ITS ALIVE
 

Ok... I still have a lot of work to do, but the Hurco KMB1 I picked up in
November 2010 finally came to life yesterday. Ok... one finger twitched
when stimulated with electricity. LOL.

I actually powered up the computer, motion control interface, breakout board
(with some on board functions also), and one of the servo drivers yesterday.
With a freshly installed 1000 line quadrature encoder on it I spun one of
the yet to be re-installed servo motors back and forth at speeds upto 1250
RPM with good position repeatability using a short g-code test program
running on the control computer. If the proximity sensors installed on the
mill are all still good and I don't have any oil line leaks I may actually
cut some parts with it in the next month or so.

1250 RPM is the speed necessary for the original factory specified rapid
speed of 250 IPM. I actually free spun the motors at 2000 RPM when testing
with direct uncontrolled power. With a 20% derating as reccommended by
Gecko that would mean a top rapid speed of 1600 RPM (upto 320 IPM) with this
setup. I'm going to set my programmed rapid speed down to 150 IPM until all
things are tested and working then bump it up with a full 500 lb load on the
table and see how it performs before settling on my final top speed. I'll
keep current low and allow for momentary overload rather than going for peak
current. If I get current or position faults during testing I'll drop speed
and/or acceleration back until I don't.

I still have a lot of work to do though before I start moving the table
under system controlled power.

ITS ALIVE
 

On Fri, 24 Aug 2012 12:58:11 -0700, Bob La Londe wrote:

Ok... I still have a lot of work to do, but the Hurco KMB1 I picked up
in November 2010 finally came to life yesterday. Ok... one finger
twitched when stimulated with electricity. LOL.

(...)

YeeeeHa!



If I get current or position faults during testing I'll drop speed
and/or acceleration back until I don't.

Good on ya!

Advise you leave headroom between the maximum error-free
acceleration and your final 'top speed' declaration.

--Winston

ITS ALIVE
 

"Winston" wrote in message
...
On Fri, 24 Aug 2012 12:58:11 -0700, Bob La Londe wrote:

Ok... I still have a lot of work to do, but the Hurco KMB1 I picked up
in November 2010 finally came to life yesterday. Ok... one finger
twitched when stimulated with electricity. LOL.

(...)

YeeeeHa!



If I get current or position faults during testing I'll drop speed
and/or acceleration back until I don't.

Good on ya!

Advise you leave headroom between the maximum error-free
acceleration and your final 'top speed' declaration.

Yeah I was originally planning on just keeping it at 150, but I have a
little Chinese CNC router I reconfigured recently that I ran all the way up
to 600 IPM before it stalled. Then it was only after about 30 passes. I
dropped it to 300 and left it there. Anyway, it kind of spoiled me for
speed. LOL. Since the no load full speed of the servo motors on the mill
is only 2000 RPM I'll never ever be able to run it as fast as I tested that
router. Kind of amazing really when I think about it. The stepper motors
machine is faster and always will be. LOL.

I think though as long as it catches up by the end of a rapid it should be
fine. Cuts will always be much slower. Not much I can feed very fast with
only a 3600 RPM spindle. I think even so I may stay at 150 for a long time.
I want to make sure braking, limits, oiler, etc all work perfectly every
time before I try to pick up any speed. Also, I want to make sure I have
soft limits programmed and a slow down zone close to limit switches etc.
I'll also want to establish a startup homing routine so I never actually hit
the limits except when homing... and of course test EVERYTHING.

I know that no load testing isn't the same as pushing around a several
hundred pound table, but the only time I saw the in position LED blink was
on full speed reversals. That's not something you would ever have happen
under normal machine operations when doing rapids. Never saw the fault
light come on at all.

Today I unplugged everything and started wiring in stuff in its "permanent"
configuration. I still want to add a drain circuit to my DC caps. The
servo amps drain them a lot quicker than just idling, but still. I would
never want to accidentally put a sweaty arm across the top of the big caps.
I'm thinking maybe just a switched resister that shorts the caps anytime the
door is opened.

I am considering cutting a hole in the door and putting a in a wire backed
plexiglas or wire core safety glass panel in there so I can button it up and
engage all the safeties and still be able to see all the LEDs.

ITS ALIVE
 

On Fri, 24 Aug 2012 21:28:33 -0700, Bob La Londe wrote:

I know that no load testing isn't the same as pushing around a several
hundred pound table, but the only time I saw the in position LED blink
was on full speed reversals. That's not something you would ever have
happen under normal machine operations when doing rapids. Never saw the
fault light come on at all.

Yeah.
I put some teeny tiny NEMA 34 steppers on my mill.
They were the most I could afford at the time.
I tested acceleration by physically pushing against the travel
of each axis until I got an error. Crank down the acceleration
and re-test etc etc. Eventually I got to a point that guaranteed
travel even when I pushed against each axis with all my available
guts. I set acceleration at 90% of that value and called it done.

Fun stuff!

--Winston

ITS ALIVE
 

On Fri, 24 Aug 2012 21:28:33 -0700, "Bob La Londe"
wrote:

"Winston" wrote in message
...
On Fri, 24 Aug 2012 12:58:11 -0700, Bob La Londe wrote:

Ok... I still have a lot of work to do, but the Hurco KMB1 I picked up
in November 2010 finally came to life yesterday. Ok... one finger
twitched when stimulated with electricity. LOL.

(...)

YeeeeHa!



If I get current or position faults during testing I'll drop speed
and/or acceleration back until I don't.

Good on ya!

Advise you leave headroom between the maximum error-free
acceleration and your final 'top speed' declaration.

Yeah I was originally planning on just keeping it at 150, but I have a
little Chinese CNC router I reconfigured recently that I ran all the way up
to 600 IPM before it stalled. Then it was only after about 30 passes. I
dropped it to 300 and left it there. Anyway, it kind of spoiled me for
speed. LOL. Since the no load full speed of the servo motors on the mill
is only 2000 RPM I'll never ever be able to run it as fast as I tested that
router. Kind of amazing really when I think about it. The stepper motors
machine is faster and always will be. LOL.

I think though as long as it catches up by the end of a rapid it should be
fine. Cuts will always be much slower. Not much I can feed very fast with
only a 3600 RPM spindle. I think even so I may stay at 150 for a long time.
I want to make sure braking, limits, oiler, etc all work perfectly every
time before I try to pick up any speed. Also, I want to make sure I have
soft limits programmed and a slow down zone close to limit switches etc.
I'll also want to establish a startup homing routine so I never actually hit
the limits except when homing... and of course test EVERYTHING.

I know that no load testing isn't the same as pushing around a several
hundred pound table, but the only time I saw the in position LED blink was
on full speed reversals. That's not something you would ever have happen
under normal machine operations when doing rapids. Never saw the fault
light come on at all.

Today I unplugged everything and started wiring in stuff in its "permanent"
configuration. I still want to add a drain circuit to my DC caps. The
servo amps drain them a lot quicker than just idling, but still. I would
never want to accidentally put a sweaty arm across the top of the big caps.
I'm thinking maybe just a switched resister that shorts the caps anytime the
door is opened.

Why a switched resistor? A simple 50k-100k 15-20 watt resistor will
bleed off a cap in relatively quick fashion. The OmniTurn CNC lathes I
use keep a resistor mounted across the caps just for that reason. Just
use a cheap wirewound ceramic. You are working with about 100 volts
DC..its not rocket science.

I am considering cutting a hole in the door and putting a in a wire backed
plexiglas or wire core safety glass panel in there so I can button it up and
engage all the safeties and still be able to see all the LEDs.



One bleeding-heart type asked me in a recent interview if I did not
agree that "violence begets violence." I told him that it is my
earnest endeavor to see that it does. I would like very much to ensure
- and in some cases I have - that any man who offers violence to his
fellow citizen begets a whole lot more in return than he can enjoy.

- Jeff Cooper

ITS ALIVE
 

Awesome!

i

On 2012-08-24, Bob La Londe wrote:
Ok... I still have a lot of work to do, but the Hurco KMB1 I picked up in
November 2010 finally came to life yesterday. Ok... one finger twitched
when stimulated with electricity. LOL.

I actually powered up the computer, motion control interface, breakout board
(with some on board functions also), and one of the servo drivers yesterday.
With a freshly installed 1000 line quadrature encoder on it I spun one of
the yet to be re-installed servo motors back and forth at speeds upto 1250
RPM with good position repeatability using a short g-code test program
running on the control computer. If the proximity sensors installed on the
mill are all still good and I don't have any oil line leaks I may actually
cut some parts with it in the next month or so.

1250 RPM is the speed necessary for the original factory specified rapid
speed of 250 IPM. I actually free spun the motors at 2000 RPM when testing
with direct uncontrolled power. With a 20% derating as reccommended by
Gecko that would mean a top rapid speed of 1600 RPM (upto 320 IPM) with this
setup. I'm going to set my programmed rapid speed down to 150 IPM until all
things are tested and working then bump it up with a full 500 lb load on the
table and see how it performs before settling on my final top speed. I'll
keep current low and allow for momentary overload rather than going for peak
current. If I get current or position faults during testing I'll drop speed
and/or acceleration back until I don't.

I still have a lot of work to do though before I start moving the table
under system controlled power.

ITS ALIVE
 

On Friday, August 24, 2012 10:00:37 PM UTC-7, Winston wrote:
On Fri, 24 Aug 2012 21:28:33 -0700, Bob La Londe wrote:



I know that no load testing isn't the same as pushing around a several

hundred pound table, but the only time I saw the in position LED blink

was on full speed reversals. That's not something you would ever have

happen under normal machine operations when doing rapids. Never saw the

fault light come on at all.



Yeah.

I put some teeny tiny NEMA 34 steppers on my mill.

They were the most I could afford at the time.

I tested acceleration by physically pushing against the travel

of each axis until I got an error. Crank down the acceleration

and re-test etc etc. Eventually I got to a point that guaranteed

travel even when I pushed against each axis with all my available

guts. I set acceleration at 90% of that value and called it done.



Fun stuff!



--Winston

Funny, I basically tested the Taig, and the MAX NC that way. The Chinese router I didn't. Just tested it for position error with an indicator.

ITS ALIVE
 

Ok... I was doing some wiring last night. After testing the servo motor I
disconnected it and started doing my "permanent" wiring. While I was
standing there admiring my work I was thinking about the various inputs. A
smoothstepper has the functionality of all the pins of two parallel ports,
but it also has a set of inputs normally used for encoders that can be
addressed like any other input as PORT 3. That gives me 15 inputs.

Since Mach doesn't tell you which limit was hit when you hit a limit switch
I figured I would wire them all up to the same input. I will also use my
limits for homing atleast to start.

Add in probing inputs and a few other misc stuff for various trouble
conditions spindel speed etc, and I would normally have only a couple inputs
left over. With those port 3 inputs however I have enough to hook simple
switches up as jog switches for manual control other than the keypad.

At first I was thinking of some cheap Radio Shack momentary push buttons,
but then I started thinking of other things I might be able to adapt. Maybe
an old Atari joystick or something else. Then I started looking on Ebay for
arcade buttons. I was almost ready to order some. One outfit has 2 HAPP
joysticks and 14 buttons for about $40.

This morning I remember something I already had. Years ago we bought an
arcade style interface to use with a PS2. I was never a serious gamer and my
son didn't like the way it played so it got put in storage.

Its perfect. No its not a pendant, but I can make a little box shelf, mount
it on the head of the mill, or hang it off one side of my keyboard tray. The
exact location can be determined later. I suppose I can even make up a
pendant style box I can hang on a hook out of the way when not in use.

Initial cost for testing the idea. Nothing.

Maybe I can add a pinball table and a tilt swinger to the machine. LOL.

ITS ALIVE
 

"Gunner" wrote in message
...
On Fri, 24 Aug 2012 21:28:33 -0700, "Bob La Londe"
wrote:

"Winston" wrote in message
...
On Fri, 24 Aug 2012 12:58:11 -0700, Bob La Londe wrote:

Ok... I still have a lot of work to do, but the Hurco KMB1 I picked up
in November 2010 finally came to life yesterday. Ok... one finger
twitched when stimulated with electricity. LOL.

(...)

YeeeeHa!



If I get current or position faults during testing I'll drop speed
and/or acceleration back until I don't.

Good on ya!

Advise you leave headroom between the maximum error-free
acceleration and your final 'top speed' declaration.

Yeah I was originally planning on just keeping it at 150, but I have a
little Chinese CNC router I reconfigured recently that I ran all the way
up
to 600 IPM before it stalled. Then it was only after about 30 passes. I
dropped it to 300 and left it there. Anyway, it kind of spoiled me for
speed. LOL. Since the no load full speed of the servo motors on the mill
is only 2000 RPM I'll never ever be able to run it as fast as I tested
that
router. Kind of amazing really when I think about it. The stepper motors
machine is faster and always will be. LOL.

I think though as long as it catches up by the end of a rapid it should be
fine. Cuts will always be much slower. Not much I can feed very fast
with
only a 3600 RPM spindle. I think even so I may stay at 150 for a long
time.
I want to make sure braking, limits, oiler, etc all work perfectly every
time before I try to pick up any speed. Also, I want to make sure I have
soft limits programmed and a slow down zone close to limit switches etc.
I'll also want to establish a startup homing routine so I never actually
hit
the limits except when homing... and of course test EVERYTHING.

I know that no load testing isn't the same as pushing around a several
hundred pound table, but the only time I saw the in position LED blink was
on full speed reversals. That's not something you would ever have happen
under normal machine operations when doing rapids. Never saw the fault
light come on at all.

Today I unplugged everything and started wiring in stuff in its
"permanent"
configuration. I still want to add a drain circuit to my DC caps. The
servo amps drain them a lot quicker than just idling, but still. I would
never want to accidentally put a sweaty arm across the top of the big
caps.
I'm thinking maybe just a switched resister that shorts the caps anytime
the
door is opened.

Why a switched resistor? A simple 50k-100k 15-20 watt resistor will
bleed off a cap in relatively quick fashion. The OmniTurn CNC lathes I
use keep a resistor mounted across the caps just for that reason. Just
use a cheap wirewound ceramic. You are working with about 100 volts
DC..its not rocket science.

I was thinking of a lower value resistor to drain the CAPs really fast,
although if I put the window in the cabinet door its not as big of a deal
because I'll be able to see the LEDs and just not open it until they all go
out.



I am considering cutting a hole in the door and putting a in a wire backed
plexiglas or wire core safety glass panel in there so I can button it up
and
engage all the safeties and still be able to see all the LEDs.



One bleeding-heart type asked me in a recent interview if I did not
agree that "violence begets violence." I told him that it is my
earnest endeavor to see that it does. I would like very much to ensure
- and in some cases I have - that any man who offers violence to his
fellow citizen begets a whole lot more in return than he can enjoy.

- Jeff Cooper

ITS ALIVE
 

On Sat, 25 Aug 2012 15:17:31 -0700, Bob La Londe wrote:


(...)

Initial cost for testing the idea. Nothing.

Maybe I can add a pinball table and a tilt swinger to the machine. LOL.

Go Bob Go!

--Winston

ITS ALIVE
 

On Sat, 25 Aug 2012 15:19:48 -0700, "Bob La Londe"
wrote:


Why a switched resistor? A simple 50k-100k 15-20 watt resistor will
bleed off a cap in relatively quick fashion. The OmniTurn CNC lathes I
use keep a resistor mounted across the caps just for that reason. Just
use a cheap wirewound ceramic. You are working with about 100 volts
DC..its not rocket science.

I was thinking of a lower value resistor to drain the CAPs really fast,
although if I put the window in the cabinet door its not as big of a deal
because I'll be able to see the LEDs and just not open it until they all go
out.



My suggestion will kill the cap in under 10 seconds.

Gunner

One bleeding-heart type asked me in a recent interview if I did not
agree that "violence begets violence." I told him that it is my
earnest endeavor to see that it does. I would like very much to ensure
- and in some cases I have - that any man who offers violence to his
fellow citizen begets a whole lot more in return than he can enjoy.

- Jeff Cooper

ITS ALIVE
 

On Fri, 24 Aug 2012 12:58:11 -0700, "Bob La Londe"
wrote:

Ok... I still have a lot of work to do, but the Hurco KMB1 I picked up in
November 2010 finally came to life yesterday. Ok... one finger twitched
when stimulated with electricity. LOL.

I actually powered up the computer, motion control interface, breakout board
(with some on board functions also), and one of the servo drivers yesterday.
With a freshly installed 1000 line quadrature encoder on it I spun one of
the yet to be re-installed servo motors back and forth at speeds upto 1250
RPM with good position repeatability using a short g-code test program
running on the control computer. If the proximity sensors installed on the
mill are all still good and I don't have any oil line leaks I may actually
cut some parts with it in the next month or so.

1250 RPM is the speed necessary for the original factory specified rapid
speed of 250 IPM. I actually free spun the motors at 2000 RPM when testing
with direct uncontrolled power. With a 20% derating as reccommended by
Gecko that would mean a top rapid speed of 1600 RPM (upto 320 IPM) with this
setup. I'm going to set my programmed rapid speed down to 150 IPM until all
things are tested and working then bump it up with a full 500 lb load on the
table and see how it performs before settling on my final top speed. I'll
keep current low and allow for momentary overload rather than going for peak
current. If I get current or position faults during testing I'll drop speed
and/or acceleration back until I don't.

I still have a lot of work to do though before I start moving the table
under system controlled power.





Greetings Bob,
Please let us know how well the Gecko servo amps work. I have three of
them and have had good luck but I haven't yet really stressed them.
Thanks,
Eric

ITS ALIVE
 

wrote in message
...
On Fri, 24 Aug 2012 12:58:11 -0700, "Bob La Londe"
wrote:

Ok... I still have a lot of work to do, but the Hurco KMB1 I picked up in
November 2010 finally came to life yesterday. Ok... one finger twitched
when stimulated with electricity. LOL.

I actually powered up the computer, motion control interface, breakout
board
(with some on board functions also), and one of the servo drivers
yesterday.
With a freshly installed 1000 line quadrature encoder on it I spun one of
the yet to be re-installed servo motors back and forth at speeds upto 1250
RPM with good position repeatability using a short g-code test program
running on the control computer. If the proximity sensors installed on the
mill are all still good and I don't have any oil line leaks I may actually
cut some parts with it in the next month or so.

1250 RPM is the speed necessary for the original factory specified rapid
speed of 250 IPM. I actually free spun the motors at 2000 RPM when
testing
with direct uncontrolled power. With a 20% derating as reccommended by
Gecko that would mean a top rapid speed of 1600 RPM (upto 320 IPM) with
this
setup. I'm going to set my programmed rapid speed down to 150 IPM until
all
things are tested and working then bump it up with a full 500 lb load on
the
table and see how it performs before settling on my final top speed. I'll
keep current low and allow for momentary overload rather than going for
peak
current. If I get current or position faults during testing I'll drop
speed
and/or acceleration back until I don't.

I still have a lot of work to do though before I start moving the table
under system controlled power.





Greetings Bob,
Please let us know how well the Gecko servo amps work. I have three of
them and have had good luck but I haven't yet really stressed them.
Thanks,
Eric

In general I do not plan to work this machine to its full cutting load with
a full 500 lbs on the table. I'll probably set my current limit for about 9
or 10 amps, but set the switches to allow for momentary max current. I will
have around 160 lbs of vises on the table all of the time if my planned use
of the machine remains the same. (4 screwless vises at 39 lbs 10.2ozs
(apx))

I'll try to keep you up to date. If you are interested I do have a build
thread going on CNC Zone he
http://www.cnczone.com/forums/hurco/...onversion.html

The latest post shows pictures of the game controller I plan to gut out and
use for manual inputs.