Archive for October, 2014
I’ll be showing my Lissagraph II drawing machine.
Come by and say hello.
Yea,the word your trying to remember is Spirograph… its not that, but kind of like that only motorized and huge.
Otherwise, just like that. It’s resemblance to any mouse-like cartoon character is purely caused by design constraints driven by gravity.
(Scroll down below the picture to read some design details.)
The Lissagraph Construction Details
By Craig Newswanger
The X and Y cranks are driven through toothed belt drives with a 5:1 ratio. The large turntable for the paper uses a direct friction drive with a rubber drive roller attached to the stepper motor. The friction drive method was chosen for the large turntable to avoid backlash effects that would be visible in the drawings. The turntable is a disk of 1/2” MDF that was cut with a router with a long compass arm.
The drive ratio with the 1 1/2” drive and the 48” diameter disk is roughly 32:1. I carefully calibrated each motor in the software to compensate for the imprecise turntable ratio and differences in the motor drive clock crystals.
There is a single idler roller opposite the motor that supports the turntable and three hard rubber casters on the back. The 20 degree tilt on the machine insures that the turntable disk is stable and rotates freely. The friction drive roller and idler were purchased from McMaster Carr.
Gravity is very dependable! You can quote me on that!
Motors and Controllers
The drive motors for the X and Y cranks are M062-FD8103 made by Superior Electric. I purchased them from Surplus Sales of Nebraska. The larger motor driving the turntable is a Nema 34 motor. I bought that from a friend. The controllers are Mforce drives, made by Schneider Electric
I used two Micro-drives for the x and y motors and a Power-drive controller for the larger motor.
The communications to the motors is through RS-422 serial link.
I use a usb to RS -422 adapter from Digikey # 768-1044-ND
My original code for the drawing machines were created in PureData, a data-flow language usually used for audio synthesis. I then wrote new code using Processing and G4P user interface tools. The software really only sends speed commands and Go and Stop commands to the motors. While the machine is drawing the computer is idle.
Speed Ratios and Sensitivity to Initial Conditions
I have been experimenting with this sort of pattern drawing in various forms for many years. I have learned what works to make interesting patterns. People ask if I know what a pattern will look like before I run a particular setting. I have a good idea of the type of image that I will get but there are often surprises that arise from small differences in the starting conditions.
One could refine the hardware so that starting states and phases could be precisely controlled but I’m not sure that would be as fun.
I have also not been interested in pre-visualizing the results using the computer. It is more interesting to see the pattern develop gradually. Observing the machine work is captivating to most people.
Although controlled digitally, the machine can be seen as an analog computer whose output is a direct graphical plot of the algorithm on a piece of paper.