Advertisement

Blog

Making the Grade in Industrial Design

As every developer knows, there are the paper specifications for a product design, and then there are the real requirements. The paper specs are dry, bland, and rigidly numeric, making them great as metrics but a poor guide to creating a design that will function properly in real-world conditions. Only experience can show you the real requirements a design must meet.

Take the specification categories to which we assign electronics parts and systems: commercial, industrial, and military grade. On paper, these relate to metrics such as shock and vibration tolerance, operating environment temperature, and voltage isolation capacity. In reality, such metrics represent only a small part of what a design must endure. Of far greater significance are the characteristics of the equipment's human operator.

Early in my working life, when I was a design engineer at JHU/APL, I got a lot of experience working on military-grade equipment designs. Beyond the typical MIL-SPEC requirements of wide operating temperature ranges and substantial shock and vibration resistance, the designs had to tolerate military (in particular, US Navy) personnel.

Most of the designs were not for mission- or life-critical systems, so they needed to tolerate rapid setup and breakdown, insufficient (or no) time spent reading operational manuals, and an attitude toward the equipment that often ranged from indifference (at best) to outright resentment. It's easy to understand why; my equipment mostly represented an additional burden on already overworked users in potentially life-threatening situations.

I learned to do things like making sure no two external connectors were identical (to avoid misconnected cables) and making sure that the housing could double as a stepstool. My project managers further trained me in the “two dumb thing” rule: My designs couldn't fail unless the user made two simultaneous dumb mistakes, like spilling coffee on the interior while holding the circuit breaker in a closed position when the chassis was powered and open for maintenance. In short, I needed to make my equipment as fault tolerant and “sailor-proof” as possible.

Of course, it's impossible to imagine all the possible contingencies, much less design for them. And it's the height of folly to imagine that you have done so. One of my co-workers went shipboard with a design he proudly described to a chief petty officer as sailor-proof. The chief took the equipment, examined it carefully, and then threw it over the railing into the ocean.

“Doesn't float,” he observed. “Not sailor-proof.”

Consumer-grade design is a lot more tolerant. For the most part, consumers figure that, if they break something while doing anything much beyond normal use of the equipment, it's their fault and thus their loss. There are exceptions, of course. The recent buzz about the iPhone 6 bending when people sat with it in their back pocket shows that consumers expect at least some degree of latitude in what constitutes “normal” use.

(Image: YouTube)

(Image: YouTube)

Industrial-grade equipment falls somewhere between those extremes. The equipment must be fairly rugged and tolerant of human abuse, but mostly not the type of abuse the military can inflict. Where exactly industrial equipment is within that range, though, only you who have industrial equipment design experience can really say.

To read the rest of this article, visit EBN sister site EETimes.

7 comments on “Making the Grade in Industrial Design

  1. Daniel
    November 25, 2014

    “The paper specs are dry, bland, and rigidly numeric, making them great as metrics but a poor guide to creating a design that will function properly in real-world conditions. Only experience can show you the real requirements a design must meet.”

    Rich, welcome to our community. You are right about the specs in design. The specs are based on best assumptions in theoretical and in practical or real-world condition it won't comes more than 80-90%. Designers has to account this while selecting components and making specs.

  2. Rich Quinnell
    November 26, 2014

    Thanks for the welcome. Agreed, the plain, dry specs are best-guess assumptions and attempts to quantify the relevant parameters. And numbers are what you need to use to test against. It sometimes helps, though, to put these things in more practical terms as a way of envisioning what you are trying to accomplish beyond mere numbers. It's fine to tell someone their box needs to tolerate 5000 N/m2 of static crushing force, but they get a better feel for the requirement if you tell them it needs to be able to have an elephant stand on it without collapsing.

  3. Daniel
    November 26, 2014

    “It's fine to tell someone their box needs to tolerate 5000 N/m2 of static crushing force, but they get a better feel for the requirement if you tell them it needs to be able to have an elephant stand on it without collapsing.”

    Rich, yes I think it's better to specify in terms of common terms. Elephant weight/strength can understand by everyone very easily instead of measuring its in N/m2  

  4. ahdand
    November 27, 2014

    @Jacob: Yes that way its easy to refer if you get lost on some formula. 

  5. FLYINGSCOT
    November 28, 2014

    Haha…..I loved hearing about the good old days designing kit that was tolerant to real abuse.  I especially liked the 2 dumb rule.  I guess there is no substitute for years of experience designing mil grade kit.

  6. Daniel
    December 1, 2014

    ” Yes that way its easy to refer if you get lost on some formula. “

    Nimantha, it's not about any formula or equation. Just equating with a very commonly knowing object; so that easy to compare than any other metrics. 

  7. Rich Quinnell
    December 1, 2014

    Thanks, Flyingscot. Sounds like you have been there yourself. Any good stories to share?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.