开云体育

Thanks all,

It's a little clearer now, the conversion of a vfd to flex-drive
sounds interresting, hope to try it some day.
Thanks

Hugo

--- In CAD_CAM_EDM_DRO@y..., Jon Elson <elson@p...> wrote:

Peter Seddon wrote:

AC servo and DC brushless motors are I believe one and the same

for all

intents and purposes.

An induction motor cannot be described as a servo motor as its low

speed

performance is generally a bit rough. To produce torque you have

to induce

current in the rotor that at very low frequencies and speeds is

not very

efficient. With permanent magnets on the rotor (ie your ac servo

motor) the

rotor flux is permanently established.

No, the newest servo drives are, indeed, using AC induction

motors! The

reason is that an induction
motor can run much hotter than the permanent magnets could stand.

They

are also cheaper, and when
rotor inertia is a factor, the rotor core can be made much lighter

than

a permanent magnet rotor.

Yes, there are tradeoffs, but a motor designed from the ground up

as a

servo motor can perform
much better than a PM brushless motor. I don't know how much worse

a

commercial 60 Hz
induction motor is, but clearly, they make little attempt to keep

the

rotor inertia down.

AC flux-vector drives are needed to control the rotor

magnetization, so

the motor can always be
accelerated instantly when needed. DSP processor chips make this
possible at very reasonable
cost. Basically, you replace the cheap CPU in a standard VFD with

a $30

DSP, add an encoder
counter, and you have an AC flux-vector drive.

Note that ANY servo motor, DC brush, brushless or AC, when

producing

torque at zero speed,
has an efficiency of zero!

Jon

Hi everybody,

I am in the finishing/debugging stages of a 3 axis stepper controller/
G-Code interpreter.

The missing part is how to handle endswitches.

My assumption is that hitting an endswitch is a catastrophic failure,
and it should never happen; therefore I was going to have the
controller freeze all movements, signal the error (a big red LED) and
let the operator shutdown the system, manually back off the tools,
etc. and restart.

I think that trying to have the tool be "smart" and back off the tool
itself, etc. could possibly lead to damaging the machine and or part
being machined.

Does this make sense ? What is usually done, or expected in such a
case ?

Please let me know what you think is the best way to handle
endswitches, and if you have experience with other systems or
software, let me know how it is done.

BTW, this system has the following features:

- communicates with the host computer through a serial RS232 port
(the host could be a palm pilot :-) ), use of hardware (RTS/CTS)
flow control to ensure dataflow integrity.
- embedded basic G-Code interpreter that supports linear and
circular interpolation in all axis
- extensible to up to 8 axis (through an extension board, to be
designed)
- automatically goes to low-current (holding) setting for a given axis
if that axis has not been active for 100 milliseconds
- active current control through integrated chipset, thus very little
wasted power. Running and holding current are configurable through
separate potentiometer, up to 3 amp per axis (although I have not
tested this much power lacking the needed power supply...)
- each axis can be configured as to what the positive or negative
direction is (to accomodate for different machine configuration)
- configurable "rapid transverse speed" (G00 code)
- configuration settings are saved in permanent memory (eeprom) so
they remain between shutdown/reset.
- up to 10'000 steps per seconds (is this overkill ? my steppers stall
after 900 steps/sec...)

Currently supports unipolar steppers, in full-stepping mode. A bipolar
stepper power board will come next.

All feedback much appreciated, thank you.

bruno

--- In CAD_CAM_EDM_DRO@y..., "Tim Goldstein" <timg@k...> wrote:

On a Shoptask you need the largest motor on the Z axis. The

requirements for X & Y are pretty much the same. You hear all the yak
about not being able to machine manually when you have ball screws. I
have ballscrews on my Shoptask and it machines manually just fine...

Tim,
I'm sure your right. When I looked at CNC'ing a lathe I was advised
by a ballscrew manufacturer to select low threads per inch if I
wanted to be able to use the lathe under manual operation. Its seems
pretty obvious is a leadsrews pitch was 1 turn/inch it would be hard
to hold. With 10 turns per inch it won't move.

My Tree mill came equipped with handles and 5tpi ballscrews and was
meant to be used either manual (learn and replay) or CNC. It doesn't
move on its own. Ditto on the Z axis.

Roger

Hi Rick,

Could you possibly describe some of the menu interactions? I like to
run MDI commands, and with logging on, I get a program I can run again.

I was thinking about adding a menu for arcs, and if one uses the 'R'
format of G02/G03, it could be relatively simple! And adding menus for
drill cycles would be a snap as well. Another "menu" I've been wanting
is for milling a simple rectangle. Should be easy.

My MDI has an eight level "history", so a simple rectangle becomes
mostly "up keys", and modifying an axis offset. I like it!

These thoughts may not be what others are thinking of for
"Conversational CNC" (I've never really seen it), so I'd like more
ideas!

Alan KM6VV


dodge1320 wrote:

On a Shoptask you need the largest motor on the Z axis. The requirements for
X & Y are pretty much the same.

You hear all the yak about not being able to machine manually when you have
ball screws. I have ballscrews on my Shoptask and it machines manually just
fine. I am sure that there are ways to get the machine to pull the table,
but it just has not happened in the limited amount of manual work I have
done. With steppers there is considerable resistance to turning the cranks
and when I had steppers on the machine it was actually a PITA to do anything
by hand. Now with the servos the machine is easier to use manually and I
still have not had it grab the work and spin the screw. Makes me wonder how
much of it is armchair musing and how much is actual experience??

Tim
[Denver, CO]
Sherline Products at Deep Discount
www.KTMarketing.com/Sherline

I am considering discontinuing the Mauch/Kulaga DRO printed circuit
board for two reasons. Sales have been slow. Only 35 in 1 year and it
was made for a rapidly disappearing ISA slot.
I have one good board and 3 defective boards left. After the good board
is sold I doubt that I will reorder them unless there is more interest
in them.
Dan

I have a question,about the Indramat motors. About 4 years ago where I worked we had a problem with Indramat motors dieing for "unknown" reasons. Bosch never gave a good answer to us either. Any thoughts, experiences etc?
Thanks,
Bill

Well, I just did what I've been telling myself that I have to stop
doing....spend more money on this recent CNC infatuation.

I just scored (2) Triple Stack Slo-Syn Nema 34 550 oz-in steppers for
$60. So now I have (3) double stack Slo-syn motors (370 oz-in) and
the (2) 550 oz.in and multiple size 23s and 17s.

When my Shoptask arrives, I'd like to CNC both the Shoptask and my
Mill/Drill. At this point I wasn't planning to change either to
ballscrew because I'd really like to maintain the manual control of
the machines, thus I know the Z axis of the Mill/Drill could easily
be controlled by any of my Nema 23 steppers. That leaves the 5 Nema
34 motors to drive the other 5 axes of the two machines.

What is typically the preference for which axes get the higher torque
motors? My understanding is that its usually the Z-axis, but because
of the quill feed configuration of the mill/drill, it won't require a
high torque motor. Is the shoptask similar if I don't convert it to
ballscrew? If I put one of the triple stacks on the Z of the
Shoptask, then I'll have to mix up one of the machines X-Y. Would
the X feed on the Shoptask be the best choice for the larger motor
because its also the lathe (Z) direction?

Thanks!
-Chris

Peter Seddon wrote:

AC servo and DC brushless motors are I believe one and the same for all
intents and purposes.

An induction motor cannot be described as a servo motor as its low speed
performance is generally a bit rough. To produce torque you have to induce
current in the rotor that at very low frequencies and speeds is not very
efficient. With permanent magnets on the rotor (ie your ac servo motor) the
rotor flux is permanently established.

No, the newest servo drives are, indeed, using AC induction motors! The reason is that an induction
motor can run much hotter than the permanent magnets could stand. They are also cheaper, and when
rotor inertia is a factor, the rotor core can be made much lighter than a permanent magnet rotor.

Yes, there are tradeoffs, but a motor designed from the ground up as a servo motor can perform
much better than a PM brushless motor. I don't know how much worse a commercial 60 Hz
induction motor is, but clearly, they make little attempt to keep the rotor inertia down.

AC flux-vector drives are needed to control the rotor magnetization, so the motor can always be
accelerated instantly when needed. DSP processor chips make this possible at very reasonable
cost. Basically, you replace the cheap CPU in a standard VFD with a $30 DSP, add an encoder
counter, and you have an AC flux-vector drive.

Note that ANY servo motor, DC brush, brushless or AC, when producing torque at zero speed,
has an efficiency of zero!

Jon

al5502 wrote:

Did somebody try to get rid of the feed screw all together
and replace it with a hydraulic cylinder?
At the first glance it may present several advantages:
no backlash, big force at low cost.

We go through this every 6 months or so. It looks neat at first glance, and if you
have lots of hydraulic gear around, might even be affordable. Proportional (Moog)
valves are REALLY expensive, like $3000 and up! The hydraulic pumps are expensive,
massively power-hungry, and LOUD! The hazards of a hydraulic leak at 3000 PSI
are very serious. You get oil injected under the skin, and it poisons you. A flexible
line blowing off can also be dangerous. If there is any air in the system, it tends to
go crazy with oscillation. The performance can be iffy, too, with jitter and stick-slip
friction.

I think if you add up all the parts costs, even surplus, you could go out and buy a complete
ballscrew retrofit and servo drives new!

Jon

gittt2000 wrote:

--- In CAD_CAM_EDM_DRO@y..., Jon Elson <elson@p...> wrote:

DC Brushless motors are really misnamed. THEY are truly

synchronous

motors. AC servo motors
are VERY similar to ordinary AC induction motors, and they ARE asynchronous, as they have to be
for the induction principle to excite the rotor.

I've been confused between DC Brushless and AC synchronous Servo motors and thought I'd sorted this out, but now I'm confused again.
The MAC AC Servo motors by Indramat (now Bosch Rexroth) are described as synchronous but they are not similar to ordinary ac induction motors. They have a wound stator, electronic commutation, and a rotor comprised of 6 (or another multiple of 3 presumably) permanent magnets, rather than the squirrel cage of conductors which form the normal induction motor rotor.

Well, this is the problem when an industry uses incorrect terminology for so long that nobody
recognizes it as being wrong. When somebody breaks the 'rules' and uses CORRECT terminology,
people are confused! Bosch is using the correct terminology. As you describe their motors,
they are truly synchronous permanent magnet motors, which most people call "DC brushless".

And, now, I've caused confusion by overgeneralizing. If it has permanent magnets or some other
means of being synchronous, then it is not an induction motor. If it is asynchronous, and requires
slip between the stator field and rotor to magnetize the rotor, then it is obviously an induction
motor.

As far as I can see there is no induction involved - the stator produces a rotating magnetic field and the magnetic rotor follows it. What makes it synchronous is the commutation which signals the servo drive to supply the correctly phased stator drive.

No, what makes it synchronous, BY DEFINITION, is that the stator and rotor fields follow each
other at all times. An induction motor always has some 'slip' between these fields, and that is
what magnetizes the rotor.

The commutation is required to make the stator poles rotate. Very similar schemes are used in
BOTH synchronous and asynchronous motors. The difference is that in an induction motor, the
slip has to be accounted for. In a torquing application, where the motor is delivering torque at
zero speed, a permanent magnet motor would have the stator fields held constant. But, an
induction motor in the same situation would require the fields to constantly rotate very
slowly, about 5 - 25 RPM.

Jon

At 07:30 PM 11/5/02 -0000, you wrote:

The missing part is how to handle endswitches.

Usually the endswitch contact disables motion in that direction but leaves
the remaining direction and all other axes active. That way you can back it
off and remediate. You may choose to include an alarm, to notify the
operator of the condition.

Regards, Hoyt McKagen

To prevent virus propogation, don't put this addy in your book
Belfab CNC -
US Best MC -
Camping/Caving -
Two-Wheel-Tech List -
Never trust a fat man

Depending on what you want to do might be of interest to you.

From their page:

This is a project to create and distribute a PLC-like program for Linux (PLC = Programmable Logic Controller), licensed under the GNU GPL.

We take advantage of the fact that we have an underlying operating system and use its features to make the MatPLC modular. One module could be a plc5 emulator, that executes plc5 code. Another module, already written, is a PID loop. A different module handles I/O.

Currently, we are in rather early stages: we have a simple IL-style language for logic modules (or C can be used), a signal-processing module which includes a PID loop, an I/O module or two and some simple HMI modules.

<snip>

I was also considering the ability to
index my spindle rotation for 5-axis (don't ask).

Don't ask? Now you have us on the edge of our seats!
What are planning to do with that 5th axis? We won't tell anyone :-)


Wayne

At 11:18 AM 11/5/02 -0700, you wrote:

by hand. Now with the servos the machine is easier to use manually and I
still have not had it grab the work and spin the screw. Makes me wonder how
much of it is armchair musing and how much is actual experience??

Chevailer CNCs with ball screws give you a manual option that completely
disables the drives. And though I said it before, the only grab and slam I
ever had was on a manual machine with Acmes. The cure is always only two
words: table locks.

Regards, Hoyt McKagen

To prevent virus propogation, don't put this addy in your book
Belfab CNC -
US Best MC -
Camping/Caving -
Two-Wheel-Tech List -
Never trust a fat man

At 10:35 AM 11/5/02 -0600, you wrote:

If we select a quadrature encoder or used a stepper motor with
gearing/belting sufficient to get that resolution, do you think that the
greater accuracy as we approach the center would still cause problems?

Of course not. But don't you then use up more CPU time calculating? IE,
either you compute all the arcs regardless of size or you compute fewer
arcs depending on the smaller radii on some parts, which latter involves an
additional function. But anyway, the error on polar machine is always going
to be greater at extremes of the machine, whereas with Cartesian machine it
isn't

Much the same applies to hexapods. The error is between two and five times
the actuator error, compared to the one times actuator error of Cartesian
machine.

Regards, Hoyt McKagen

To prevent virus propogation, don't put this addy in your book
Belfab CNC -
US Best MC -
Camping/Caving -
Two-Wheel-Tech List -
Never trust a fat man

Hi Hoyt

I can see the "inherent differences in accuracy" issue but let me pose a
question based on the assumption that we build Fred's 12" square
prototype. In order to do this we would need a 17 inch platter.

My math may be all squirreled up so correct this. The distance out to a
corner of the 12 inch square should be be about 8.5 inches. That works
out to about 53.41 inches around the circumference at that size. (zero
diameter tool for outside cutting) Assume that we want 0.0005 for the
smallest move at a corner of the square. There should be about 106814
of those arcs.

If we select a quadrature encoder or used a stepper motor with
gearing/belting sufficient to get that resolution, do you think that the
greater accuracy as we approach the center would still cause problems?

Ray

Bill/Group,

What do you mean by conversational CNC?

Conversational CAM where you're prompted for parameters?

Or, conversational CNC controls, where the programming is done at the
machine?

The former is nothing new.

Conversational CNC controls are something else. Two of my controls
are advertised as "conversational". Basically, as I see it, they're
just controls with very powerful canned cycles. The two controls
work slightly differently. One of them gives you a screen full of
possible parameters for you to choose based on which options of the
particular canned cycle you're utilizing. The user is not prompted
for parameter values. The other control gives a similiar screen with
a graphical representation of the canned cycle and prompts the user
for parameter values.

Both controls will dry run the program graphically to verify it.
Neither control checks for out of range values or omission of
required parameters until dry running.

It's the power of the canned cycles that makes these controls
special. They can accomplish in a couple of lines of easily
programmed, understandable and shop-floor-modifiable code what a CAM
system might require 100's of lines using plain vanilla Gcodes.

Example: a part having a rectangular pocket with radiused corners
oriented 30 degrees off the Y axis with peck milling to go to full
pocket depth and finish pass....five lines of code, first line to
load tool and turn spindle on, second line rotate machine axis, third
to position tool at start of pocket, fourth line to mill pocket, last
line M30. How many lines would that take using G00, G01, G02, G03
and how easily could you modify pocket size, corner radius or depth?

Using the conversational controls we do about 98% of our production
part programming while standing in front of the machines without need
for CAM.

Doug



--- In CAD_CAM_EDM_DRO@y..., wanliker@a... wrote:

When this list was first started there was quite a lot about

Conversational

CNC, but nothing since, for a long time.

Would someone undertake explaining to me and the list, what, and

why

Conversational CNC is desirable, its advantages and disadvantages.

Also would any of the programmers on the list like to tackle a

project like

that? If so I would set up a group for you to handle the technical
conversations. And I would be happy ti take care of any overhead

for the

list. Hopefully this would add another option to our

members????????????

Thanks,
bill
List Mom

Doug more details please.........................Give a bit more detail on
how everything is set up and how the machine knows where to do the canned
cycle. There has been very little detail given on these programs. Be our
teacher on this aspect.
thanks
bill

--- In CAD_CAM_EDM_DRO@y..., "Peter Seddon" <peter@s...> wrote:

AC servo and DC brushless motors are I believe one and the same

for all

intents and purposes.

That's the only conclusion I can come to.

With permanent magnets on the rotor (ie your ac servo motor) the
rotor flux is permanently established.

Yes, and being smaller diamter, lighter and longer means it's much
lower inertia. Also means that if you drive the input shaft with no
power to the stator, then it generates, so making it easy to adjust
the commutation timing without phase shift inducing currents and
losses to allow for.