Re: [Emc-users] machine opinion please

[Brent Loschen]

How about something like this 
Gene?  
My LCNC conversion to one of these was a piece of cake, even for this 
LCNC novice!  A little more spend up front than the one you found, but 
probably long run cheaper since it's already CNC'd with quality lead 
screws and powerful servo drives.  This one looks to be in great shape.  
Variable speed (split sheave), Hi/Low spindle.  It does need 220/3 phase 
power, which for me required a $350 rotary phase converter.



On 2/6/2021 5:33 PM, Gene Heskett wrote:

On Saturday 06 February 2021 19:03:27 andy pugh wrote:


On Sat, 6 Feb 2021 at 22:52, Gene Heskett  wrote:



The column does look to be of adequate size, unlike most.

I still think you would possibly be better-served by a used industrial
machine.

https://www.ebay.com/itm/264882512556

A Hardinge would be nice, but how hard is it to put motors on?

Is cute, and has a horizontal spindle option too.

But this is lovely, and has tooling:
https://www.ebay.com/itm/353246842626


Yes, but is 30" deeper than I have room for.


Neither is a trivial CNC conversion, though.


Cheers, Gene Heskett



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] LinuxCNC with Robot Arm Setup

[Brent Loschen]

Nothing on MovIt, but the following recent article in Servo Magazine 
provides a cookbook for building a ROS based system that addresses 
several of the features you mention (at least conceptually), out of a 
Roomba and an Xbox Kinect for vision.  I appreciate examples like this 
that steps one through an otherwise complicated install procedure 
allowing you to get something working with minimal time and mental 
investment.  Once I get the overall concepts down, and some hands on 
skills,  it's much easier to dig deeper into specific documentation 
rather than feeling the need to try to understand it all before I can 
even begin.


https://www.servomagazine.com/magazine/article/a-self-guiding-robot-build-using-ros-and-a-single-board-computer

Brent

On 11/17/2021 7:43 PM, Ian Charnas wrote:

Chris, many thanks for the reply.

I am indeed planning on using computer vision to set targets for the robot
arm to move to. Thanks for sharing the link to MoveIt. One solution could
be to have my computer vision code send the waypoints I want to MoveIt,
which would plan the path, and then send the path to LinuxCNC.

In this scenario, I'm imagining I'd run the computer vision code and MoveIt
on a higher end workstation with a GPU (I can write gpu-accelerated
computer vision code), and then send the results perhaps over gigabit
ethernet using UDP (since it's faster than RS485 or CAN) to a separate
computer running LinuxCNC in real time. The second computer doesn't need to
be as powerful in terms of CPU and GPU which is nice.

I had no idea about the learning curve on MoveIt. I looked around and found
that the authors just published a nice set of video tutorials
 on MoveIt
2 so I'll start there.

I really appreciate the kind and helpful responses here, y'all are great!
Ian

On Wed, Nov 17, 2021 at 8:54 PM Chris Albertson 
wrote:


On Wed, Nov 17, 2021 at 3:07 PM theman whosoldtheworld <
bleachk...@gmail.com>
wrote:


wow movelt . and there are some ethercat connection on movelt?

thanks a

lot


As you plow through the stacks of documentation, you find that MoveIt
outputs a "joint trajectory"  Which is a set of messages with (time, x, x',
x'')  The hardware interface is up to the user.  MoveIt  stops short of the
device drivers.

This is why I said a hybrid with LCNC might be interesting. LCNCis goodat
low levelmotor control andhas the Mesa FPGA ecosystem that work very well.

MoveIt is complex and has a really steep learning curve.  It is built on
top of ROS and ROS has a steep learning curve.   I'd guess the stack is a
couple orders of magnitude more complex than LCNC.  It all runs on Linux
and if you are not a Linux software developer then you have ever more to
learn.  It takes most new people about a year to come up to speed.A
Movit-based 6-DOF robot arm is a tough introduction.  You are best off
starting with basic ROS tutorials and making system that can print "Hello
World" to the screen.

Also note the current transition from ROS-1 to ROS-2.   New projects really
should be using ROS-2. So check that you are reading the correct
tutorials.  Many are not labeled 1 or 2 because there was no need back when
only ROS-1 existed.So look for explicit labeling as "2".




Il giorno mer 17 nov 2021 alle ore 18:22 John Dammeyer <
jo...@autoartisans.com> ha scritto:


Very cool.  Thanks for the link.
John


-Original Message-
From: Chris Albertson [mailto:albertson.ch...@gmail.com]
Sent: November-17-21 8:45 AM
To: Enhanced Machine Controller (EMC)
Subject: Re: [Emc-users] LinuxCNC with Robot Arm Setup

If you are taking the right approach or not depends on the tasks you

want

this robot arm to reform.  Are they entirely scripted such that you

can

write G-code to control the arm?   Or is this arm using a camera and

vision

system to pick up randomly placed objects?

If the former, LCNC will work fine and self-collision is a
non-issue because you programmed the G-code so as not to let that

happen.

But if the arm's path through space is determined in real-time and is

not

pre-programmed, LCNC is not the best tool, and you should be looking

at

"MoveIt". https://moveit.ros.org/

But I think it is possible to build a hybrid system that uses BOTH

software

systems.  There is a part of LCNC that does motion control and has
programmatic API and this could be used to build an arm controller.

MoveIt

then computes a trajectory.

For LCNC experts not familiar with MoveIt...   MoveIt outputs a

series

of

"target points". for each motor.  A joint has the following

information

- Time this target point shall be reached
- The location of the point
- The velocity
- The acceleration

Question.  Given that you have a user space process running that has

that

above information how to best pass this to LCNC so that LCNC can

output

the

stepper pluses or whatever is needed to drive the hardware.

On Tue, Nov 16, 2021 at 9:12 PM Ia

Re: [Emc-users] G-Code issue with IJ

[Brent Loschen]

 This entire thread has been very interesting to me.  I've learned a 
lot about arc moves and creating spirals, but I'm curious why the large 
number of short G1 moves (generated by PathPilot) is "better" than fewer 
G2/G3 moves?  Seems like a nearly constant feedrate/chipload (similar to 
adaptive clearing) would be better than thousands of 
start/stops/accel/decel - what am I missing?


Brent

On 9/24/2019 2:45 PM, Ken Strauss wrote:

I decided to see what Tormach generates with their conversational programming
in PathPilot (LinuxCNC pretty face). I don't recall the original parameters so
I requested a circular pocket, 0.5 deep, 1.0 diameter, at 0,0, 0.5 DoC,
1/4-inch cutter (what was in the spindle). It spirals down in a 1/2 pocket to
full depth and then does a spiral out. I'm happy to try different parameters.
(Mill - Circular Pocket G-code generated: Tue Sep 24 16:36:41 2019 )

... cut ...


(Spiral)
F 15.0 (Arc Feed, inches/minute)
G1 X 0.1250 Y 0.
F 15.0 (Arc Feed, inches/minute)
G1 X 0.1252 Y 0.0110
F 15.1 (Arc Feed, inches/minute)
G1 X 0.1245 Y 0.0219
F 15.1 (Arc Feed, inches/minute)
G1 X 0.1228 Y 0.0329
F 15.2 (Arc Feed, inches/minute)
G1 X 0.1201 Y 0.0437
... cut ...
(- End of Circular Pocket -)

M30 (End of Program)




___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] misters for really small setups?

[Brent Loschen]

Gene - rather than a pressurized Coke bottle/regulator, have you 
considered using a small peristaltic pump?  I just looked on Ebay and 
found  several for under $20-25. This one 
 
($12) fastens to a NEMA 23 stepper.  (unclear how the motor shaft 
attaches to the pump)  Easy enough to build a stand alone speed 
controller for it, but if you want to get fancy, you could even control 
it with LCNC and have very precise, automated flow control based on 
spindle RPM/feedrate/stock/??.  This 
one 
includes 
a stepper motor.


I used one like this 
for 
my mister.  Very precise flow control from .6  to  600 mL/hour.
Another nice one 
, 
but missing a 24v power supply.



On 10/1/2019 2:01 PM, Gene Heskett wrote:

On Tuesday 01 October 2019 12:05:52 Les Newell wrote:


A while back we discussed misters and the mods I did to a mister. The
jet is 2mm with a pretty small annular ring around it. That sounds
pretty much like what you want.

Precisely. I had one made on the old hf mill, but it kept running into
fixture bolts and eventually destroyed itself.

The machining work involved is just drilling one 2mm hole in a lathe
and cutting the 2mm tube to length. It needs pressurized coolant but
your coke bottle setup would do the trick.


Generally that pressure in the bottle is just enough to lift the liquid
to the top of the gantry. Max of say 13" of water, or about 1/2 psi.
Difficlt to set with a $10 regulator. After that, the siphon effect
keeps it flowing rather gently. The bottle is, unless  its plumb full,
below the nozzle, so flow stops when the air solenoid is turned off. The
bottle sits on a small sheet metal shelf about an inch above the bottom
of the gantry riser. Turned off, its half an hour back draining into the
coke bottle to empty the plumbing.  And about a minute to refill when
the mist button is enabled again. The air valve vents when off, taking
pressure off the whole thing in one grand thup.

I can't remember if there was a time when the mist button was active
while running a program, but it would be handier than bottled beer,
hint, hint... OTOH, putting the M7 M9 around the active cut in the gcode
works well, and reduces the AC duty cycle a just noticable bit. Makes me
do it right that way ;-)

I'll see what I can do once I see about the missus lunch and todays fish
wrap, if she's awake, she wasn't at noon local. I'll nip off a 1/4" of
that cap tube, clean up the ends, and see about soldering that into the
inner nozzle, then rigging a 1/4" of something to bring the annular gap
to the end of the brass adjustable cone into scale for the cap tube's
OD.

Except I've used up the brass tubing I had in 1/32" increments so they
fit into the next one up or onto the next one down.  And the old boy
that ran the hobby stop up in bridgeport closed up and retired about 3
years back. So I need to find another place to source it.

Google and hints welcomed.


Les


Thanks Les

Cheers, Gene Heskett



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] misters for really small setups?

[Brent Loschen]

On 10/2/2019 7:18 AM, Gene Heskett wrote:

On Wednesday 02 October 2019 00:36:41 Brent Loschen wrote:


<https://www.ebay.com/itm/Mini-Peristaltic-Pump-Head-With-Tube-Small-F
low-Stepper-Motor-OEM-Package/273419600046?hash=item3fa91400ae:g:h1MAAO
SwMdZbe77a>


I sorta went crazy, ordered two kits of telescoping brass tubing to help
make better, smaller nozzles. I can solder up a 1/16 od, and redrill it
at #80 to get an even smaller tip. Then off this link, which takes a
short nema 42 motor with a 10mm shaft (no way do I have a motor that
big), but I did find a 12 volt rated peristaltic that looked usable at
$6.32/copy. So since even the silicon tubeing has a limited lifespan in
service of only a year or so, I bought 4 of those. But they're in china,
so a Nov 1 estimated delivery date. Spit.
If you check out the mechanical drawing of the pump in the link above 
(last thumbnail to the right), it appears to have both 31mm and 47mm c-c 
mounting holes which match  NEMA 17 and 23 steppers respectively.  Not 
sure why they reference a 42 motor which has a larger bolt pattern (89) 
than the overall dimensions of the pump (57)!
Although it would be nice to have adjustable spring tension on the pinch 
arm to accommodate a wide range of pump tubing, it's hard to beat the 
price ($13 delivered) for a 6 roller pump.  Precise speed control of a 
stepper should be a lot simpler than to PWM a DC motor.  Would it be 
possible to treat that pump/stepper as a joint and set the RPM via 
feedrate in Gcode?


Brent


Now if I can write hal code to allow enabling a PWM module to drive this
for speed control, at the same time the m7 is on, and some more pyvcp to
dial the speed via controlling the pulse width, I should be set better
than I am right now. Enough air pressure to keep the work area clear,
and just enough fluid to cool things without soaking the neighborhood.

The stepper drive is a nice idea, and I may yet pull one of the 225's off
the remains of the hf and put it on this 6040's z drive.  OTOH, unless I
start doing a lot of 3d woodworking, this is fast enough. Famous last
words...
==
Speaking of PWM, is it possible we (meaning LCNC) have a means to detect
the rise in back emf as the motor starts to turn, effectively pulsing
the motor so it has enough power to overcome roller friction, but would
them turn it off after 2-10 degrees of rotation?  Should make a much
more controllable low flow speed control.

Somehow, I doubt we could turn it off and sample the back EMF with one of
a 7i76's 12 bit A/D's fast enough to make that work, and the docs don't
say how long to a valid reading, but can you tell me, Andy or Peter?

That way I could use a solid 12 volts to bang it with, detect that the
motor is moving, extending the pwm's off time for speed/flow control and
that should allow flows that hardly wet a finger.  Or have I let my
imagination out to play w/o a chaperone again?

Won't be the first time, and I hope not the last. :)

Thanks.

Cheers, Gene Heskett




___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] help with spindle speed

[Brent Loschen]

Almost a year ago I converted my servo'd Bridgeport/Boss to LinuxCNC and 
it works sooo much better than that old and slow 68000 control that I've 
been reluctant to change anything.  The only remaining todo is to 
convert the spindle speed control from the current mechanical switch to 
machine/digital control, and it's a perfect winter project.


To that end I've set up as a small test bed in the comfort of my home 
with a 5i25/7i76 controlling 3 steppers, 3 home switches, and a PWM 
driven DC motor.  I configured LCNC-2.7.14 with pncConf using 
"5i25-internal Data" and "7i76x2 - With one 7i76".  All steppers, 
switches, homing, etc work as expected.  I've fabricated a rotary 
encoder with slots and an IR sensor for the motor  that outputs +5v 
pulses and verified it with an oscilloscope.  The DC motor speed is 
currently controlled with a pot wiper on a small PWM driver board. The 
goals of this test bed are to show spindle RPM , control the DC motor 
with 0-5vdc from the 7i76 , and use knowledge gained to convert the BP.


My first goal was to simply show the RPM on the Pyvcp spindle speed 
panel, so in addition to the 3 stepgens, I also enabled encoder 0 in 
pncConf by selecting the "i/o connector 3" tab, and changed the "unused 
Spindle Encoder 0" dropdown to "Spindle Encoder 0".  I then attached the 
output of my encoder to TB3 pin 7, which according to the 7i76 manual, 
should be ENCA+.   Got no RPM indication on the panel, so I verified 
that I had pulses on TB3-7 with my scope.  Next I did a dmesg and saw 
that encoder #0 had been instantiated on Port 3-13, and checked my .hal 
file to make sure that encoder.00.counter-mode was set to 1, (as I only 
have an A input). I then looked at the hal configuration and put a watch 
(also halmeter) on a number of pins associated with spindle and encoder 
to ensure machine was homed, spindle was enabled, and to look for any 
changes on hm2_5i25.encoder.00.input-a, but nothing.   Toggling the home 
switches, and the axis spindle enable button showed that my pin watches 
were working as expected, and that the spindle was truly enabled.  All 
good.  I'm to the point where I'm wondering if TB3 pin 7 is the correct 
input, or if I should be using a different 7i76 input pin.
I'm somewhat of a hal newby, and although I feel I understand the 
basics, I have a lot more to learn.  Thanks in advance for any help you 
can offer. Hal and ini files attached.



Brent

# Generated by PNCconf at Tue Jan 14 11:43:52 2020
# If you make changes to this file, they will be
# overwritten when you run PNCconf again

loadrt trivkins
loadrt [EMCMOT]EMCMOT servo_period_nsec=[EMCMOT]SERVO_PERIOD 
num_joints=[TRAJ]AXES
loadrt hostmot2
loadrt hm2_pci config=" num_encoders=1 num_pwmgens=0 num_stepgens=4 
sserial_port_0=20" 
setphm2_5i25.0.watchdog.timeout_ns 500
loadrt pid names=pid.x,pid.y,pid.z,pid.s
loadrt abs names=abs.spindle
loadrt lowpass names=lowpass.spindle
loadrt scale names=scale.spindle
loadrt near

addf hm2_5i25.0.read  servo-thread
addf motion-command-handler   servo-thread
addf motion-controllerservo-thread
addf pid.x.do-pid-calcs   servo-thread
addf pid.y.do-pid-calcs   servo-thread
addf pid.z.do-pid-calcs   servo-thread
addf pid.s.do-pid-calcs   servo-thread
addf scale.spindleservo-thread
addf abs.spindle  servo-thread
addf lowpass.spindle  servo-thread
addf near.0   servo-thread
addf hm2_5i25.0.write servo-thread

# external output signals


# external input signals


# --- HOME-X ---
net home-x <=  hm2_5i25.0.7i76.0.0.input-00

# --- HOME-Y ---
net home-y <=  hm2_5i25.0.7i76.0.0.input-01

# --- HOME-Z ---
net home-z <=  hm2_5i25.0.7i76.0.0.input-02

#***
#  AXIS X
#***

setp   pid.x.Pgain [AXIS_0]P
setp   pid.x.Igain [AXIS_0]I
setp   pid.x.Dgain [AXIS_0]D
setp   pid.x.bias  [AXIS_0]BIAS
setp   pid.x.FF0   [AXIS_0]FF0
setp   pid.x.FF1   [AXIS_0]FF1
setp   pid.x.FF2   [AXIS_0]FF2
setp   pid.x.deadband  [AXIS_0]DEADBAND
setp   pid.x.maxoutput [AXIS_0]MAX_OUTPUT
setp   pid.x.error-previous-target true
setp   pid.x.maxerror .0005

net x-index-enable  <=> pid.x.index-enable
net x-enable=>  pid.x.enable
net x-pos-cmd   =>  pid.x.command
net x-vel-cmd   =>  pid.x.command-deriv
net x-pos-fb=>  pid.x.feedback
net x-output=>  pid.x.output

# Step Gen signals/setup

setp   hm2_5i25.0.stepgen.00.dirsetup[AXIS_0]DIRSETUP
setp   hm2_5i25.0.stepgen.00.dirhold [AXIS_0]DIRHOLD
setp   hm2_5i25.0.stepgen.00.steplen [AXIS_0]STEPLEN
setp   hm2_5i25.0.stepgen.00.stepspace   [AXIS_0]STEPSPACE
setp   hm2_5i25.0.stepgen.00.position-scale  [AXIS_0]STEP_SCALE
setp   hm2_5i25.0.stepgen.00.step_type0
setp   hm2_5i25.0.stepgen.00.control-type 1
setp   hm2_5i25.0.stepgen.00.maxaccel [AXIS_0]STEPGEN_MAXACCEL
setp   hm2_5i25.0.stepgen.00.maxvel 

Re: [Emc-users] help with spindle speed

[Brent Loschen]

On 1/14/2020 9:13 PM, Peter C. Wallace wrote:

On Tue, 14 Jan 2020, Brent Loschen wrote:


Date: Tue, 14 Jan 2020 20:06:36 -0700
From: Brent Loschen 
Reply-To: "Enhanced Machine Controller (EMC)"
    
To: "Enhanced Machine Controller (EMC)" 


Subject: [Emc-users] help with spindle speed


Almost a year ago I converted my servo'd Bridgeport/Boss to LinuxCNC 
and it works sooo much better than that old and slow 68000 control 
that I've been reluctant to change anything.  The only remaining todo 
is to convert the spindle speed control from the current mechanical 
switch to machine/digital control, and it's a perfect winter project.


To that end I've set up as a small test bed in the comfort of my home 
with a 5i25/7i76 controlling 3 steppers, 3 home switches, and a PWM 
driven DC motor. I configured LCNC-2.7.14 with pncConf using 
"5i25-internal Data" and "7i76x2 - With one 7i76".  All steppers, 
switches, homing, etc work as expected. I've fabricated a rotary 
encoder with slots and an IR sensor for the motor that outputs +5v 
pulses and verified it with an oscilloscope.  The DC motor speed is 
currently controlled with a pot wiper on a small PWM driver board. 
The goals of this test bed are to show spindle RPM , control the DC 
motor with 0-5vdc from the 7i76 , and use knowledge gained to convert 
the BP.


My first goal was to simply show the RPM on the Pyvcp spindle speed 
panel, so in addition to the 3 stepgens, I also enabled encoder 0 in 
pncConf by selecting the "i/o connector 3" tab, and changed the 
"unused Spindle Encoder 0" dropdown to "Spindle Encoder 0".  I then 
attached the output of my encoder to TB3 pin 7, which according to 
the 7i76 manual, should be ENCA+.   Got no RPM indication on the 
panel, so I verified that I had pulses on TB3-7 with my scope.  Next 
I did a dmesg and saw that encoder #0 had been instantiated on Port 
3-13, and checked my .hal file to make sure that 
encoder.00.counter-mode was set to 1, (as I only have an A input). I 
then looked at the hal configuration and put a watch (also halmeter) 
on a number of pins associated with spindle and encoder to ensure 
machine was homed, spindle was enabled, and to look for any changes 
on hm2_5i25.encoder.00.input-a, but nothing. Toggling the home 
switches, and the axis spindle enable button showed that my pin 
watches were working as expected, and that the spindle was truly 
enabled. All good.  I'm to the point where I'm wondering if TB3 pin 7 
is the correct input, or if I should be using a different 7i76 input 
pin.
I'm somewhat of a hal newby, and although I feel I understand the 
basics, I have a lot more to learn.  Thanks in advance for any help 
you can offer. Hal and ini files attached.



Brent





Is the 7I76's encoder input set for Single ended inputs?
If not, you will not get any signal from a single ended source
That solved it Peter!  I checked the field power jumpers before I 
installed the 5i25, but never went back and changed the spindle jumpers.

Thanks Again!


Peter Wallace
Mesa Electronics


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users




___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] help with spindle speed

[Brent Loschen]

Thanks for your reply Andy, and yes, PWM would be odd choice for the 
BP.  I chose PWM for this "lab" project, because in addition to the BP 
mill, I have a LCNC router with an analog controlled VFD which is next 
in line for a conversion.   Getting an analog controlled PWM to work 
properly seemed the most challenging, and it forces me to dig a little 
deeper into hal.  I feel much of what I learn here is applicable to the 
Bridgeport, even though I won't be using PWM there.


The Bridgeport does have the split sheave/air motor "vari-speed" setup.  
I've removed the pot used by the BOSS control, and fabricated an encoder 
wheel with I/R sensor that sits inside a 3D printed housing which bolts 
to existing tapped holes in the top of the spindle housing.  As much as 
I would like to have rigid tapping, I feel the mechanics of that old BP 
aren't up to the task, which is why I have just a single encoder and no 
index.   I've seen a couple of posts from others who have used various 
combinations of spindle-at-speed plus  others to control the air motor 
which I hope to leverage.


Brent

On 1/15/2020 3:06 AM, andy pugh wrote:

On Wed, 15 Jan 2020 at 03:09, Brent Loschen  wrote:


The DC motor speed is
currently controlled with a pot wiper on a small PWM driver board. The
goals of this test bed are to show spindle RPM , control the DC motor
with 0-5vdc from the 7i76 , and use knowledge gained to convert the BP.

Are you planning to use a VFD and control spindle motor speed, or use
a mechanical variable speed?

If you have a 7i76 then using PWM seems a bit odd, as the 7i76 has a
built-in digital potentiometer for spindle speed control.





___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] help with spindle speed

[Brent Loschen]

On 1/15/2020 3:22 PM, Gene Heskett wrote:

I should probably interject here that with a good high-res encoder, it
won't care about the wobbly speed mechanics as long as it keeps on
turning under load. The z axis rigid tapping motion is 100% controlled
by the spindle position as it turns.  Since the Z is typically the
slowest axis, you will have maximum spindle speeds to deal with because
Z can't keep up. My G0704 can reverse directions much faster than Z, so
I have some hal trickery in the form of a limit3 to shape the reversal
decel/accel's into something resembling what the tv folks call a "sine
squared" signal, where the peak of the delta in speed matches the zero
speed and the z motor _can_ keep up because it has maximum torque at
zero speed.  This "shaping" of the reversal of the spindle has almost no
effect on the actual reversal times.

Hi rez in the encoder allows much higher Pgain in the spindle motors PID,
and with the speed being helped by the air drive if its wired up to the
high.on.low signals and fast enough in reaction times, rigid tapping
should not be a problem. Haveing some experience with split sheeves
needing several revs to adjust, I can see that being a problem, but it
would be an educational experiment to see how big a problem.
In addition to the split pulleys, my "mechanical" concerns are more 
along the lines of the heavy duty electrical contactors clunking every 
time the fixed speed, 3 phase AC spindle motor is called to switch back 
and forth; increased heat in the motor windings due to those reversals; 
and the air powered, mechanical spindle brake snapping on and off 
hundreds of times.  Yes it was very tempting, and yes, I'm sure some of 
my concerns could be mitigated, however I feel that this old mill (cira 
1985) just isn't the right tool for the job.



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] 7i76 TB4 Spindle + - question

[Brent Loschen]

I seem to recall stumbling across a post regarding a high current draw 
between the Spindle + and Spindle - (7i76-TB4 pins 1 and 3), but I'll be 
darned if I can find it now using the full might of Google.  My question 
is this:  is there any problem using 5V from the cable/host interface to 
power those pins, or do I need a separate power supply?  (I currently 
power the card from the cable) If current draw is no problem, does 
anyone see a problem with stability or noise from that source?



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Question regarding power requirements for 7i76 Spindle+ and Spindle-

[Brent Loschen]

I currently power a 7i76 from the 5i25/cable, and am wondering if it's 
OK to power the Spindle + and - (TB4 pins 1 and 3) using that 5v source, 
or if it requires a separate 5v supply?  If current draw from the cable 
is no problem, are there any other drawbacks to using that 5v source? 
(i.e. noise, stability, etc.)  And finally, Spindle + and - pins appear 
to be isolated from the rest of the card, and it seems like I should 
connect Spindle - to 5v neutral regardless of which 5v I use.  Do I have 
that right? (it will be connected to a VFD with 0-5 input requirement)


Thanks,
Brent


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Probing a Profile

[Brent Loschen]

On 2/7/2020 5:31 PM, Andy Pugh wrote:



On 7 Feb 2020, at 23:45, Greg Bernard  wrote:

That is interesting and useful, but I wonder if probing the important
features and simply tracing a scan or a photograph of the profile in CAD
would be quicker and easier for that particular part.

Possibly, but at least this way it is definitely scaled in mm, and furthermore 
the same mm as my mill.



I either do a direct scan of the part, or trace on paper then scan, and 
bring the bitmap into Inkscape.    From there I convert it to svg, and 
let Inkscape find the edges, save the profile as a dxf, and bring that 
into CAD.  It's usually easy to get < .5mm accuracy. With extra care and 
some tweaking in Inscape it's not impossible to hit .25 - .30 mm edge to 
edge accuracy, especially from a direct scan.  If the part is scan-able, 
it also works for locating holes and pockets.  Large parts can be 
scanned in sections and stitched together in the tool.  Unfortunately , 
even .25 isn't good enough for some applications.


When I saw the video, I couldn't help but wonder if there isn't a faster 
way to probe yet still maintain accuracy.  I thought about a hybrid 
approach, where you take that profile from Inkscape and feed it to a 
routine that probes the actual part, but now from within a 1-2mm 
envelope of the provided profile.  If the profile has long straight 
edges, the routine need only probe the beginning and the end of that 
line.   "Smart probing"! Just a thought.



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Emc-users Digest, Vol 166, Issue 122

[Brent Loschen]

On 2/29/2020 8:14 AM, chris wrote:

I went to my shop the other day and it's been so long since I used my little 
mill I cant remember my login. Any suggestions? I really dont want to have to 
do a clean install.Chris
null
___

You should be able to boot the machine into single user mode, then from 
the command prompt, issue the "passwd" command to change the old 
password to something new.  If you are unfamiliar with single user boot, 
google "boot linux single user" and you'll find several explanations on 
how to do it through the grub bootloader.


Brent


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] RXTX problem

[Brent Loschen]

I've used jssc, 
 the successor 
to RXTX, with much success.  It's multi platform and open source.  Never 
had much luck with RXTX.


On 4/29/2020 1:14 PM, Gene Heskett wrote:

This has nothing to do with linuxcnc other than this install is from one
of our installers.

I have a utility that talks to a 35 yo legacy computer, but needs the
RXTX libraries to do it.

It hasn't worked since the old asus mobo burned up and was replaced with
a newer, faster board and cpu also an Asus.

Several of these libraries including one I just installed, are here and
there but it still won't load. I suspect its a path problem, but even
dropping it into the same directory with the .jar that needs it isn't
working.

I need a fresh idea. Anybody have a clue?

Thanks.

Cheers, Gene Heskett



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] RXTX problem

[Brent Loschen]

On 4/29/2020 3:52 PM, Gene Heskett wrote:

On Wednesday 29 April 2020 16:05:11 Brent Loschen wrote:


I've used jssc,
<https://github.com/scream3r/java-simple-serial-connector> the
successor to RXTX, with much success.  It's multi platform and open
source.  Never had much luck with RXTX.


But how do I convince the .jar that wants RXTX, that jssc is the same
flavor its looking for.

Thanks.
The api's for RXTX are different than those for jssc, so it's not a drop 
in replacement.   In the Java world, the jssc library is considered 
superior to the ancient Java wrapper library for RXTX native code, but 
to use it would require changes to the Java source code for your 
application (your .jar file).  Apologies if I misunderstood your 
original request.


If it's potentially a configuration problem, this documentation 
<http://www.jcontrol.org/download/readme_rxtx_en.html>may help you 
determine if the RXTX components are in the right locations relative to 
your Java installation.


What does the Java application do?  Do you have the source code for it?  
If so, I'd be happy to take a crack at changing it to utilize jssc.





On 4/29/2020 1:14 PM, Gene Heskett wrote:

This has nothing to do with linuxcnc other than this install is from
one of our installers.

I have a utility that talks to a 35 yo legacy computer, but needs
the RXTX libraries to do it.

It hasn't worked since the old asus mobo burned up and was replaced
with a newer, faster board and cpu also an Asus.

Several of these libraries including one I just installed, are here
and there but it still won't load. I suspect its a path problem, but
even dropping it into the same directory with the .jar that needs it
isn't working.

I need a fresh idea. Anybody have a clue?

Thanks.

Cheers, Gene Heskett

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Fwd: missing feature of openscad

[Brent Loschen]

The best solvent based adhesive for PLA that I've found is 
Dichloromethane (AKA methylene chloride).  I dissolve PLA filament in it 
until I have a thick liquid about the consistency of model cement or 
slightly thicker.  Apply quickly and don't waste any time assembling the 
parts, as the solvent evaporates quickly.  Thicker mixtures can be used 
to fill in voids (color matched) or build up areas of a print for 
finishing by other processes.  Due to its super low viscosity, it can be 
dribbled into a joint and it wicks in nicely, although not as strong as 
the cement method (and it can be messy).  Of the many adhesives I've 
tried, it creates the very best bond in both adhesion and strength - 
much like model cement on polystyrene.  Small amounts are available from 
a company called Flex-i-file.com.  Product is called Plasti-i-Weld.  It 
costs around $20 on Amazon, however I was able to buy a 500ml bottle of 
it for about the same price from a chemical supply place.  Use 
responsibly in a well ventilated area.  FWIW, the MSDS says it's listed 
on the iarc & ntp carcinogen list but not by OSHA - (but what isn't?).


On 6/4/2020 4:46 PM, Gregg Eshelman via Emc-users wrote:

For a toothed pulley with a flange on both sides I'd modify the 3D model to 
make one flange a separate part, with features on it and the body of the pulley 
to align it so that when glued on it will be concentric.

To glue PLA, cyanoacrylate (super glue) works well. So does tetrahydrofuran, 
which is in some types of PVC pipe cleaner/primer. The PVC cleaners without THF 
won't bond PLA.
The ridged surfaces of FDM prints can be used to make parts that fit together 
like they're snap fit. Sizing part fits just right then adjusting extrusion can 
make the same models easily drop together, a light press fit that holds but is 
separable, or a press fit that's not tight enough to distort but is never 
coming apart without something breaking.

 On Thursday, June 4, 2020, 3:14:31 PM MDT, Chris Albertson 
 wrote:
  If you have not enabled supports and are printing an object like a pulley
you WILL have a total mess.    3D printer need a surface and an part of the
part that overhangs with just air onder it will not print well as the hot
plastic will just fall out the nozzle.

But that cross hatch pattern looks like support base so it must be enabled.
     You can enable support either "from the baseplate up" or "everywhere"
and you need "everywhere" as at to support the top flange.

Yes, removing support is a PIA because on a pulley the teeth will be
completely 100% inside a ring of support material.

For a beginner you picked a hard project.  Most people would start by
printing a one-inch cube or a 1/2 inch flat washer.  Pulleys have need of
support and have printed threads for the set screw.  Both as
"intermediate level" skills.    Horizontal threads always have
support inside that needs pin punch (hex key) to remove

I had to modify the pulley design so it could print with minimal support.
Take all the ridges off the outside, make the flange thicker and radius the
inside of the flange

It is also possible to control the support density and pattern.

Getting all this right nly tak seconds if you know to do it.  I start in
Fusion and vew thepart from al sides to see if it can be better designed to
reduce the need for support.  Then in Cura I try a few suport techniques
and use preview to see where it is going to go.  If it looks bad, then
back to Fusion to change an angle or whatever.    This is why engineers
like really fast workstatins as design is iterative.

I'll look at a file if you can e-mail it.
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Design for the materials and process

[Brent Loschen]

On 6/7/2020 8:41 AM, Gene Heskett wrote:

On Sunday 07 June 2020 01:21:39 Gene Heskett wrote:

I went back to the 12 tooth I'll use for the motor pulley on the BS-1
clone, and made it with the newer configuration.

Surprise, no junk in the access hole or the nut pocket, but I'd increased
the nut size to a 5mm nut, which left the outside wall a bit thinner
than I'd like, so I'll add a couple mm's to the hub diameter.

But before I do that again, perhaps someone could give me better
measurements for both its width and thickness for a std 5mm nut?

8 x 4 check out: https://www.mcmaster.com/90592A095


This pulley would be usable, but belt tension is going to be quite high
as the teeth, at this diameter are rounded enough to allow it to hop.
Its a good looking render other than that.  Would slowing the movement
speeds down improve that?  Render time was just a few seconds over an
hour, which is a huge improvement.

Adhesion problems are gone if the first layer is under .1mm thick.  Thats
about onion skin if using a paper for gage.

Many thanks for all the help folks.

Cheers, Gene Heskett




___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] ot: need a motor control switch

[Brent Loschen]

Without putting a pencil to it, what you've outlined sounds suspiciously 
like a R/S flip flop (or latch) driving 2 SPST relays or SSR's.  Not 
quite sure of your "open center" requirement though. What do you mean by 
"electrically controllable"? - digital logic (ie microcontroller?), or 
low voltage electro-mechanical?  More detail required.


On 8/11/2020 9:58 PM, Gregg Eshelman via Emc-users wrote:

Just did some looking and found that DPDT solid state relays are uncommon 
beasts, expensive and often discontinued. I also found this interesting thread 
on stack exchange
https://electronics.stackexchange.com/questions/174995/creating-dpdt-solid-state-relay-with-fets

 On Tuesday, August 11, 2020, 7:26:10 AM MDT, grumpy--- via Emc-users 
 wrote:
  i have a small 12v dc gearmotor rated at 50 watts
the motor will be run at full voltage forward and reverse
i need a switch for this that is electrically controllable
something like a double pole double throw relay and a seperate double pole 
single throw relay
or a double pole three position with an open center
any suggestions
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users




___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Success story and a question

[Brent Loschen]

I've been following this list for a couple of years, hoping that perhaps 
through osmosis, I would learn enough about Linuxcnc to convert my old 
Bridgeport R2E4 to a modern control.  The first step towards that 
endeavor was to build a Mesa/stepper motor based, cnc router and learn a 
little LCNC along the way.   The router project went very smoothly, 
especially given how little I initially knew about LCNC.  Tackling the 
servo based R2E4 was still pretty far outside my comfort zone however.  
Getting all of the servo stuff set up properly seemed like it posed 
enough challenges that I didn't want to kill a working mill just for the 
sake of upgrading the control.   I kept putting off changing out that 
old, slow, Motorola 68k based control until it gave up the "axis enable" 
ghost a couple months ago.  For all intents and purposes, the machine 
had become a boat anchor overnight.  At that point I figured I had 
little to loose by tearing into it.  Since I had all of the schematics 
to the machine, I started mapping out all of the gazinta's and 
gazouta's.  All pretty straightforward stuff.   Next I moved into the 
cabinets and removed everything but the large analog axis drive power 
supply, the large contactors (for drive enable, spindle forward/revers, 
etc), the wiring that went to the axis drives/encoders, and much of the 
fundamental cabinet-cabinet wiring and power distribution.  Snip Snip 
and I had a heaping pile of circuit boards, card cages, wires, etc.  
Since the old servo drives were now in the junk pile, I needed to find a 
replacement that would work with Mesa hardware.  My good friend pointed 
me to these step/direction 
 
based servo drives which have servo tuning firmware built into them.  
(anyone else use them?  feedback??)  I installed new encoders on the 
servos and wired everything up.  A few solid state relays were added to 
drive the air brake solenoid, spindle speed motor/sheave, 
forward/reverse contactors, etc. from the Mesa card.   Even added a 4th 
axis channel.  For the smoke test, I opened all of the circuit breakers, 
and slowly brought the machine up one breaker at a time.  No smoke!  
After basic LCNC configuration, we had motion and made chips.  
Everything worked like a champ and it was a fun project that only 
consumed about 4 weekends of my time.  (thankfully, that mill is a 
hobby, and not my livelihood!)


And now for my question.  The old Bridgeport had a button on the front 
panel titled "No Z" that turned off all z motion and let me "air mill" a 
part as a sanity check of my X & Y boundaries/fixtures.  I can't find 
the equivalent functionality in LCNC and was wondering if anyone here 
has done this, or knows the best way to implement it.  (the LCNC display 
certainly shows the part and the toolpath, but for me, it's so much 
better to see the part actually move under a motionless cutter)  
Ideally, I would like to have a button in AXIS that performs that task 
much as it did on the original Bridgeport control.


Many thanks to all of you talented individuals who contribute here!

Brent
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Success story and a question

[Brent Loschen]

Thanks for those pin names Andy, and great point about a premature 
toggle of the "No Z" button.  I'm not sure what the original BP did if 
you attempted to disable the No Z  button while a program was running.  
Given the safety features of that old BP control, I tend to think that 
the button was probably disabled/ignored during program execution, but I 
never tested it.   Whenever I used the "No Z" feature, I ran as much of 
the program as I needed for my XY sanity check, then stopped the program 
and turned off No Z before rerunning the program with Z motion.


In order to get that behavior in LCNC, I would need a signal that 
indicates whether or not a program is running.    I did a few searches 
and it looks like halui.program.is-running AND'd with halui.mode.is-auto 
is potentially one solution (IOW, a program is running, and it's not an 
MDI or MPG command).  Sound right?  Is there a better way?


Now I'm off to learn a little about AXIS and how to add a button!

Thanks,
Brent

On 8/8/2018 3:18 AM, andy pugh wrote:

On 8 August 2018 at 03:17, Brent Loschen  wrote:


And now for my question.  The old Bridgeport had a button on the front panel
titled "No Z" that turned off all z motion and let me "air mill" a part as a
sanity check of my X & Y boundaries/fixtures.  I can't find the equivalent
functionality in LCNC and was wondering if anyone here has done this, or
knows the best way to implement it.

I suspect this would start off looking easy and turn out to be
surprisingly difficult.

Turning off the Z is pretty easy, you would disable the Z stepgen and
(at the same time) connect  joint.2.motor-pos-fb to
joint,2,motor-pos-cmd rather than to hm2_5i25.0.stepgen.2.position-fb.

Now the machine thinks that Z is moving, so it is happy, and the
stepgen is disabled, so no steps are sent. That just needs a mux2
component in the HAL.

But what happens when you release the button? Suddenly the machine
will realise that the Z axis is not where you told it it was, and you
will get an immediate following error. Also, the Z will attempt to
move to the "real" position at a speed only limited by the stepgen
accel and velocity limits.

So you probably need some Hal-trickery to gently move the Z back in to
commanded position, while at the same time lying about the position to
prevent following error being triggered.




--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Question on DM860 stepper drivers

[Brent Loschen]

On 4/21/2016 7:09 AM, Gene Heskett wrote:
> On Thursday 21 April 2016 04:04:16 Gregg Eshelman wrote:
>
>> Belt drive from the stepper? If there's enough metal in the pulley,
>> drill and tap three holes into the face then get six flanged rubber
>> bushings. Mill out a circle of metal with three holes. Put bushings in
>> holes, bolts through bushings. Screw to pulley. Hooray! A damper. :)
>>
> Chuckle.  Sounds workable Gregg, but its only a 17 tooth pulley in quite
> close proximity to the 42 toother driving the nut, so figuratively
> speaking, no room to swing that cat.  Besides its a double shaft motor,
> and it does have a spool of 1.75" fender washers with elastomer washers
> stacked about 2.25" high on the upper end of the motor shaft.  See best
> view of it at the sig's link + emc/36.html.  But thats not the same
> motor, that 252 was pulled and a 432 put in its place later.  It shows
> the whole new Z drive before it was installed on that toy.  Back up to
> about 29.html to see it installed.  Those dampers work very well, took
> the motor stall from about 9 ipm to around 34 ipm if everything is
> right. But the psu is only 28 volts, and I have a new 44 volter ready to
> go in at the same time as I put a 5i25 into that system.  Even with
> software stepping, I have carved a boat load of stuff on that little
> mill.
>
> But there is very little insulation in that building, so I only keep
> enough wintertime heat in it to keep it above the dew point so its off
> limits to this old diabetic for long enough to setup and do a job in
> cold weather.  I'll also need to round up a 5 volt wall wart to run the
> rest of the logic in that mills electronics box when I make the psu
> switch as the 28 volter has a 5 volt output its all running on now.
> I probably have a suitable one of those as I never throw out the wall
> warts even if the rest of the kit has died & been thrown out.
>
> But it looks like I'll need to chew up some 1/2" stock in the lathe,
> making anchors from the screen door springs that will do some counter
> weighting on the G0704. A 6mm Allen head cap screw with a 20mm long
> spacer, and a 1/8" deep and wide groove should make it a pull out the
> spring and drop the loop over it anchor point.  Easy to change springs
> if I should stumble over some wound from a larger wire.  These, at .052"
> are a bit lighter than I'd like, but they should help.
McMaster has a nice selection of 3' long springs of various OD's, wire 
diameters, and spring rates.  They seem reasonably priced (~$5-$10) in 
sizes likely used for this type of application - 
http://www.mcmaster.com/#extension-springs/=122u4yj.

I'm curious if anyone has compared moving mass (pulley/cable/mass) vs 
springs for counter weighting a G0704.  Other than ease of 
implementation, space requirements, etc., any real world 
advantages/disadvantages of either?  ie Z axis repeatability? Any 
thoughts on the "right" amount of counterweight?  1/2 the head weight, 
more/less?


--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Any Interest or Ideas for a Linuxcnc Fest 2016 ?

[Brent Loschen]

Sebastian,

I am in the Denver area and I know thee are several others here who are 
LCNC enthusiasts (or perhaps newbies like me) who would love to get 
together and swap stories, tips, and a beer.  If you are ever so 
inclined to lend your expertise to such a group, I would be happy to 
organize a meeting place, Meetups, etc.  Please let me know if this is 
of interest.

Brent Loschen


On 6/14/2016 12:35 PM, Sebastian Kuzminsky wrote:
> On 05/02/2016 12:17 PM, Stuart Stevenson wrote:
>> If there is no interest or opportunity somewhere else you are all welcome
>> here.
>> I am in a shop of my own now. It may not be as interesting.
>> I am no longer associated with MPM in Wichita, Kansas nor EMI in Eminence,
>> Missouri.
>>
>> I think sometime in the next few weeks or sometime early fall would be the
>> best time to be here.
>>
>> If we have interest shown we can set a date and let it happen.
> Thanks for the offer.  I think it's better than the offer I made on IRC
> for a meeting in Boulder, CO.
>
> Do you have something like the conference room at MPM, where we can all
> sit comfortably around a table with power and internet?  If so I think
> we'll gratefully accept your offer!
>
> Since the next few weeks came and went, we're looking at some time in
> early fall.  September?  October?  When works for everyone?
>
>


--
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are 
consuming the most bandwidth. Provides multi-vendor support for NetFlow, 
J-Flow, sFlow and other flows. Make informed decisions using capacity planning
reports. http://pubads.g.doubleclick.net/gampad/clk?id=1444514421&iu=/41014381
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Moving axis while paused

[Brent Loschen]

On 6/27/2016 9:56 AM, andy pugh wrote:
> On 27 June 2016 at 15:28, Chris Kelley  wrote:
>> I would say that the "proper answer" is to teach the users to properly use
>> holding-tabs, hold-downs, etc
> That reminds me, I used holding tabs a lot in CamBam, but haven't even
> found them in Fusion360 or Inventor HSM. I wonder if they are there to
> be found?
I can can confirm that Fusion360 does support tabs.  Options exist for 
tab lengths, straight or "ramped" (triangular?) tabs, and either 
automatic (you specify the number of tabs) or manual placement.  I found 
the tab feature to be quite easy to use, especially compared to some of 
the actual 3d modeling capabilities - but that problem is mine alone, 
and not specific to the software!  I'm pretty sure I read that the CAM 
portion of F360 is standard HSMWorks, but I've not actually used that 
product.

--
Attend Shape: An AT&T Tech Expo July 15-16. Meet us at AT&T Park in San
Francisco, CA to explore cutting-edge tech and listen to tech luminaries
present their vision of the future. This family event has something for
everyone, including kids. Get more information and register today.
http://sdm.link/attshape
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Introduction and a couple questions

[Brent Loschen]

Greetings everyone!  New guy here.  I've been following the group for 
several weeks and feel it's time for an intro and to get some 
suggestions for a small project I'm working on.

First, my motivation for looking into LinuxCNC.  I own an older 
Bridgeport R2E4 CNC mill (Boss 9 with DC servo motors) that works 
flawlessly at the moment. However, every time I open the cabinets and 
look at the mass of wires and circuit boards I know at some point, 
something inside is going to fail.  Even though I have the schematics 
and am comfortable doing basic electronic troubleshooting, I know that 
sooner or later I will be faced with replacing that rats nest with a 
modern servo control and software. In anticipation of that event, I 
would like to experiment with different control software to see which I 
like best.  I generally prefer Linux over Windows and am currently 
favoring LinuxCNC.  To that end, I plan to start on a small project and 
go big if/when the BP fails.

To start that journey I would like to convert a Shapeoko type desktop 
mill, from TinyG, to a real CNC control in order to become familiar with 
the CNC software.  My question for the group is, what hardware would you 
all recommend for this task?  I've looked at some of the ~$150 Chinese, 
3 axis, PLC/control/stepper drive combos available on Ebay, but I get 
concerned when I read articles regarding the poor quality and wide 
variation in the actual driver chips used in many of those stepper 
drives.  I'm not sure if those problems are actually related to junky 
hardware, or possibly just operator error, so I'm left wondering if 
anyone has found anything in that realm that actually works?  (I only 
need to drive NEMA 23's)  If not, I would appreciate hearing what some 
of you may be using for motion control/drive hardware for an inexpensive 
mill like the Shapeoko.  One final consideration is whether to use a 
parallel port solution, or if there are USB or network controls 
available in a wallet friendly price range?   I've read that USB 
controls tend to be susceptible to induced electrical noise, but I have 
no idea how widespread the problem is or if a good quality, short, 
shielded USB cable plus shielded stepper cables can resolve noise 
issues?  My preference would be to consider an alternative to the 
parallel interface, but the price of other options may be overkill for 
this small application.  From what I've read, LinuxCNC seems to have 
good support for parallel, but I'm unclear if/how it supports network or 
USB controls and would like to hear your thoughts on that aspect. Down 
the road (hopefully WAY down the road!)  I may poll the group again for 
Bridgeport servo control recommendations, but for now, any suggestions 
on the most appropriate hardware (ie low cost, reliable, LinuxCNC ready) 
for this small 3 axis mill would be greatly appreciated.

Looking forward to hearing feedback from the experts!

Thanks in advance,
Brent

--
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Introduction and a couple questions

[Brent Loschen]

On 2/21/2015 6:27 PM, Kirk Wallace wrote:
> On 02/21/2015 03:42 PM, Brent Loschen wrote:
>> Greetings everyone!  New guy here.  I've been following the group for
>> several weeks and feel it's time for an intro and to get some
>> suggestions for a small project I'm working on.
> Welcome to the world of LinuxCNC.
>
Thank you Andy P., Mark W., and Peter W. for your responses to my 
original post - most helpful!.  My apologies for this delayed response.

I've been reading everything I can, and I'm weighing the various options 
for the desktop mill (to practice on with steppers) vs the Bridgeport 
(brushed DC servo ) conversion down the road.  I'm not sure if some of 
the components used in CNC systems really do have multiple names for the 
same functionality, or if there are subtle differences between them.  
For instance is a motion controller the same as a PLC or FPGA?  In 
certain contexts, I get the impression that they are the same, yet in 
others, a "motion controller" seems to be a  programmable device used 
for fixed functionality i.e. robotic pick and place.  In my world, PLC's 
and FPGA's are discrete semiconductor components, not complete boards 
controlled by them!

Another question I can't seem to answer is, what's the fundamental 
difference between using a standard "bit banged" pport vs using 
something like 7i43 or a bus connected 5i25?In the default case (no 
special hardware), my understanding is that LCNC toggles the lines of 
the pport (step/direction signals?) which are connected to a stepper 
drive.  What I've not been able to answer, is how the fundamentals 
change when using, say a 7i43, connected to the same port?  Instead of 
sending pulses out the pport, LCNC must instead be sending some kind of 
higher level data stream (possibly lower bandwidth? ) to the 7i43 where 
it gets converted to precisely timed step/dir pulses - is that right?  
If so, what is that data and how is it different from step/dir signals 
in the default case?  Can anyone comment on the differences or point me 
to documentation that explains it?

Along the same lines, I've wondered how acceleration/deceleration is 
handled.  Is it always manged by LCNC regardless of bit banging or 
additional hardware, or do FPGA/PLC based boards ever handle that too?

Thanks,
Brent

--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Introduction and a couple questions

[Brent Loschen]

Thanks for this explanation.  It answered a question that I've had for a 
long time.
Brent

On 2/27/2015 4:15 PM, andy pugh wrote:
> On 27 February 2015 at 22:24, Brent Loschen  wrote:
>> If so, what is that data and how is it different from step/dir signals
>> in the default case?  Can anyone comment on the differences or point me
>> to documentation that explains it?
> It might help to start with the parport stepgen.
>
> That has two functions. The slow function runs every time the
> servo-thread runs, typically every 1mS. It reads in the next position
> target from HAL, looks at where it is now, and calculates a step-rate
> required to get to that position by the time the next update comes in
> in 1mS.
>
> Then the base-thread runs about every 25uS and either makes a pulse or
> doesn't make a pulse depending on whether one is needed to maintain
> the step rate requested by the servo thread.
>
> In the case of a Pico or Mesa p-port connected FPGA card the first
> step looks just the same. Code running on the PC looks at position and
> target and decides what step-rate is needed. But this data is then
> passed to the FPGA as data on the EPP bus, and the actual steps are
> generated by the FPGA.
>
> The big difference is step-rate resolution. The "base clock" of the
> bit-banged solution is 25kHz. The FPGA card timer resolution is around
> 10Mhz.
>


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Introduction and a couple questions

[Brent Loschen]

My ultimate goal is to upgrade the DC servo drive in a Bridgeport R2E4, 
if/when it ever goes out.   Before I go there, I want to practice with a 
LinuxCNC installation using steppers on a small desktop mill just to get 
familiar with the the software.  I'm comfortable with Linux but never 
used LinuxCNC before, so there's plenty to learn before I tear into a 
working Bridgeport!

On 2/28/2015 7:59 AM, Stuart Stevenson wrote:
> Why are we worrying about step generation when he stated in his original
> post his machine has dc servos?
> On Feb 28, 2015 8:27 AM, "Brent Loschen"  wrote:
>
>


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Matrix Wiring

[Brent Loschen]

It's not the push-pin/patch cord kinda thingy you described, but how 
about simply using a component carrier and solder jumper wires to make 
the pin-pin cross connections.  You would wire as many carrier jumper 
boards as you have cable combinations.

example component carrier - https://www.peconnectors.com/component-carriers/

Then simply drop one of the jumper boards  into a ZIF socket that you've 
soldered the wires from each cable to each row of pins.  The ZIF socket 
pin -> wire connection would work best on a small perf board with 
holes/solder pads for a sturdier connection.
example ZIF socket - 
https://www.mouser.com/ProductDetail/3M-Electronic-Solutions-Division/224-5248-00-0602J?qs=RC2ne4458IIcDFg0luAdug%3D%3D


Brent

The ZIF socket would probably need to be soldered to a perf board to 
actually make those solder connections sturdy.

On 7/12/2022 3:40 AM, andy pugh wrote:

Not CNC related, but I thought that someone here might know the answer.

I want to make a rewirable adaptor (OBDII to DB15)

One neat way to do this would be with a matrix, where you plug something
into holes and that connects the row to the column.

As a CNC link (almost) I think that something like this was used on the MEC
Microsprint programmable lathes where things often called "diodes" were
plugged into holes to program the machine.

I feel that a couple of PCBs. spaced a small distance apart, could have
plated-through holes linked by "something" being plugged through them. The
problem being to find the "something" that will make contact with two holes
at the same time.

A single PCB and an array of jumpers would work, but that would be 480 pins
to keep track of

The plated-hole idea has 240 holes, but only needs 8 plugs to do the
programming I need.

2mm banana plugs do exist.
https://uk.rs-online.com/web/p/banana-connectors/7632907

But I am not sure if they will reliably connect two plated holes.





___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] no longer receiving the list!!

[Brent Loschen]

Yes!  My subscription also stopped on 8/2.  I too simply re-subscribed 
and everything worked fine after that.  The listserv even kept my old 
password (for my email address/userid), which seems kinda strange.  It's 
like I my subscription was suspended rather than my userid deleted.


I also subscribe to several other other lists that are hosted by the 
same listserver software and none of them dropped me.  Perhaps the 
emc-users admins can explain why some users were dropped/suspended?


Brent

On 8/10/2017 8:00 PM, Cecil Thomas wrote:
I've been on this list for over 15 years with no loss of service for 
more than a day or so when the list was down.


I received my last posting on 8/2/2017.
I thought maybe the list was down but when I checked:

https://sourceforge.net/p/emc/mailman/emc-users/

I see that postings are still happening.

I just re-subscribed to the list and it apparently worked.

Has anyone else been un-subscribed involuntarily

Thanks,
Cecil


-- 


Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users



--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users