Magic
Shhhhh… don’t tell anyone! 
<curtsies>
I usta be a technician (avionics) too, and I’ve had the chance to run a lathe, a mill, and a drill press, tho I’d not DARE call myself a machinist.
Not sure what my point is, but I think it had something to do with learning all you can and teamwork and cooperation and we are the world. OK, time for a group gro… er, hug.
I think the OP has been answered, but FWIW…
There are different type of engineers…electrical, mechanical, chemical, environmental, etc.
And then there are subsets (mechanical = machine design, fluid dynamics/aerodynamics, etc.)
Each of these disciplines have slightly different design mindsets. Some are nuts-and-bolts, some are process-related, some are more conceptual.
Personally, I’m an engineer who does more consulting than designing.
One of my best friends majored in agricultural engineering at Virginia Tech. He described a manchine design problem that I found intriguing - the class had to design a chicken vac.
In order to gather the most chickens in the least amount of time its more efficient to just suck them up into a hose and deposit them into crates. The problem is that chickens are notoriously variable in shape and size, are delicate (after a fashion), and are not particularly aerodynamic. He said one of the hardest parts was figuring out how strong to make the motor on the machine so that it didn’t propell the chickens into the crate with so much force that they died.
Well, I’m hoping to join the ranks of the magical engineers soon. I’m majoring in Electrical Engineering right now (senior). Concentrating in DSP. Basically, I’ve done virtually NO circuit design (one lab class with analog circuits, and one digital), while others in my major with different concentrations have done 5 or 6 courses of circuit work through 3 years. My stuff so far seems to be almost exclusively theoretical and mathematical. I hope you learn a lot about design on the job, because right now, I couldn’t design a damn thing. I could give you design specs for some communications stuff, or do matrix algebra, FFTs, DFTs, DTFTs, Laplace transforms, Z-transforms, etc, but nothing much in the way of actual design.
Jman
Or they make things that fail:
Leaning Tower of Pisa
Titanic
Apollo “1” & 13
Challenger
Chernobyl
70’s Chevy Truck fuel tanks
90’s Chrysler Minivan doors
Trial and error is still alive and well… Reminds me of the scene from Fight Club:
Narrator: My job is to apply the formula. Take the number of vehicles, A, times the probable rate of failure, B, times the cost of the average out-of-court settlement, C. A times B times C equals X. If X is less than the cost of doing a recall, we don’t do one.
Woman: Are there a lot of these kinds of accidents?
Narrator: You wouldn’t believe.
Woman: What auto company did you say you worked for?
Narrator: A major one.
There are three principal ways to lose money: wine, women, and engineers.
While the first two are more pleasant, the third is by far the more certain.
– Baron Rothschild, ca. 1800
Not that I necessarily agree, but it made me chuckle.
I’m a real-life design engineer (EE). Not to boast, but I’ve designed many electronics systems (w/ discrete analog, digital, and hybrid circuitry), designed multi-layer PCB’s, programmed microcontrollers, written data acquisition applications, etc. It’s my job.
So how do you design something?
You begin by determining the system requirements. What are the inputs? What are the outputs? What are the power requirements? Size? Volume? Weight? Environmental considerations? Human factors considerations? The list goes on and on…
After you’ve nailed down the requirements, you formally begin the design phase. This really isn’t that difficult. For the most part, it is just a matter of designing or obtaining “building blocks,” where each building block has input and output parameters. Once the building blocks are verified to be working properly on an individual basis, you connect them together and verify & validate the system as a whole.
It’s understandable Mr. Blue Sky is somewhat in awe after looking at the innards of a VCR. But it must be understood that a VCR (or anything else) is not “one big circuit,” but hundreds of small, manageable, interconnected circuits that have well-defined parameters.
I’d like to chime in with the “hail to the technicians and machinists” crowd. I’m an engineer, but I was a tech first and still mostly think like one. Never underestimate the contributions these people make. I hate to design anything without the input of the people who are actually going to put it together and maintain it. I look upon good machinists with something very like awe.
As a programmer in a large-company commercial setting, I think I can say that my work falls in between that of engineers and that of technicians, in its essentials. We
have to design stuff, we usually have to adapt or tweak an
earlier solution, and we also end up doing most of the building and testing. Unfortunately due to time constraints
we can’t always produce the most elegant design possible.
Interesting question, and cool responses.
If you’re interested in this sort of thing, you might check out the book The Soul of a New Machine by Tracy Kidder. It’s about a team of engineers designing a brand-new mainframe computer, essentially from scratch, back in 1980. The time period means a lot of the specific information is dated, but as a portrait of a bunch of techies figuring out how to build something, it’s very informative, and a hell of a lot of fun. Basically, it’s a book-length real-life answer to the OP’s question.
FairyChatMom
Finally, someone who has been there.
Welcome to the board.
I’ll look foreward to your posts.
Design experience during school would be helpful (find as much as you can). But in general, it seems like an undergraduate degree just gives you the background (familiarity) and the tools (mindset, references) to be an engineer. Once you get a job, you’ll be working with more senior engineers. They’ll task you with small components at first, and as you figure out that particular business’ design, you’ll move to more complex design work. On-the-job training is not uncommon.
The big problem is determining exactly what is required as opposed to doing what you want to do.
There is a syndrome so common that it has been dignified with a name - creeping featurism - which has been responsible for many an overrun.
Its where you put in facilities because you can but this often winds up being unnecassary, costly, unreliable.
Sometimes a customer will try to give a requirement that demands technology that only exists in the realms of science fiction.
Another huge problem is changing the requirements well into the design process. This happens in military projects where an innovation in another field may compromise the viability of the original idea.
If you work in a commercial field you might be 2 years into a major project and a competitor beats you to the punch so you try to add features to make your product more attractive.Cars are a classic case study.
Last of all there is knowing when to compromise. You may have done the calculations and determined that the most efficient design uses a part such as a bearing of a singular dimension but this might put the cost up hugely so you use the nearest available which you can buy cheaply off the shelf rather than a custom made item.
Above all a good design engineer listens.
Speak of science fiction, here’s a site on StereoLithography (pure fluid magic)
Yep - listen to the customer, listen to the boss, listen to coworkers, listen to artisans… you just never know where you might get exactly the information you need.
Wow You Guys
I’m impressed.
I’ve been posting for 8 months now and didn’t realize how many technical people were on board.
I have been out of tech school for 30 years now except for picking up some classes that didn’t amount to anything.
In my day an electronics tech worked to the component level.
The government used modules in some of the navy gear I worked on, but they were pretty much still repairable. I worked on a lot of service and repair contracts and we produced new gear for Great Britain France Germany Australia and Japan.
The first computer modules I worked on was the GPATS system. General Purpose Auto Test System. Transistor Flip/Flops.IIRC most problems consisted of mismarked transistors and diodes and shorted diodes.
This was in the day of $60.00 AM transistor radios.
Troubleshooting to the component level for most of us ended with the major TV companies going to Japan.Motorola was the first company, I believe, to begin using modules. That almost ended troubleshooting as I knew it.
By the way no one in my electronics class had a calculator.They only add subtracted multiplied and divided anyway.
Well enough of ancient history.
Thought you’d like to know how it was.
justwannano, I still troubleshoot TVs, VCRs, and audio equipment at the component level (when I go back to help out at the family repair shop). Manufacturers are moving back away from the modular approach–even a lot of TV tuners are discrete these days.
Balance
You mean there might be a use for the old tube caddy tuner sub and picture tube rejuvinator that I have in the basement?BIG GRIN
Engineers don’t build stuff, workers do. Engineers design stuff. That’s their job, and I’m not saying it’s easy.
Architects don’t build anything, either.
If you don’t believe me, watch a bridge going up. The guy/gal in the white hard hat, waving his/her arms around, is the engineer.
The sweaty people, the ones with the tools, are building the bridge.
Nitpick? Yeah.
Peace,
mangeorge (Worker}