The Hard(ware) Facts

by John J. Smith



Issue 12

Nov/Dec 84

Next Article >>

<< Prev Article




I believe it was Marconi who said that invention is 1% inspiration and 99% perspiration. To relieve some of the hard work in trying to get a recognisable graphics image on the screen, I needed some inspiration for something that would be better than POKEs and DRAWTOs.

What about a light pen? There are, after all, locations for it (54284 and 54285) but I could not find one for the Atari, although rumour has it that the one for the Apple might work. I could of course have built my own light pen but then suddenly graphics tablets came along. Could this be the answer? Not at 90 at time, at least not for me. I put on my thinking cap and came up with the 1% inspiration for a minimum cost device in the hope that someone could write the software and make it work with the Atari. Although this particular version is untried, I did make a working bar-decoder some six or seven years ago but there was little interest at the time. Anyway this simplified version should enable you to produce pictures on a screen.

This is no ordinary light pen. Light pens react to the presence or absence of light from a particular area of a CRT screen when drawing across the screen with the pen. This idea is to place a drawing flat on a table and use the special pen which has its own light source to manually scan across the page line by line to detect light and dark areas of the image. This will produce an equivalent electrical output along the wire to the computer and the signal would be interpreted by software to produce an exact reproduction on the CRT screen. The idea is shown in a simplified form but if you have the time and a Meccano set or something similar, you could perhaps produce a rotating drum around which the picture is fitted. The drum would be fitted on to a threaded rod and driven at constant speed! The pen would be fixed in position vertically and moved horizontally to scan a different part of the picture as the drum rotates. The result would be the same as scanning by hand but would possibly be more accurate and faster.

Due to limited space, I will only describe the 'pen-reader'. The circuit diagram is quite simple and the only tricky bit is getting the spacing right from the paper to the lens. (Figure 1).

The preset pot used for sensitivity adjustments is the type with a screwdriver slot and is best set to midway and then adjusted after construction so that it reacts properly during the transition from black to white. A PEEK at locations 54284 and 54285 would give the necessary reading. There is obviously software needed for this but this should present no problems to PAGE 6 readers!

The light sensitive transistor has three connections but the base (centre) connection is not used. Looking at the flat side with pins pointing down the E (Emitter) is on the left, B (Base) in the centre and C (Collector) on the right The light sensitive part is that circular lens type part and must be fitted so that the light reflected off the page, as seen through the hole in the box, passes through the 8mm lens and falls on this spot on the transistor. I would again emphasise that this particular version is untried and it is probable that some hardware experts among the readership will be inspired by this article to come up with a better solution. If you do, please let me know. I want to try out your ideas!

The 'pen' used to read a drawing is not really a pen at all but instead consists of a small box or other suitable container approx. 4" x 2" x 1". It sounds a little large but remember that it has to contain all the parts. These are as follows.

Light sensitive transistor 2N5777

Lens - 8mm cine movie projector type. 

Lamp -1.5 volt or 1.1 volt torch bulb with lens cap on one end.

Battery - 1.5 volt cell type HP7 

Potentiometer - 470K preset pot

Integrated circuit - 74C04


The inside of the box should be painted matt black to prevent reflections. The hole in the bottom of the box should be approximately 1/2" from one end and the size of the hole approximately 1/8". Size of the hole equals definition to sensitivity and should be adjusted by experiment.

The lens has a threaded portion on the outside allowing a certain amount of zooming enabling adjustment to be made. This adjustment is critical and is best explained by using figure 2. When the lens is the correct distance from the page the black line will fill the whole lens as magnification is quite powerful on this type of lens. Moving it across the stripes will appear to make the lens alternate from dark to light The distance from the page should be measured and the transistor mounted behind the lens at equivalent distance in its own light proof box looking down at the hole through which the page is read. Make sure that the lamp is slightly behind and to one side of the lens and positioned so that it is aimed at the hole. A tall piece of plastic or PCB will help.

There it is then, an idea for a cheap graphics device. There you have the inspiration, can you provide the perspiration and make it work?