start cranking on
X and as soon as the indicator moves start making adjustments WHILE I'm
cranking. When you get the hang of it, it should take no more than one
pass to align it - then tighten it - and then a final check pass.
Indicator sweeping for center: I've seen people sweep one side
of a
block then the other - then back - and forth - and back - etc. I set my
indicator sweeper a little larger than the block on both ends - move the
block/table against the indicator and zero the indicator - then zero the
axis. Bring the indicator down on the other side of the block - move
the block/table against the indicator till it reads zero - check what
your axis readout says - divide it by two and move over that amount.
Then reset your indicator to read zero against one side of the block -
and then bring it up and down on the other side. You should be pretty
close, make any fine adjustments necessary. It takes longer to write
about it than it does to actually do it. <g>
If you want zero to be on one edge of the block rather than
center,
just measure the block and move half that amount and reset your part
zero point.
=======================================================
Local: Sat,Jan 6 2001 5:57 pm
Subject: Re: Setup Tips

To pick up an edge using a test indicator you sweep the edge, and zero the
indicator dial at the 'sweet spot'...(indicator reversal point)

Then you pick the quill up a bit and hold a gage block against the surface,
again finding the sweet spot but this time on the block and at 180 deg
.....moving the table till they both agree.

When you get the same reading on the part as on the gage block, then you are
directly over the part edge.

All this assumes the quill as being absolutely perpendicular to the machine
x y travels (not necessarily to the table surface ).......on an accurate jig
borer, you should be able to index an edge datun to within .0001 this way
without much of a problem.

Variations of this method can also be used to very accurtaly locate the
center of any shaft as well as for picking up a gear tooth, space or keyway
when clocking work on a rotary fixture.

--

SVL

[Nocturnal Dragon]

Post by PrecisionMachinisT
To pick up an edge using a test indicator you sweep the edge, and zero the
indicator dial at the 'sweet spot'...(indicator reversal point)
Then you pick the quill up a bit and hold a gage block against the surface,
again finding the sweet spot but this time on the block and at 180 deg
.....moving the table till they both agree.

Reminder to self --

Bring home "Holes, Contours and Surfaces book by Moore" and scan and
post the pictures of this technique that are published in the book.

It's a very rare occasion I use an edge finder to pick up an edge. Using
an indicator insures the pickup is going to be as accurate as you need
it to be, whereas using an edge finder might get you pretty close but
you'll /never/ be 100% sure.

[PrecisionMachinisT]

Post by Nocturnal Dragon

Post by PrecisionMachinisT
To pick up an edge using a test indicator you sweep the edge, and zero the
indicator dial at the 'sweet spot'...(indicator reversal point)
Then you pick the quill up a bit and hold a gage block against the surface,
again finding the sweet spot but this time on the block and at 180 deg
.....moving the table till they both agree.

Reminder to self --
Bring home "Holes, Contours and Surfaces book by Moore" and scan and
post the pictures of this technique that are published in the book.

My post was clear enough.


Your going on here like much of what you've spent probly 2 decades in
learning / perfecting methods to obtain tolerences under .0005 has no
intrinsic value--yet I doubt without those skills you would be able hold
much in the way of a secure/ meaningful job.

If someone inexperienced can't / won't take the time to fill in the blanks
and FUCKING THINK about it...then tough shit, afaik.

Post by Nocturnal Dragon
It's a very rare occasion I use an edge finder to pick up an edge. Using
an indicator insures the pickup is going to be as accurate as you need
it to be, whereas using an edge finder might get you pretty close but
you'll /never/ be 100% sure.

An edge finder is a handy tool to get you close enough to use an indicator.

--

SVL

[Anthony]

Post by PrecisionMachinisT
An edge finder is a handy tool to get you close enough to use an indicator.

Might there be a use for a tool such as this, loaded in a collet in the
spindle.
I can think of a couple of uses...one of which would maybe be a
programmable end stop for repetitive work.

http://www.machines-cnc.net:81/projects/

[PrecisionMachinisT]

Post by Anthony

Post by PrecisionMachinisT
An edge finder is a handy tool to get you close enough to use an indicator.

Might there be a use for a tool such as this, loaded in a collet in the
spindle.
I can think of a couple of uses...one of which would maybe be a
programmable end stop for repetitive work.
http://www.machines-cnc.net:81/projects/

We used to use something similar to that all the time in the setting up of
hydraulic tracer mills....a pin with a flat ground to center, installed into
the valve in place of the stylus proper......a 'pointer' or a test indicator
was placed into the profiler spindle.

BTW:

Nice website so far--seems to be coming along quite nicely......suggest
perhaps you should put an email link up and request submittals from
newsgroup participants.

--

SVL

[Nocturnal Dragon]

Post by PrecisionMachinisT

Post by Nocturnal Dragon
Reminder to self --
Bring home "Holes, Contours and Surfaces book by Moore" and scan and
post the pictures of this technique that are published in the book.

My post was clear enough.

A picture is worth a thousand words...

Post by PrecisionMachinisT
Your going on here like much of what you've spent probly 2 decades in
learning / perfecting methods to obtain tolerences under .0005 has no
intrinsic value--yet I doubt without those skills you would be able hold
much in the way of a secure/ meaningful job.
If someone inexperienced can't / won't take the time to fill in the blanks
and FUCKING THINK about it...then tough shit, afaik.

You're torling, right? I hope.

If not, that's a rotten fucking attitude and you ought to be whailed
upside the head for it.

[ deleted 40 line rant I just typed, ain't worth it ]

Post by PrecisionMachinisT

Post by Nocturnal Dragon
It's a very rare occasion I use an edge finder to pick up an edge. Using
an indicator insures the pickup is going to be as accurate as you need
it to be, whereas using an edge finder might get you pretty close but
you'll /never/ be 100% sure.

An edge finder is a handy tool to get you close enough to use an indicator.

Nah. Extra work.

[PrecisionMachinisT]

Post by Nocturnal Dragon
You're torling, right? I hope.

Nope...

I am Weldor--from the planet Zorg.

--

SVL

[Nocturnal Dragon]

Post by PrecisionMachinisT

Post by Nocturnal Dragon
You're torling, right? I hope.

Nope...
I am Weldor--from the planet Zorg.

Thanks for clearing that up. :-)

[PrecisionMachinisT]

Post by Nocturnal Dragon

Post by PrecisionMachinisT

Post by Nocturnal Dragon
You're torling, right? I hope.

Nope...
I am Weldor--from the planet Zorg.

Thanks for clearing that up. :-)

All your machine are belong to us.

--

SVL

[Nocturnal Dragon]

Post by Nocturnal Dragon

Post by PrecisionMachinisT

Post by Nocturnal Dragon
Reminder to self --
Bring home "Holes, Contours and Surfaces book by Moore" and scan and
post the pictures of this technique that are published in the book.

My post was clear enough.

A picture is worth a thousand words...

And here it is;

Loading Image...

[PrecisionMachinisT]

Post by Nocturnal Dragon

Post by Nocturnal Dragon

Post by PrecisionMachinisT

Post by Nocturnal Dragon
Reminder to self --
Bring home "Holes, Contours and Surfaces book by Moore" and scan and
post the pictures of this technique that are published in the book.

My post was clear enough.

A picture is worth a thousand words...

And here it is;
http://bdhi.homeunix.net/images/edge-indicate.jpg

I suggest Anth should put it up on his page also.

--

SVL

[Anthony]

Post by PrecisionMachinisT

Post by Nocturnal Dragon
http://bdhi.homeunix.net/images/edge-indicate.jpg

I suggest Anth should put it up on his page also.

I'd like to have BD's permission to use the image first.

[Kathy]

"Anthony" <***@hotmail.com> wrote in message news:***@216.77.188.18.
..

Post by Anthony
http://www.macines-cnc.net:81/

fix that. the h is missing

[PrecisionMachinisT]

Post by Anthony

Post by PrecisionMachinisT

Post by Nocturnal Dragon
http://bdhi.homeunix.net/images/edge-indicate.jpg

I suggest Anth should put it up on his page also.

I'd like to have BD's permission to use the image first.

Better yet would be to get permission from Moore Tool.

--

SVL

[Nocturnal Dragon]

Post by Anthony

Post by PrecisionMachinisT

Post by Nocturnal Dragon
http://bdhi.homeunix.net/images/edge-indicate.jpg

I suggest Anth should put it up on his page also.

I'd like to have BD's permission to use the image first.

Go right ahead. There's a new one at the same URL above with Moore's
copyright notice added. I don't think reproducing a picture or few
from the book violates any copyright laws, right to fair use and all,
but they have contributed a lot to the industry with their fine machines
and tools so why not.

We have a couple Moore jig grinders (had a Moore jig grinder and jig
bore at the last place I was at) that are as old as I am, and they're
both still cranking out precision work on a daily basis.

[PrecisionMachinisT]

Post by Nocturnal Dragon
We have a couple Moore jig grinders (had a Moore jig grinder and jig
bore at the last place I was at) that are as old as I am, and they're
both still cranking out precision work on a daily basis.

I used to babysit one at the ole lazy b every so often...IIRC it had a OM
fanuc on it ....XYZA&C axis...with fully coordinated contouring...

It was a very nice machine.

===

Heres a machining center that they are currently offering :

http://www.thomasregister.com/olc/mooretool/5stand.htm

[
Optional Hydrostatic Spindle

High-speed oil hydrostatic spindle with speeds up to 80,000 rpm. Designed,
manufactured and built by Moore Tool. Superior stiffness, trueness of
rotation and surface finish are some of the hydrostatic advantages.

]

--

SVL

[Dave]

If you go to the Moore website, you can still buy the book from them
for $65.00 plus shipping and handling. Thanks for the tip, I'm now
thinking about buying a book I had never heard of until 2 minutes ago.

http://www.thomasregister.com/olc/44626000/publicat.htm

Thanks,
Dave

[PrecisionMachinisT]

Post by Dave
If you go to the Moore website, you can still buy the book from them
for $65.00 plus shipping and handling. Thanks for the tip, I'm now
thinking about buying a book I had never heard of until 2 minutes ago.
http://www.thomasregister.com/olc/44626000/publicat.htm
Thanks,
Dave

Thank BD, not me.

--

SVL

[Nocturnal Dragon]

Post by PrecisionMachinisT

Post by Dave
If you go to the Moore website, you can still buy the book from them
for $65.00 plus shipping and handling. Thanks for the tip, I'm now
thinking about buying a book I had never heard of until 2 minutes ago.
http://www.thomasregister.com/olc/44626000/publicat.htm
Thanks,
Dave

Thank BD, not me.

Don't thank me. Buy me something.

<g>

[dlevy]

Be careful ordering this book. They do not list shipping charges and the
support email doesn't work.

<snip>
Bring home "Holes, Contours and Surfaces book by Moore" and scan and
post the pictures of this technique that are published in the book.
<snip>
--
Black Dragon
That which does not kill us, makes us stranger.
-- Trevor Goodchild - AEon Flux

[Nocturnal Dragon]

Post by dlevy
Be careful ordering this book. They do not list shipping charges and the
support email doesn't work.

How much support do you need to read a fucking book?! If you can't do it
yourself, get your mommy to read it to you. Sheeze...

Post by dlevy

<snip>
Bring home "Holes, Contours and Surfaces book by Moore" and scan and
post the pictures of this technique that are published in the book.
<snip>

[BottleBob]

Post by dlevy
Be careful ordering this book. They do not list shipping charges and the
support email doesn't work.

Dlevy:

Try Barns & Noble, out of print books.

http://search.barnesandnoble.com/OopBooks/OopResults.asp?userid=u56k9pNB8a&sourceid=M000000269&title=Holes%2C+Contours+and+Surfaces

--
BottleBob
http://home.earthlink.net/~bottlbob

[Garlicdude]

Post by PrecisionMachinisT
To pick up an edge using a test indicator you sweep the edge, and zero the
indicator dial at the 'sweet spot'...(indicator reversal point)
Then you pick the quill up a bit and hold a gage block against the surface,
again finding the sweet spot but this time on the block and at 180 deg
.....moving the table till they both agree.
When you get the same reading on the part as on the gage block, then you are
directly over the part edge.
All this assumes the quill as being absolutely perpendicular to the machine
x y travels (not necessarily to the table surface ).......on an accurate jig
borer, you should be able to index an edge datun to within .0001 this way
without much of a problem.
Variations of this method can also be used to very accurtaly locate the
center of any shaft as well as for picking up a gear tooth, space or keyway
when clocking work on a rotary fixture.
--
SVL

Or you can use one of these magnetic edge finders, if the part isn't
magnetic you can crazy glue, or push with your fingers to hold it agains
the edge:

http://www1.mscdirect.com/CGI/NNPDFF?PMPAGE=1322&PARTPG=NNLMK32

We used to call the "h" shaped edge finder a "Queens chair" at the place
where I learned to machine. We used to make a few when we had free
time.

[PrecisionMachinisT]

Post by Garlicdude
Or you can use one of these magnetic edge finders, if the part isn't
magnetic you can crazy glue, or push with your fingers to hold it agains
http://www1.mscdirect.com/CGI/NNPDFF?PMPAGE=1322&PARTPG=NNLMK32
We used to call the "h" shaped edge finder a "Queens chair" at the place
where I learned to machine. We used to make a few when we had free
time.

Steve,

Well, that would be cheating.

<g>

Definately not very fast, but I was consistantly resolving to +/- .00025 or
less when I was running jigbore, according to QC...that's equal to or better
than the cmm that layout had on the floor at the time....

IOW, neither layout nor inspection dept could definitively tell me which
side of the tolerance band I was within.

Still....I have to give most of the credit to the machine tool
manufacturer--that lil Swiss jigbore thingy was so damned accurate it was
*almost* easy to hold that kind of tolerances.

--

SVL

[Charlie Gary]

Post by Laz
CNC machining question ( guess this is the right place for this question )
can anyone describe the fastest way possible to locate a part using a
dial-indicator mounted in a toolholder in the spindle of a vmc ? the
part is without tooling holes, the indicator is fairly simple, in
that I would not know what the offset is
the part might be a 3d form so I might not be able to use a mic to
establish its size
Thanx in advance
Laz

Here's a little info, but each part can present its own challenges. Imagine
a part with 4 parallel sides, like a rectangle, only tapered. The taper is
constant, so you don't have to worry about figuring out what different
angles do. Bring the indicator over to the x or y minus side. Starting on
the minus side just means you don't have to type in a minus sign when
calculating the center position. Anyway, get the indicator close enough to
the part you can turn the spindle with your hand and make the needle touch.
Ease the part over until the indicator reads zero at the highest point. Set
zero for that axis. Now either don't move the spindle up to clear the part,
or write down the z value shown on the screen, because all subsequent
readings must be at the same z level to account for the tapered walls. Now
go to the other side of the part and repeat the sweeping process. Did I
mention it's VERY IMPORTANT you don't bang, bump or move the needle. This
whole process relies on the distance from spindle centerline to indicator
tip being the same each time. Sorry, back to finding the center. After
getting the indicator to read zero at the highest point on the other side,
divide the number shown on the screen by two and move the part that far.
You are now over the center on that axis. Repeat for the other axis, and
you have found center both ways. If you want to check your work, set zero,
sweep the sides again and write down the numbers shown on the screen. If
you did it right, the only difference between the numbers should be the plus
or minus sign.

[Benthere]

Post by Charlie Gary
Here's a little info, but each part can present its own challenges. Imagine
a part with 4 parallel sides, like a rectangle, only tapered. The taper is
constant, so you don't have to worry about figuring out what different
angles do. Bring the indicator over to the x or y minus side. Starting on
the minus side just means you don't have to type in a minus sign when
calculating the center position. Anyway, get the indicator close enough to
the part you can turn the spindle with your hand and make the needle touch.
Ease the part over until the indicator reads zero at the highest point. Set
zero for that axis. Now either don't move the spindle up to clear the part,
or write down the z value shown on the screen, because all subsequent
readings must be at the same z level to account for the tapered walls. Now
go to the other side of the part and repeat the sweeping process. Did I
mention it's VERY IMPORTANT you don't bang, bump or move the needle. This
whole process relies on the distance from spindle centerline to indicator
tip being the same each time. Sorry, back to finding the center. After
getting the indicator to read zero at the highest point on the other side,
divide the number shown on the screen by two and move the part that far.
You are now over the center on that axis. Repeat for the other axis, and
you have found center both ways. If you want to check your work, set zero,
sweep the sides again and write down the numbers shown on the screen. If
you did it right, the only difference between the numbers should be the plus
or minus sign.

Wouldn't a co-ax indicator be easier?

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

[Nocturnal Dragon]

Post by Benthere

Post by Charlie Gary
Here's a little info, but each part can present its own challenges. Imagine
a part with 4 parallel sides, like a rectangle, only tapered. The taper is
constant, so you don't have to worry about figuring out what different
angles do. Bring the indicator over to the x or y minus side. Starting on
the minus side just means you don't have to type in a minus sign when
calculating the center position. Anyway, get the indicator close enough to
the part you can turn the spindle with your hand and make the needle touch.
Ease the part over until the indicator reads zero at the highest point. Set
zero for that axis. Now either don't move the spindle up to clear the part,
or write down the z value shown on the screen, because all subsequent
readings must be at the same z level to account for the tapered walls. Now
go to the other side of the part and repeat the sweeping process. Did I
mention it's VERY IMPORTANT you don't bang, bump or move the needle. This
whole process relies on the distance from spindle centerline to indicator
tip being the same each time. Sorry, back to finding the center. After
getting the indicator to read zero at the highest point on the other side,
divide the number shown on the screen by two and move the part that far.
You are now over the center on that axis. Repeat for the other axis, and
you have found center both ways. If you want to check your work, set zero,
sweep the sides again and write down the numbers shown on the screen. If
you did it right, the only difference between the numbers should be the plus
or minus sign.

Wouldn't a co-ax indicator be easier?

On a rectangularish part? Okaaaay...

[Charlie Gary]

Post by Benthere

Post by Charlie Gary
Here's a little info, but each part can present its own challenges.
Imagine a part with 4 parallel sides, like a rectangle, only
tapered. The taper is constant, so you don't have to worry about
figuring out what different angles do. Bring the indicator over to
the x or y minus side. Starting on the minus side just means you
don't have to type in a minus sign when calculating the center
position. Anyway, get the indicator close enough to the part you
can turn the spindle with your hand and make the needle touch. Ease
the part over until the indicator reads zero at the highest point.
Set zero for that axis. Now either don't move the spindle up to
clear the part, or write down the z value shown on the screen,
because all subsequent readings must be at the same z level to
account for the tapered walls. Now go to the other side of the part
and repeat the sweeping process. Did I mention it's VERY IMPORTANT
you don't bang, bump or move the needle. This whole process relies
on the distance from spindle centerline to indicator tip being the
same each time. Sorry, back to finding the center. After getting
the indicator to read zero at the highest point on the other side,
divide the number shown on the screen by two and move the part that
far. You are now over the center on that axis. Repeat for the other
axis, and you have found center both ways. If you want to check
your work, set zero, sweep the sides again and write down the
numbers shown on the screen. If you did it right, the only
difference between the numbers should be the plus or minus sign.

Wouldn't a co-ax indicator be easier?

The only, and I mean only time I've ever found a coax to be the better
choice is when I was setting tool holders to lathe spindle centerline. Then
they're just the ticket, but any other time I prefer my trusty BestTest
Brownie. Besides, it's an opportunity to see what the spindle components
feel like before you smell them getting hot.

[Proctologically Violated©®]

This thread has really made me appreciate my edgefinders! goodgawd...

Bottle's edgefinding suggestion was good. I too go in quick at .01, but
it's readily apparent when it first touches (less wobble), and then I go in
.001.

You all really do all this indicating stuff on non-circular material??
Damn...
I just indicate circular stuff and vises.
Good thing I got near-zero overhead....
----------------------------
Mr. P.V.'d
formerly Droll Troll

Post by Laz
CNC machining question ( guess this is the right place for this question )
can anyone describe the fastest way possible to locate a part using a
dial-indicator mounted in a toolholder in the spindle of a vmc ? the part is
without tooling holes, the indicator is fairly simple, in that I would not
know what the offset is
the part might be a 3d form so I might not be able to use a mic to establish
its size
Thanx in advance
Laz

[J. Nielsen]

On Fri, 29 Apr 2005 14:34:27 -0400, "Proctologically Violated©®"

Post by Proctologically Violated©®
You all really do all this indicating stuff on non-circular material??
Damn...
I just indicate circular stuff and vises.
Good thing I got near-zero overhead....

One word: "measure probe" OK, so it's two words, but who's counting.

Position it over a corner. Punch in a few numbers. Press The Green Button,
and by magic your X Y Z zero is stored in the setup register... unless the
gremlins interfered, in which case you just ruined a $1000 probe.

[Laz]

TO ALL:

Thanks for all the info- I'm going to print out the new methods here put
them in my toolbox and try them when I get a bit of time; certainly
experience is one thing you can't buy; I can learn from my mistakes or on
the advice of others- I prefer the latter. I especially found the method on
locating the center as described by PrecisionMachinis T interesting. What I
do is make my indicator large enough to give a measurement on opposite
sides, sweep at 3, 6 and 9 o'clock noting the difference, divide the total
by 3 and adjust my location at 2 places 90 deg to each other to bring my
indicator to read the average.

Laz


"J. Nielsen" <***@post8.tele.dk> wrote in message news:***@4ax.com...
On Fri, 29 Apr 2005 14:34:27 -0400, "Proctologically Violated©®"

Post by Proctologically Violated©®
You all really do all this indicating stuff on non-circular material??
Damn...
I just indicate circular stuff and vises.
Good thing I got near-zero overhead....

One word: "measure probe" OK, so it's two words, but who's counting.

Position it over a corner. Punch in a few numbers. Press The Green
Button,
and by magic your X Y Z zero is stored in the setup register... unless the
gremlins interfered, in which case you just ruined a $1000 probe.

[Cliff]

Post by BottleBob
What I
do is make my indicator large enough to give a measurement on opposite
sides, sweep at 3, 6 and 9 o'clock noting the difference, divide the total
by 3 and adjust my location at 2 places 90 deg to each other to bring my
indicator to read the average.

??

[Laz]

Post by BottleBob
What I
do is make my indicator large enough to give a measurement on opposite
sides, sweep at 3, 6 and 9 o'clock noting the difference, divide the total
by 3 and adjust my location at 2 places 90 deg to each other to bring my
indicator to read the average.

??
--
Cliff

yeah; it's easier if I just show you.

Laz

[Cliff]

Post by Laz

Post by BottleBob
What I
do is make my indicator large enough to give a measurement on opposite
sides, sweep at 3, 6 and 9 o'clock noting the difference, divide the

total

Post by BottleBob
by 3 and adjust my location at 2 places 90 deg to each other to bring my
indicator to read the average.

yeah; it's easier if I just show you.
Laz

When it reads the *same* in all 3 places the spindle's
centerline is directly above the location (assumig you are
indicating a round feature)?

[Laz]

The centerline is in the centre of the diameter; when I sweep with my
indicator the number on he dial remains constant- sometime I might
(handwheel) tweak it a bit, but only a few tenths to fine tune my location
because I am roughly doing the math in my head as I work.

Laz

Post by Cliff

Post by Laz

Post by BottleBob
What I
do is make my indicator large enough to give a measurement on opposite
sides, sweep at 3, 6 and 9 o'clock noting the difference, divide the

total

Post by BottleBob
by 3 and adjust my location at 2 places 90 deg to each other to bring my
indicator to read the average.

yeah; it's easier if I just show you.
Laz

When it reads the *same* in all 3 places the spindle's
centerline is directly above the location (assumig you are
indicating a round feature)?
--
Cliff

[Cliff]

Post by Laz
The centerline is in the centre of the diameter; when I sweep with my
indicator the number on he dial remains constant- sometime I might
(handwheel) tweak it a bit, but only a few tenths to fine tune my location
because I am roughly doing the math in my head as I work.

Ummm ... then what are you dividing something by 3 for?

[Laz]

I'm checking the radius at 3 places, 90 deg to each other, totaling the
difference and diving by three to obtain an average distance any of those
locations should be. My centre is a constant distance from anywhere on the
circumference. My handwheel allows me to adjust on x and y axis on a flat
plane perpendicular to the head of the vertical mill; therefore, if I sweep
my indicator in line with either of those axis and adjust my distance to the
average, then sweep 90 deg to the other axis and adjust for that I am very
near my centre; a subsequent sweep to check will reveal if and by about how
much I have to adjust my location.

For example:
After placing my indicator at the eyeballed centre of the radius and
adjusting my indicator to give me a measurement over the entire sweep area:
My first check my .001 indicator tells me the following:
(-)x axis ( 9 o'clock ) -2.5
(+)y axis ( 12 o'clock) +1.0
(+)x axis ( 3 o'clock) +3.0

my total variation is conveniently +1.5
divided by 3 gives me +0.5

if I go back to my 1st location (-)x axis ( 9 o'clock ) and (handwheel)
adjust my location on the x axis to make my indicator read +0.5;
then my 2nd location (+)y axis( 12 o'clock) and adjust my location on the y
axis to make my indicator read +0.5
I can sweep the entire radius and see little variation meaning depending on
the circumstances, I may or may not have to fine tune my location.

Laz

Post by Cliff

Post by Laz
The centerline is in the centre of the diameter; when I sweep with my
indicator the number on he dial remains constant- sometime I might
(handwheel) tweak it a bit, but only a few tenths to fine tune my location
because I am roughly doing the math in my head as I work.

Ummm ... then what are you dividing something by 3 for?
--
Cliff

[Cliff]

Post by Laz
I'm checking the radius at 3 places, 90 deg to each other, totaling the
difference and diving by three to obtain an average distance any of those
locations should be. My centre is a constant distance from anywhere on the
circumference. My handwheel allows me to adjust on x and y axis on a flat
plane perpendicular to the head of the vertical mill; therefore, if I sweep
my indicator in line with either of those axis and adjust my distance to the
average, then sweep 90 deg to the other axis and adjust for that I am very
near my centre; a subsequent sweep to check will reveal if and by about how
much I have to adjust my location.
After placing my indicator at the eyeballed centre of the radius and
(-)x axis ( 9 o'clock ) -2.5
(+)y axis ( 12 o'clock) +1.0
(+)x axis ( 3 o'clock) +3.0
my total variation is conveniently +1.5
divided by 3 gives me +0.5
if I go back to my 1st location (-)x axis ( 9 o'clock ) and (handwheel)
adjust my location on the x axis to make my indicator read +0.5;
then my 2nd location (+)y axis( 12 o'clock) and adjust my location on the y
axis to make my indicator read +0.5
I can sweep the entire radius and see little variation meaning depending on
the circumstances, I may or may not have to fine tune my location.
Laz

Post by Cliff

Post by Laz
The centerline is in the centre of the diameter; when I sweep with my
indicator the number on he dial remains constant- sometime I might
(handwheel) tweak it a bit, but only a few tenths to fine tune my

location

Post by Cliff

Post by Laz
because I am roughly doing the math in my head as I work.

Ummm ... then what are you dividing something by 3 for?

That sounds like a hard & complicated way to do things ......
and probably not all thaa accurate at getting first approximations.

Why not just bring it in in X, then in Y? Then check 360 degrees
as a fail safe?

Think about that divide by 3 stuff ....

[Laz]

Post by Cliff

Post by Cliff

Post by Cliff

Post by Laz
The centerline is in the centre of the diameter; when I sweep with my
indicator the number on he dial remains constant- sometime I might
(handwheel) tweak it a bit, but only a few tenths to fine tune my

location

Post by Cliff

Post by Laz
because I am roughly doing the math in my head as I work.

Ummm ... then what are you dividing something by 3 for?

That sounds like a hard & complicated way to do things ......
and probably not all thaa accurate at getting first approximations.
Why not just bring it in in X, then in Y? Then check 360 degrees
as a fail safe?
Think about that divide by 3 stuff ....
--
Cliff

A hard way? Hardly. For me numbers are easy, fiddling around, little here
little there little bit the other way etc. was just a frustrating waste of
time. My method gives me a quick and direct way to find the center-and I use
it because it works. By the way, it is my nature to double check my work
before proceeding- sometimes to the chagrin of my supervisors

[jon_banquer]

Post by Laz

Post by Cliff

Post by Cliff

Post by Cliff

Post by Laz
The centerline is in the centre of the diameter; when I sweep with my
indicator the number on he dial remains constant- sometime I might
(handwheel) tweak it a bit, but only a few tenths to fine tune my

location

Post by Cliff

Post by Laz
because I am roughly doing the math in my head as I work.

Ummm ... then what are you dividing something by 3 for?

That sounds like a hard & complicated way to do things ......
and probably not all thaa accurate at getting first approximations.
Why not just bring it in in X, then in Y? Then check 360 degrees
as a fail safe?
Think about that divide by 3 stuff ....
--
Cliff

A hard way? Hardly. For me numbers are easy, fiddling around, little here
little there little bit the other way etc. was just a frustrating waste of
time. My method gives me a quick and direct way to find the

center-and I use

Post by Laz
it because it works. By the way, it is my nature to double check my work
before proceeding- sometimes to the chagrin of my supervisors
A hard way? Hardly.

If it works for you that's all that matters.

jon

[Cliff]

Post by jon_banquer

Post by Laz
A hard way? Hardly.

If it works for you that's all that matters.

I see that you could follow none of it <G>.
How's Kinko's?

[Cliff]

Post by Laz

Post by Cliff

Post by Cliff

Post by Cliff

Post by Laz
The centerline is in the centre of the diameter; when I sweep with my
indicator the number on he dial remains constant- sometime I might
(handwheel) tweak it a bit, but only a few tenths to fine tune my

location

Post by Cliff

Post by Laz
because I am roughly doing the math in my head as I work.

Ummm ... then what are you dividing something by 3 for?

That sounds like a hard & complicated way to do things ......
and probably not all thaa accurate at getting first approximations.
Why not just bring it in in X, then in Y? Then check 360 degrees
as a fail safe?
Think about that divide by 3 stuff ....

A hard way? Hardly. For me numbers are easy, fiddling around, little here
little there little bit the other way etc. was just a frustrating waste of
time. My method gives me a quick and direct way to find the center-and I use
it because it works. By the way, it is my nature to double check my work
before proceeding- sometimes to the chagrin of my supervisors

Laz,
You have two independant variables: X & Y.
Using part of the error in Y to adjust X just adds steps
and noise to the method as it is unrelated.

Consider: You are dead on in X but off by .2 in Y.
Are you going to adjust X by .2/3? Which way? + or -?
See? It does not get you to the correct position as fast ..
takes more measurements & adjustments.

Not that jb knew what the subject was <G>.
Kinko's must be nice this time of year ...

[Laz]

Post by Cliff
Laz,
You have two independant variables: X & Y.
Using part of the error in Y to adjust X just adds steps
and noise to the method as it is unrelated.
Consider: You are dead on in X but off by .2 in Y.
Are you going to adjust X by .2/3? Which way? + or -?
See? It does not get you to the correct position as fast ..
takes more measurements & adjustments.
Not that jb knew what the subject was <G>.
Kinko's must be nice this time of year ...
--
Cliff

I did it out on my CAD here, working with the typical rounded values given
in my example as absolutes and a typical .5 in diameter tooling hole I found
my method puts me .00035 in. of center. Working with my .001 indicator, that
is the result of an initial adjustment.

The divide by two on x and then y method won't work with my indicator, maybe
an edgefinder, but then I would have to put my indicator in and sweep to
check my location. Whatever works for you. My way works for me.

Laz

[Cliff]

Post by Laz

Post by Cliff
Laz,
You have two independant variables: X & Y.
Using part of the error in Y to adjust X just adds steps
and noise to the method as it is unrelated.
Consider: You are dead on in X but off by .2 in Y.
Are you going to adjust X by .2/3? Which way? + or -?
See? It does not get you to the correct position as fast ..
takes more measurements & adjustments.
Not that jb knew what the subject was <G>.
Kinko's must be nice this time of year ...

I did it out on my CAD here, working with the typical rounded values given
in my example as absolutes and a typical .5 in diameter tooling hole I found
my method puts me .00035 in. of center. Working with my .001 indicator, that
is the result of an initial adjustment.
The divide by two on x and then y method won't work with my indicator,

No need to divide by two ..... just get on center in X first then do
the same in Y. Then sweep the circle to doublecheck everything.

Post by Laz
maybe
an edgefinder, but then I would have to put my indicator in and sweep to
check my location. Whatever works for you. My way works for me.

Edgefinders find edges one at a time .... no numbers are given,
usually.

Not that jb knew what the subject was <G>.

[Nocturnal Dragon]

Post by Proctologically Violated©®
You all really do all this indicating stuff on non-circular material??

If you're working with precision, there are few choices. An indicator is
the most reliable and economical choice.

As others have pointed out, there are tools like "h" or "chair" gages
to aid in picking up edges, but burrs and stuff that otherwise would be
seen when using an indicator can cause errors when using those tools

The only times I don't use an indicator is when doing stuff like the
biohazard symbol we carved in the side panel of my sons pimped out
computer case. Here, eyeballing it was plenty close enough... :-)

Loading Image...