Difference between revisions of "CNC plasma cutter"

From RevSpace
Jump to navigation Jump to search
 
(21 intermediate revisions by 4 users not shown)
Line 9: Line 9:
 
This goal is still a while away for now, but as always, we are full of Ideas and this is a convenient place to put them.
 
This goal is still a while away for now, but as always, we are full of Ideas and this is a convenient place to put them.
  
== CNC plasma cutter based on VSL X-Z table ==
+
== System overview ==
 +
The diagram below shows the proposed system setup.
  
=== The working mechanics ===
+
[[File:plasmacnc.svg]]
  
Once upon a time, three consecutive interns at VSL made a X-Z traversing unit to automatically measure wind speed velocity distribution of one of VSL's primary standars. This traversing unit was put out of use over half a decade ago, and was finally put on the scrap metal list. If it wasn't for a hackers intervention, it would have been there by now. Luckily it can now have a second life in the spark shack.
+
From left to right:
 +
* design software (like inkscape) is used to design the shape to be cut and to export it as svg.
 +
* job setup software (unknown, but something like visicut) is used to position the shape and convert it to g-code.
 +
* linuxcnc (or possibly grbl) is used to "execute" the g-code through the control hardware.
 +
* the control hardware powers the motors of the cutting table and controls the plasma torch
  
The X-Z traversing units has a DC motor for each direction of travel. These motors are coupled to a gearbox and a rotary encoder. A toothed belt then transfers the rotational motion into a linear motion.  
+
We're basing this project on an old X-Z traversing unit originally made for measuring air velocity distributions in wind tunnels at a metrology institute. The current motor controllers, however, seem to be made for a fixed speed and it's probably easiest to replace the electronics altogether.
  
=== Hardware that works ===
+
<gallery>
 +
File:X-Z_traverseerunit_front.jpg|X-Z traversing unit before any modifications
 +
File:X-Z_traverseerunit_stuurelektronika.jpg|Current control electronics with DC-motor controllers and Mitsubishi PLC
 +
</gallery>
  
The motors work when operated by the joystick controller. Speed is adjustable by means of knobs inside the contol cabinet.
+
A Z-axis seems required to control the torch height. It won't need to move very far, quickly or precisely, though.
The two motorcontollers work [[http://www.alpatek.com/file.html?id=419]]
 
  
=== Hardware that needs to be sorted ===
+
== To-do list ==
*The onboard RS-232 PLC (Mitsubishi something or other) may or may not work. If it still works, we need a way to program it. Possibly it's easiest to take out the PLC and put an Arduino or Rpi in there instead
+
# Decide how to control the machine. There seem to be two basic options: GRBL and LinuxCNC
*Interface with the plasma cutter (relay, some cable and a plug to match the plug that's already on there)
+
#* [[User:Smeding|Smeding]] knows a little bit about LinuxCNC and is willing to do some of the work
*mounting for the plasma cutter head
+
# Design and build interface/driver hardware for motors and encoders
*some rig to reliably hold the sheets of steel vertical (very important for thin sheets that may bulge or buckle otherwise)
+
#* [[User:Smeding|Smeding]] is willing to help with this
*a PC that will survive the Spark Shack
+
#* [[Benadski]] has motor controller hardware that takes PWM as input.
 +
# Set up a PC that can withstand the general environment of the SparkShack (EMI, humidity, temperature)
 +
# Get basic motion control working so that the machine moves
 +
# Design and build interface hardware for the plasma torch, so it can be turned on and off. Will probably want voltage sensing for torch height control.
 +
#* [[User:Smeding|Smeding]] is willing to help with this, too
 +
# Add a third axis to the machine for torch height control
 +
# Put everything together -- implement the software side of things for torch height control and reading e.g. SVG files
  
  
=== Software that needs to be sorted ===
+
connection diagram:
*Program for the hardware controller
+
<br>
**Motion control
+
[[File:XY-tafel connection diagram bewerkt.JPG|1024px]]
**Speed control
+
<br>
**Plasma cutter on/off
 
**position feedback
 
**...
 
*CNC software to draw/cut
 
**Perhaps use the software from the laser cutter?
 
  
== Original Ideas ==
 
  
 
+
blokdiagram sturing plasmasnijder<br>
=== Specs ===
+
<br>
* 3x3 m working area
+
[[File:Blockdiagram sturing rev 2.1.PNG|1280px]]
* movement speed of 1000mm/s
+
<br>
* Calibrated accuracy and precision of +/- 0.1mm
 
 
 
=== Axis motors ===
 
[[User:Smeding|smeding]] is going to design and build a proof of concept linear (synchronous) motor for use in the X- and Y-axes. This test rig will be used to determine whether a DIY LSM is at all useful for our application and to finalize the design if we choose to go with the concept. Specifically, this involves figuring out:
 
* How the attainable accuracy compares to the wanted specifications
 
* How the attainable movement speed compares to the wanted specifications
 
* A final design for the drive electronics
 
* Coil and armature geometries
 
* Rail geometries (magnet size, spacing and orientation pattern and any backing)
 
* Optimal control schemes
 
 
 
After this, we'll be ready to construct the three rails (2x X and 1x Y) that make up the bulk of the CNC machine.
 
 
 
==== prototype BOM ====
 
* angle extrusion, probably aluminium for the prototype
 
* skate bearings (we already have some)
 
* magnets
 
* copper wire (we already have some)
 
* electronics:
 
** Microcontroller ([[http://www.dickbest.nl/index.php?_a=viewProd&productId=8128|AT90PWM3B]]?)
 
** MOSFET drivers (half-bridge or BLDC)
 
** MOSFETs (N-channel, high current)
 
 
 
== Some rough ideas by Yotson ==
 
 
 
*1.5 x 3.0 Meter max. material size. Spec of plasma cutter mentions 12mm as max thickness.
 
    Table should manage to hold the weight of a 1.5m x 3.0m x 20?mm sheet of steel.
 
 
 
*weight of plasma nozzle + hoses/wires?
 
 
 
*length of nozzle + hoses/wires?
 
 
 
*speed of nozzle over material > 10 cm per second. (non cutting)
 
 
 
*nozzle placement accuracy: < 1mm
 
 
 
*transportable. Able to fit through 'standard' door (on its side, obviously)
 
 
 
*Maybe:
 
'upgradeable' by placing rotary cutting tool, pencil, knife, spray can, .... ???
 
 
 
---------------------------------------------------
 
 
 
Transport axes:
 
x, y:
 
angle iron and skate wheels + bearings?
 
 
 
 
 
linear motors?
 
threaded rod, DC motors, encoders?
 
 
 
z  :
 
? + threaded bolt, manual adjustment between height of 0mm and 50mm.
 
 
 
 
 
---------------------------------------------------   
 
 
 
== Threaded rod info ==
 
ISO standard. From HTTP://en.wikipedia.org/wiki/ISO_metric_screw_thread
 
{| class="wikitable"
 
|-
 
!scope="col" | Diameter in mm.
 
!scope="col" | Pitch (fine)
 
!scope="col" | Pitch (coarse)
 
!scope="col" | Rotational speed 10 cm/s lin. motion (coarse)
 
!scope="col" | Rotational speed 10 cm/s lin. motion (fine)
 
|-
 
|20 or 22||2.5||1.5 or||40rps, 2400rpm||50rps, 3000rpm
 
|-
 
||      ||  ||2    ||              ||66.7rps, 4000rpm
 
|-
 
|30 or 33||3.5||2||28.6rps, 1714rpm||66.7rps, 4000rpm
 
|}
 

Latest revision as of 02:43, 21 October 2015

Project CNC plasma cutter
Status Initializing
Contact [[Project Contact::Walter, Smeding, Gori, Yotson, Semafoor]]
Last Update 2015-10-21

After acquiring a plasma cutter at the space, it seemed only logical to try and fabricate a CNC version eventually.

This goal is still a while away for now, but as always, we are full of Ideas and this is a convenient place to put them.

System overview

The diagram below shows the proposed system setup.

Plasmacnc.svg

From left to right:

  • design software (like inkscape) is used to design the shape to be cut and to export it as svg.
  • job setup software (unknown, but something like visicut) is used to position the shape and convert it to g-code.
  • linuxcnc (or possibly grbl) is used to "execute" the g-code through the control hardware.
  • the control hardware powers the motors of the cutting table and controls the plasma torch

We're basing this project on an old X-Z traversing unit originally made for measuring air velocity distributions in wind tunnels at a metrology institute. The current motor controllers, however, seem to be made for a fixed speed and it's probably easiest to replace the electronics altogether.

A Z-axis seems required to control the torch height. It won't need to move very far, quickly or precisely, though.

To-do list

  1. Decide how to control the machine. There seem to be two basic options: GRBL and LinuxCNC
    • Smeding knows a little bit about LinuxCNC and is willing to do some of the work
  2. Design and build interface/driver hardware for motors and encoders
    • Smeding is willing to help with this
    • Benadski has motor controller hardware that takes PWM as input.
  3. Set up a PC that can withstand the general environment of the SparkShack (EMI, humidity, temperature)
  4. Get basic motion control working so that the machine moves
  5. Design and build interface hardware for the plasma torch, so it can be turned on and off. Will probably want voltage sensing for torch height control.
    • Smeding is willing to help with this, too
  6. Add a third axis to the machine for torch height control
  7. Put everything together -- implement the software side of things for torch height control and reading e.g. SVG files


connection diagram:
XY-tafel connection diagram bewerkt.JPG


blokdiagram sturing plasmasnijder

Blockdiagram sturing rev 2.1.PNG