![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Ain't that a kick in the head?
Nov. 3rd, 2013 03:42 pmI started my engineering career late, five years after graduating from the university. The intervening time was spent working in the Soviet Union, at a number of temporary jobs as I tried unsuccessfully to make my way as a free-lance photographer in New York City, and as a production editor at Plenum Publishing Corporation, where I participated in and oversaw the process of taking a pile of translations and turning them into a published product for a dozen Soviet physics journals.
That all changed in the summer of 1979 after an interview and offer of employment from Reynolds, Smith & Hills, Inc., Architects–Engineers–Planners, of Jacksonville, Florida. My interview came about through the efforts of my half-brother Steve, who had set out to prove that my self-defeating attitude of "nobody'd be interested in me as a potential engineer given my track record" was a lot of bunk.
Originally, the company had the idea of bringing me on board as a mechanical engineer, but that only lasted until the company's "advanced energy" division got wind of the fact that I could string words together in a lucid and persuasive manner—which distinguished me from most engineers—and so I spent a couple of years helping out with proposals and feasibility studies for everything from solar power to energy retrofits for buildings to ethanol production. Then the company's control systems department landed a job that required software to be written to integrate various hardware components into a materials handling system for a coal-fired power plant.
The fact that I was the only engineer in the company who had ever programmed in BASIC—and only as an undergraduate, at that—pretty much assured my immediate transfer to the control systems department. Fortunately, I was a quick study.
I did my work on an 8080-based computer with 8 kB of memory (that's kilobyes, for those unfamiliar with the abbreviation). The machine had no operating system, as such, and no random disk access. Files had to fit onto a 90 kB floppy disk. To make sure our hardware–software combination would work in the real world, we set all the hardware up in our office area, together with a set of "simulator" panels (shown in the first photo below) that had a toggle switch mounted for every input and a light bulb for every output to be installed at the power plant.
All those inputs and outputs were connected to a set of programmable controllers (made by Gould, if memory serves), and it took me a while to develop and debug the ladder logic controlling how inputs affected outputs (and vice versa). Once that was done, I developed routines that enabled the computer to "talk" to the controllers and obtain snapshots of I/O status, which then picked out the relevant bits and displayed them on the screen, as shown below.
Operator interaction took place using an early form of touchscreen technology in which the display was surrounded by light-emitting diodes (LEDs) and corresponding optical sensors. Touching the screen of a CRT equipped with this kind of screen interrupted two light paths (one horizontal, the other vertical) showing the location of the "touch." Touching various displayed components changed their state, depending on the component (e.g., touching a pump that was off turned the pump on, and vice versa).
That was the state of the art, back in the early 1980s. These days, most of us walk around with more computing power in our telephones than existed on the planet at about the time I did this work.
Like the man said...
That all changed in the summer of 1979 after an interview and offer of employment from Reynolds, Smith & Hills, Inc., Architects–Engineers–Planners, of Jacksonville, Florida. My interview came about through the efforts of my half-brother Steve, who had set out to prove that my self-defeating attitude of "nobody'd be interested in me as a potential engineer given my track record" was a lot of bunk.
Originally, the company had the idea of bringing me on board as a mechanical engineer, but that only lasted until the company's "advanced energy" division got wind of the fact that I could string words together in a lucid and persuasive manner—which distinguished me from most engineers—and so I spent a couple of years helping out with proposals and feasibility studies for everything from solar power to energy retrofits for buildings to ethanol production. Then the company's control systems department landed a job that required software to be written to integrate various hardware components into a materials handling system for a coal-fired power plant.
The fact that I was the only engineer in the company who had ever programmed in BASIC—and only as an undergraduate, at that—pretty much assured my immediate transfer to the control systems department. Fortunately, I was a quick study.
I did my work on an 8080-based computer with 8 kB of memory (that's kilobyes, for those unfamiliar with the abbreviation). The machine had no operating system, as such, and no random disk access. Files had to fit onto a 90 kB floppy disk. To make sure our hardware–software combination would work in the real world, we set all the hardware up in our office area, together with a set of "simulator" panels (shown in the first photo below) that had a toggle switch mounted for every input and a light bulb for every output to be installed at the power plant.
All those inputs and outputs were connected to a set of programmable controllers (made by Gould, if memory serves), and it took me a while to develop and debug the ladder logic controlling how inputs affected outputs (and vice versa). Once that was done, I developed routines that enabled the computer to "talk" to the controllers and obtain snapshots of I/O status, which then picked out the relevant bits and displayed them on the screen, as shown below.
Operator interaction took place using an early form of touchscreen technology in which the display was surrounded by light-emitting diodes (LEDs) and corresponding optical sensors. Touching the screen of a CRT equipped with this kind of screen interrupted two light paths (one horizontal, the other vertical) showing the location of the "touch." Touching various displayed components changed their state, depending on the component (e.g., touching a pump that was off turned the pump on, and vice versa).
That was the state of the art, back in the early 1980s. These days, most of us walk around with more computing power in our telephones than existed on the planet at about the time I did this work.
Like the man said...
