Cnc project

Aim of this project

To design and build a table-top CNC machine on a budget of £150.

Background

Waking up the first day of 2009 with a mild hangover and randomly looking around the internet for non-porn I came across a few pages for CNC mills and figured I really needed one to build a spaceship, but the price tag on most CNC's are prohibitive for a hobbyist. Luckily there are many design plans out there for home-build machines. Eager to jump on the bandwagon, I have set myself the goal of spending a little time and money this year designing and building my own CNC machine, and posting the development here.

Specifications

(Subject to change)

Interface: USB (probably)
3-axis design :

• bed size (x,y) : 600x600mm
• bed depth (z) : 150mm

Tool type: Interchangeable, but will probably start with a router
Possible future upgrades include automatic tool switching
resolution: better than 0.1mm

Software

Most likely will use currently available open-source software. Suggestions are welcome.

Electronics

Currently considering micro-controller based solution with upload memory/hard disk to run off projects whilst in standalone mode (dependant on costs).

Design considerations and challenges

Resolution
Resolution of the device depends primarily on two factors, with a third secondary consideration:

• Thread separation - This is the distance between threads on the drive rods. Assuming a nominal separation of 1mm (1000 threads per meter), a single rotation of the rod will move the machine head 1mm.
• Drive resolution - This is the step angle (for stepper motors). A 1.8 deg/step motor will provide 200 steps per full revolution
• Measurable resolution - positions of the axis can be kept track of in software by counting the steps made on the motor, however over time the system may become uncalibrated - for example if the drive motors stall when the software thinks its still moving. Some mechanism for measuring the precise position of the machine head should be employed to provide feedback. This can be through an optical disk attached to the drive mechanism or (my personal favourite) using an optical range finder on each axis.

Speed
The resolution of the device directly affects its speed - stepper motors are designed for low-rpm usage, and the particular ones chosen for this project loose significant torque above 1800rpm (30 cycles / sec). Thus, whilst a system using 1.8 deg/step motor, with a 1mm thread separation gives a resolution of (1mm / 200 = 5um), it produces a rather slow movement even at its top speed - 30mm / second. for a 600x600 work area this gives an end to end movement time of 20 seconds. By using a larger thread width (say a twin/quad helix screw at 4mm separation), or gearing the motor down we can attain faster speeds whilst keeping the resolution well within design tolerances. However motor torque becomes an issue in both cases. I may consider sourcing 7.2 deg/step motors to solve this issue

Clamping
How to keep the piece we're working on in place

Components

Drive
Each axis will run along two parallel cylindrical metal rods of appropriate length, and driven by a long threaded rod powered by a motor. the moving components will be attached to the threaded rod by 2 long (1") bolts - one fixed bolt, and one 'floating' bolt attached with a stiff spring to act as a backlash arrestor (compensates for play in the mechanism).

Progress report

05 jan 2009

I have had a look through my spare parts bin and found 3 stepper motors that should be suitable for the drive system of the CNC, they all appear to be 1.8 degree/step which gives the desired precision. They have been sitting in my parts bin for some time, and if memory serves they have been salvaged from old printers (probably dot-matrix) recovered from skips or throw-aways from friends. Salvage seems to be the best source of stepper motors since these seem to carry a price tag of around £30 and up (each) from most reputable dealers.
The main question is weather they can provide sufficient torque, to that uncertainty the system will need to be designed to house larger motors if necessary.

Additionally I have been working on some design schematics and will post them when they are finished.

More to come as soon as I can be arsed, with pictures