Like most computer systems, ORAC resides in a box. Although the processing takes place distributed through the galaxy, the ORAC unit is the focus to which the information flows via telespace. The original ORAC is quite large and would be inconvenient for me to store as well as somewhat expensive to build, so I aimed at a device of approximately half the volume. This is an appropriate reflection of the technological advances made since the original ORAC device was created.
To make the box, I had to acquire some perspex rectangles from Cadilac Plastics, Oak Flats. Because of the small size of these items and the fact that I know one of the guys working there, this cost me nil.
In order to fit the sides of the box together, I needed to sand the edges to 45° angles prior to glueing them. Otherwise the box would have looked messy. After unsuccessfully trying to achieve this by hand, I borrowed an electric sander from the van Vliet twins.
In the original ORAC, the brain sphere was a clear perspex globe containing a bright pulsating white light and eight red lamps around the circumference (vertical, front-to-back) which flashed on and off. This looked a bit dicky so I determined to make something that looked more high-tech and mysterious. I determined to fill the sphere with steel wool and to light it from the inside. The steel wool would look unrecognisable for what it really is in this context. I devised to place a smaller sphere inside to hold the light source and to place the steel wool between this and the outer sphere, thus obscuring the light source. This assembly was mounted around a perspex tube which in turn spanned the box.
Initially the light source was to be eight independently controllable ultra-brite LEDs, the idea being that the controller could light different parts of the globe at different times. After consideration, I decided eight globes would be insufficient and considered 20 arranged in a [20-sided regular solid surface] This also appeared too few and I realised I would need several dozen lights to achieve the effect of running patterns through the sphere. Also, I thought red LEDs would be too plane and wanted multi-colours. At $2 per red LED, $8 for blue and no secure source for green, I thought too expensive and too hard.
Instead, I thought to use a red and a blue LED under independent brightness control to change the colour of the whole globe. Although bright, individual LEDs were not bright enough and I decided to use four red and four blue, arranging them as two opposing and intersecting triangular pyramids. I also devised a circuit to flash the LEDs too fast to see. This allows me to run them at a higher than normal brightness without destroying them (we see the briter light, but to the LED the average power is the same). This LED block was inserted into the sphere which was glued into the box with its support poles.
The orbital lamps were added as an afterthought and required me breaking the glue bonds holding the globe in the box. I bent two strips of perspex into half circles and drilled holes through them at eight evenly-spaced intervals each. Into these I glued 16 red 1000mcd LEDs and wired them to a ribbon cable. This assembly was then glued around the brain sphere.
Testing the internals of the brain-ball
These are clear perspex tubes containing a strip of coloured translucent perspex in the original ORAC. Their function is not recognisable which is appropriate for highly-advanced technology. These were created by slotting strips of coloured translucent perspex into perspex tubes. I decided to light the rods by inserting regular LEDs into the ends of the perspex strips and putting them under electronic control as well.
I created the first copy of this unsuccesfully. The circuit was botched and I had forgoten to separate the function into two parts - the counters (which could later be replaced by a computer) and the LED drivers (which would stay permonantly). After much trouble to de-solder the chips, I used a fresh proto-board to manufacture two new circuits. Shock of all shocks, all parts worked first test as I assembled them. Cables were attached and the boards glued into the main box.
The control circuit also incorporates a 5 volt regulator to supply a constant voltage from the 6 volt supply.
Testing ORAC to discover why that one dang LED won't light up!
I breadboarded a simple R-C multivibrator oscilator hooked to a speaker to provide the characteristic hum of ORAC running. By hooking a big capacitor across the power supply to the oscilator, I was able to get the power-on rising hum of orac very closely.
Getting the power-down sound proved too difficult in the time provided and, after trying several approaches, I gave up. Since I intend eventually putting a small computer inside this ORAC to give it basic speech recognition and synthesis capabilities, this wasn't a major problem for me as this CPU would take over the sound effects too.
I presented ORAC in class towards the end of session. Everyone was very impressed (ie. couldn't believe what I had constructed). Although individual marks for this assignment was not given, I recieved a High Distinction for the subject for which this project represented most of the marks.
Patchy the mouse performs a quality control inspection and gives it the "tails up".
Some other things I intend to do with my ORAC are...