Homebrew NC Mill

[Spehro Pefhany]

I read in sci.electronics.design that Garrett Mace <[email protected]>


That's 'thread chasing'. It's getting on for 200 years old.

I note that turning metal on lathes is getting on for (at least) 1800
years old.

Best regards,
Spehro Pefhany

 

[John Woodgate]

I read in sci.electronics.design that Spehro Pefhany <speffSNIP@interlog

I note that turning metal on lathes is getting on for (at least) 1800
years old.

Yes, but specifically making threads with the type of tool described is
'chasing'. You don't need a lead-screw, just a swift hand.

 

[Garrett Mace]

That's 'thread chasing'. It's getting on for 200 years old.
--

Thread chasing is not the same as thread milling. Thread chasing is more
suited for lathe operations, it's basically the same as single-point
threading except it has multiple teeth with a tapered profile, meaning only
one pass is necessary. The thread chasing tool is run the entire length of
the threads. A thread milling tool has the same tooth profile along its
entire length, and cuts all or most of the threads at the same time on the
inside of the hole. Only one circular move around the hole is required,
interpolated with one vertical move of one thread pitch. A thread milling
tool can cut threads almost entirely to the bottom of the hole, while a
thread chasing tool has several incomplete threads left at the bottom of the
hole.

 

[Spehro Pefhany]

I read in sci.electronics.design that Spehro Pefhany <speffSNIP@interlog


Yes, but specifically making threads with the type of tool described is
'chasing'. You don't need a lead-screw, just a swift hand.

I see what you mean. There's another meaning of "chasing threads", at
least on this side of the pond, to refer to cleaning up existing
threads with a die. The kind of tool you are talking about is a
multi-point tool shaped like a slice taken radially from a tap.


(view from above)

--------------- (rotating work)

- - - - - - - - - <- axis of rotation

---------------

/\/\/\/\
| |
| |
| | <--- fixed chasing tool, free to move <--->



Best regards,
Spehro Pefhany

 

[Brian Whatcott]

....specifically making threads with the type of tool described is
'chasing'. You don't need a lead-screw, just a swift hand.

As it happens, I can agree with John Woodgate.

Though the original writer TALKS about thread-milling (which
surprise, surprise, is done with a rotating mill (looks like a tap)
he actually described a tool with thead profiles cut down
one side. This is a description of a thread-chase, which does not
need to have a progessive cut - they often have threads to full depth
all the way, and moves axially while the work rotates....

Take a look at this URL for a little more on milling threads (with a
ROTATING mill!)

<http://www.sct-usa.com/millhelp.asp>

Brian w

 

[Garrett Mace]

As it happens, I can agree with John Woodgate.

Though the original writer TALKS about thread-milling (which
surprise, surprise, is done with a rotating mill (looks like a tap)
he actually described a tool with thead profiles cut down
one side. This is a description of a thread-chase, which does not
need to have a progessive cut - they often have threads to full depth
all the way, and moves axially while the work rotates....

Take a look at this URL for a little more on milling threads (with a
ROTATING mill!)

<http://www.sct-usa.com/millhelp.asp>

Brian w

A common type of thread-milling tool uses a single carbide insert, which has
a row of cutting teeth. A thread chase, as the name implies, is supposed to
chase the thread for most of the length. A thread chase is most likely to be
used in a lathe, replacing a single point threading tool for quicker
material removal.

 

[Rich Grise]

Why hack into the readout? I think you want to just hack into the cable
and tap into the 2 quadrature signals for each axis. If it has a
connector you could do it with a breakout box.

Well, yeah, that's why I was asking about the signal.

Little to go on? Obviously it's a pair of signals from a relative shaft
encoder. You know, they're 90 deg out of phase, like a sine and cosine.

Well, the sensor is linear, so what are the pulses telling me? Some length
per pulse, I'd surmise - I suppose I should just build the breakout
cable and poke at stuff with a scope or something. Yeah, that's what
hackers do! (the real kind, not the pirate kind, thank you very much. )

Cheers!
Rich

 

[Tom Del Rosso]

In Rich Grise typed:

Well, the sensor is linear, so what are the pulses telling me? Some
length per pulse, I'd surmise - I suppose I should just build the
breakout
cable and poke at stuff with a scope or something. Yeah, that's what
hackers do! (the real kind, not the pirate kind, thank you very much.
)

You have to identify 2 signal pins for each axis. The first few google
hits have waveform diagrams.

http://www.google.com/search?hl=en&ie=UTF-8&oe=UTF-8&q=quadrature+encoder

 

[Rich Grise]

O.K. the mill got there..... pity about the cutter :-(

Speed ( feed rate ) control is very important.

Yes, that's very true. But I've just now started getting an inkling that
the signal I'm looking at might be one pulse per 0.0005". The thought
of that being true makes me drool. I have complete control of the
motor - and today, Jimmy brought in a stepper motor: the 34T2FEHH at
http://www.bodineonline.com/hy-sync.html
so I have about a billion options. And everything's already here -
I don't even need a budget! Maybe a couple of good power devices
or just a driver board - that's the kind of thing it's cheaper to
buy.

My plan for the first demo is to go to some point and zero the
unit, and tell it, "OK, go to 24.3470 and stop." And have it
go, "Whirrrrrr-r-r--r---r----up." and stop dead nuts. I've done
that open loop with steppers, and the feeling when the boss and
about 6 cow-orkers see a stepper do that, which the programmer
before me had going clackety-clackety-clack, when they see and
hear it go "whirrrrrzzzzp.", and they all go, "OOhh, Aaaahh."
there's no feeling that compares to that, except maybe singing
in front of 300 adoring parents in the talent show and getting
thunderous applause.

Cheers!
Rich

 

[Tom Del Rosso]

In Rich Grise typed:

Yes, that's very true. But I've just now started getting an inkling
that the signal I'm looking at might be one pulse per 0.0005".

I'm amazed that a linear encoder could go that small. As Wild Bill just
pointed out, a linear encoder usually has limited resolution. A rotary
encoder on the screw gives more resolution but has backlash.

hear it go "whirrrrrzzzzp.", and they all go, "OOhh, Aaaahh."
there's no feeling that compares to that, except maybe singing
in front of 300 adoring parents in the talent show and getting
thunderous applause.

What did you do, Man of La Mancha?

To hack
the im-pos-si-ble hack...

 

[Mark Zenier]

In Rich Grise typed:

I'm amazed that a linear encoder could go that small. As Wild Bill just
pointed out, a linear encoder usually has limited resolution. A rotary
encoder on the screw gives more resolution but has backlash.

The typical scale is a row of stripes, 500 per inch, (or the metric
equivalent), chrome on quartz glass. Then there's a scale in the sensor
that's a bit different pitch so that light projected through both scales
gives a moire effect. Like the vernier scale on (non-dial) calipers.
And two photosensors, set to that they read the the fringes 1/4 of a
cycle out.

If these sensors are analog, they give you a sine and cosine signal
that can interpolated further. (One system I worked on by a factor of
320, so that the resolution was 1/160000 of an inch. Using a laser
interferometer, accuracy was about 5 times worse than that, depending
on defects on the scales). The resolution limit with machine tools
is more a result of vibration than the limits of the encoder system.

Mark Zenier [email protected] Washington State resident

 

[Rich Grise]

What did you do, Man of La Mancha?

To hack
the im-pos-si-ble hack...

Actually, "My Way." The teacher said that it was an appropriate
choice. ;-)

Cheers!
Rich

 

[David Billington]

And he says he can get a position transducer, and kept talking
about the sine and cosine. Well, a very little digging shows
that the transducer, above, uses "quadrature TTL," which is
tantalizingly little to go on.

Quadrature has been explained by others. TTL is transistor transistor
logic. Basically either 0V or 5V to indicate off or on. Actually off is
from about 0V to 0.8V and on is about 3.8V to 5V IIRC. This makes it
easy to interface to other computer interfaces also using TTL. Some DRO
scale units output sine waves at other voltages which means you would
have to convert the sine waves to something usable. I chose a Newall DRO
for my lathe because it had RS 422 compatible differential quadrature
outputs and I can use a standard 422 converter chip to convert to
quadrature TTL to input into a motion controller.

 

[Tom Del Rosso]

In David Billington typed:

Quadrature has been explained by others. TTL is transistor transistor
logic. Basically either 0V or 5V to indicate off or on. Actually off
is from about 0V to 0.8V
and on is about 3.8V to 5V IIRC.

2.0 to 5 V

 

[Tom Del Rosso]

In Rich Grise typed:

Actually, "My Way." The teacher said that it was an appropriate
choice. ;-)

In the Sid Vicious style I'd imagine.

 

[Jon Grimm]

linear encoders with resolutions of .0002" and finer are very common

we have linear encoders on both of our cnc mills, we think they are
superior, because they measure the actual machine movements, eliminating the
possibility of losses due to backlash

 

[Mark J.]

In news:[email protected] (Jon Grimm):

linear encoders with resolutions of .0002" and finer are very common

we have linear encoders on both of our cnc mills, we think they are
superior, because they measure the actual machine movements,
eliminating the possibility of losses due to backlash

How in the heck do they maintain that kind of linearity over 24-36" travel?

 

[Tom Del Rosso]

In Jon Grimm typed:

linear encoders with resolutions of .0002" and finer are very common

we have linear encoders on both of our cnc mills, we think they are
superior, because they measure the actual machine movements,
eliminating the possibility of losses due to backlash

How do they work? Optically, or are they pots?

 

[Richard Crowley]

"Mark J." wrote ...

How in the heck do they maintain that kind of linearity
over 24-36" travel?

They've been perfecting the process for several decdades?

 

[Mark J.]

In news:[email protected] (Richard Crowley):

"Mark J." wrote ...

They've been perfecting the process for several decdades?

That's not "How"...