petergraves

For more on microcontrollers and wearables, check out Make: Volume 43.  Don't have this issue? Get it in the Maker Shed.

For more on microcontrollers and wearables, check out Make: Volume 43.
Don’t have this issue? Get it in the Maker Shed.

Projects that fail are rarely published — but they teach valuable lessons that often lead to success. If you’ve spent much time designing and building projects, you know this well. I certainly do. Some of my failed projects made a major impact on my career in electronics and science.

During my senior year at Texas A&M University in 1966, Texas Instruments announced the development of a powerful LED that emitted several milliwatts of invisible near-infrared, about the same power output as a small flashlight. My great-grandfather had been totally blinded by a dynamite explosion when he was a young man, and the new LED gave me an idea for building a travel aid for the blind. So I hitchhiked to Dallas to meet Edward Bonin, one of TI’s LED engineers.

The new LEDs cost $356 each, about $2,671 in today’s money. Dr. Bonin said he would give me, a rank amateur, one of the LEDs if I could build a circuit that would generate the pulses needed to make the travel aid. I modified a 2-transistor Morse code practice oscillator board sold by a radio and TV repair shop for 99 cents and sent it to Bonin. He approved the circuit and sent it back, together with three of the sophisticated LEDs.

Mims_MAKE_42_Figure_A_

Mims_MAKE_42_Figure_B

I quickly built and documented a prototype and within a few days built a working travel aid that measured 2″×2″×4″. Flashes of invisible infrared emitted by the aid were reflected by objects up to 10 feet away. The reflected IR was detected by a silicon solar cell, and the resultant photocurrent was amplified by a transistor amplifier (salvaged from a hearing aid) and sent to an earphone, which emitted a tone. The closer the object, the louder the tone.

Mims_MAKE_42_Figure_C

I tested the travel aid with more than 20 blind children and adults. It worked well, but the need to hold it in one hand was a drawback. Eventually I assembled the entire device on a pair of sunglasses. All the electronics were installed inside two ⅜”-diameter brass tubes mounted on the temples, the LED transmitter in one tube and the receiver in the other. A tiny hearing aid earphone in the receiver tube was coupled to the user’s ear through a short length of plastic tubing.

The eyeglass aid worked well. It also received an Industrial Research 100 Award and a 1987 runner-up Rolex Award. But in the end, the project I had spent years developing was a failure. The hearing aid companies I approached about manufacturing the travel aid responded that the potential liability was much too risky. What would happen if a blind user wearing the travel aid fell into a hole or was otherwise injured?

Though the travel aid was never manufactured, it taught me more about solid-state electronics and optics than my friends majoring in electrical engineering were learning. They were building old-fashioned vacuum tube circuits in their lab courses, while I was working with transistors and state-of-the-art infrared-emitting diodes.

Mims_MAKE_42_Figure_D

The travel-aid circuits led to several new projects. I used the LED pulse generator circuit to flash a tracking light in night-launched model rockets I was flying to test a new kind of guidance mechanism. After George Flynn, the editor of Model Rocketry magazine, watched one of those flights, he asked me to write an article about the light flasher. It was published in September 1969.

Mims_MAKE_42_Figure_E

Ed Roberts and I were then assigned to the Laser Division of the Air Force Weapons Laboratory. We often talked about selling electronics kits through magazines like Popular Electronics and Radio-Electronics. When the light flasher article was published, we decided to form a company to build and sell light flashers and other model rocketry gear. We called it Micro Instrumentation and Telemetry Systems (MITS).

I eventually left MITS to pursue a new career as an electronics writer. Ed stayed and introduced a string of new products. I wrote the instruction manuals for some of them. I also introduced Ed to Leslie Solomon, the technical editor of Popular Electronics.

In 1974 Ed learned about the 8080, Intel’s new 8-bit microprocessor, and soon began work on a microcomputer based on the new chip. The hobby computer era took off when Ed’s Altair 8800 appeared on the cover of the January 1975 issue of Popular Electronics. When Paul Allen saw the magazine, he immediately bought a copy and took it to show his friend Bill Gates. They soon called Ed to say they were developing a version of BASIC for the Altair, and you know the rest of the story (see my column in Make: Volume 42, “The Kit That Launched the Tech Revolution”).

Sometimes I wonder how this story might have ended had my great-grandfather not been blinded, or if TI hadn’t invented the first infrared LED. Of course it’s impossible to know in advance what might come from a failed or abandoned project — and that’s motivation enough to press ahead.

Going Further

Have you developed a project that failed for technical or other reasons? Think about how it might have advanced your knowledge, and tell us in the comments below.

And consider beginning a new project that’s got only a marginal chance for success. My view is that every project is like a course in tech school or college, for the spinoffs from a failed project are sometimes as significant as those from the great successes.