To most people in the early 70s, electronics equalled TV, radio, giant computers with whirling things and blinkenlights tended by white-coated priests in cavernous rooms, the relatively new concept of “solid-state hi-fi” audio and, if they thought about it, RADAR. So what has and hasn’t changed?
There were a few electronics magazines containing practical projects with just enough theory to understand how they worked, plus many adverts with very small print. Occasionally articles included a PCB layout, and with luck you could buy a ready-made PCB from someone. When you wanted something the local hardware store didn’t have, you used snail-mail to order items, and a couple of weeks later they arrived. Always presuming you were lucky and the advertisers weren’t flying a kite, that is.
Electronic component shops were confined to major cities, and typically had components that could be used in TVs (e.g. high power 10% resistors and horrible wax-coated capacitors) or heaps of surplus equipment and PCBs. You got used to going along just to see what was available this week, having a guess at what was on the PCBs, and using ingenuity to make circuits out of whatever you had to hand. You often desoldered expensive components such as resistors and capacitors and switches, and reused them in the next project. Circuit layout was adjusted according to whatever was available.
Debug and test techniques were primitive, but with a little imagination you can extract useful information using only very basic equipment. The only equipment available when I designed and debugged a 6800-based computer (like the Altair 8080) was:
- great big clunk switches, often originally intended for car dashboards
- LEDs and light bulbs
- analogue multimeters
- homemade power supplies using transformers, bridge rectifiers, enormous 1000uF capacitors, zener diodes and 2N3055 power transistors; 7805s were some years in the future
That mindset of extracting the maximum information from inadequate equipment proved professionally useful. For example, if you are creating the next generaton of high-end oscilloscopes, what scope are you going to use!
Wiring was a real pain, being based on individual single-core wires. Ribbon cables and IDC connectors were some years in the future, so if a 16-bit bus ran between two PCBs then you used 16 individual wires and twisted them together.
Circuit construction techniques were primitive, because decent quality PCBs were out of the reach of amateurs.
Manhattan and Dead Bug techniques were normal, and could still be found in commercial equipment that was in use every day. They worked well for audio, radio and TV circuits, but didn’t scale to the large number of small connections in the new-fangled digital world.
Professional PCBs were layed out by hand in the form of 2:1 scale versions with black tape on a transparent plastic, or with red and blue rubylith tape for “advanced” two-sided PCBs. The artworks were then sent away to fabrication houses where they were photo-reduced during manufacture. Consequently PCBs were expensive and not used for prototyping. Such red/blue artworks seem to have disappeared so thoroughly that I can’t even find any example images on the web, but you can still see the PCBs in, for example, old HP and Tektronix test equipment.
The commercial techniques available to hobbyists were:
- stripboard plus point-to-point wiring, which worked (and still works) well for low frequency applications, but is a pain for wide digital busses
- PCB etch resist made with Deco-Dalo pens dispensing a thick blue ink. They never worked well: either the ink was too thin or it came out of the pen in an enormous splodge. I believe you can still get these things, but I can’t imagine why anyone would want to use them. Etching used ferric chloride, and spent etching fluid was poured down a drain without a second thought.
- spring-loaded breadboards, which are still appallingly unreliable 40 years later. I suppose they have their place if you don’t mind spending all your time debugging your construction technique. I prefer to debug my design.
- wire-wrap, which was reliable, fast, expensive and worked well for TTL frequency circuits. Some commercial computers used this internally, with the wiring being done by early computer-controlled machines
- and 10 years later we had perfboards populated with low-profile IDC sockets. These were fast, reliable, stunningly expensive, and worked well for newer faster logic families.
Hence many hobbyists resorted to Manhattan techniques, sometimes using nails hammered into wood as a support/anchor points. Or invented new ways, based on their skillset and whatever was to hand – but that’s a story for another time.
Paying for items was either very simple, or slow, or difficult and slow:
- by cash, or cheque plus cheque guarantee card (up to £50) within the UK. Turnaround time was 2 weeks, since the recipient waited 10 days until the cheque had cleared before dispatching the goods. People that didn’t have bank accounts, and there were many, used “postal orders” which are essentially a pre-paid cheque issued by the GPO
- credit cards did exist, but very few people had them. People realised that credit cards didn’t give you credit, they gave you debt – and debt was still seen as something faintly dishonourable that only poor people needed. Of course poor people couldn’t actually have credit cards since they were a poor credit risk that would lose the banks’ money!
- if you wanted to buy something from overseas you had to pre-order and buy an expensive “International Money Order”, then put it in the snail-mail and hope it reached its destination before the price had changed. Turnaround time was 6-8 weeks at best.
Makes me wonder why we ever had enough patience to build anything. The lack of distractions probably helped. No doubt in 50 years time my kids will be saying something similar about biohacking.