I was talking to a non-technical friend the other day, and mentioned something about breadboarding a circuit. He looked at me and asked, “What's that?” He had a point: “Breadboard” is one of those standard terms that engineers toss around casually, as both a noun and verb, as in, “The breadboard worked” or “I'm going to breadboard my circuit later.”
Unlike many phrases where the actual derivation is unclear, shrouded in mystery, or subject to different possibilities and myths, we know the origins of the term precisely. Today's breadboards, of course, are anything but a “bread board.”
Way back in the day, a circuit breadboard was literally just that: a wooden cutting board with nails as wiring posts; see here. Wires and components were connected to these nails, and that was the circuit, right out in the open, no secrets. This type of breadboard was even used for vacuum tube projects (watch out for the high voltages!) and some discrete transistor ones extending from DC to lower RF ranges.
Sometimes the nails were replaced with clips, such as Fahnestock clips (which you can still buy; see here) but I found those to be unreliable. In fact, I developed my early soldering skills by learning to solder my wires into these clips, thereby defeating the clip-in/clip-out functionality they offered but greatly enhancing their reliability.
Another breadboarding alternative for prototypes was wire-wrap; I wrote a short column about them for our sibling site, The Connecting Edge. You can read it here (it received many comments, by the way). There were also perforated PC-board material, with and without copper plating.
Then came the DIP (dual inline package) IC, which changed everything about breadboarding. While the DIP was certainly not compatible with a wooden board and nails, various solderless boards were developed for these packages, and they worked pretty well, such as the Global Specialities brand.
(Source: Cornell University)
But DIPs and discrete components are now a small and shrinking part of the circuit's bill of materials, of course. We have surface mount ICs and passives, and ICs with tens and even hundreds of contacts underneath their package. Sure, you can buy adapters and carriers to bring these contacts out to be visible and accessible, such as from Schmartboard, but that has a major effect on layout and signal integrity.
Reality is that for many of today's circuits, the only viable breadboard is an actual PC board. Clock rates are so high, bandwidths so wide, and there are so many signals to route here and there that only a genuine PC board can do it.
For awhile, many hobbyists were making their own with photoresist pens, acid etch baths, and more. But today's circuit density makes that sort of do-it-yourself PC board a real challenge — plus drilling any needed holes with precision placement is difficult.
Of course, a circuit board does not have to be the standard copper-etched type. You can also make your own with a milling process by using one of those excellent machines from LKPF Laser & Electronics. If you haven't seen these in action, they are amazing, and the density, precision, and speed of fabrication they achieve is impressive.
There's still a place for the original breadboard, I suppose, if you're doing a project such as LED lighting, where it's mostly power signals and routing. Beyond that, it's tough. One of my many unprovable hypotheses is that a major reason aspiring EEs (junior high school and up) don't get involved with circuit design — whether analog or digital — is that building a circuit with today's components is quite hard: either you have to figure out a way to use the tiny, surface-mount parts, or try to find DIP and non-SMT versions, and good luck on that. (And when you're done, you can't easily probe or change parts, anyway.)
Many years ago, I met the late Jim Williams in his lab at Linear Technology Corp., and he lamented the challenges of breadboarding with these super-tiny packages. Then he showed me his latest “breadboard.” It was a circuit in mid-air, like a spider's web, with components scattered like trapped flies among the strands. He said this allowed him to check out a circuit's basic concept, and the physical separation minimized crosstalk.
But he also admitted that when reduced to its PC board layout, things could go one of two ways. It could work better than the open-air circuit, due to better ground planes and shorter traces; or it could work worse, due to signal proximity, crosstalk, and noise. He said that even he just never knew what the result would be despite his experience and pretty good hunches.
What's your breadboard look like? How do you breadboard your circuits? Post pictures if possible.