MechEng is confusing

Article By : Michael Dunn

Is it foolhardy to attempt tasks outside our domain? Know anything about screws?

Designing a six-pole 2 dB Chebyshev lowpass, coding an HDL block, or writing some cycle-pinching real-time assembler—all tasks I can confidently execute. Tightening a screw just so? Not so much.

How far we can stretch our skillsets?

On the one hand, EEs have, in large companies at least, become very specialized. Members of IC design teams tend to focus on very narrow pieces of the puzzle. Cellphone companies have desense engineers, whose sole function is to make sure the RF receivers aren’t swamped (desensitized) by all the other electronics packed into close proximity.

On the other hand, in smaller companies, we sometimes need to wear many hats, including ones we’ve never worn before, and which may not even fit at first.

A few jobs ago, I had to develop fairly strong expertise in thermal design, because we were making chassis that were tightly packed with PC motherboards and HDDs, which had to be quiet, and operate in places like Denver Colorado, and even higher up. Solving these problems with reasonable confidence involved lots of research and lots of measurements, and we fortunately had enough budget for things like airflow & temperature instrumentation, and even a good environmental chamber. That was a productive and rewarding skillset stretch.

But my current project is closer to home. The floor in our “new” 50-something house is wavier and slopier than one might hope, and one unsupported inside corner in particular got me thinking about jack posts.

Can I stretch my skillset to handle a bit of simple MechEng?

The basement already has a half-dozen original posts – mostly down the middle of the house’s long dimension. Perhaps I can adjust some of those to improve floor flatness. But what about a new support?

I picked up a standard-issue post from a certain big-box home store. Its maximum load is 41 kN. While that’s useful information, the manufacturer neglected to install a load cell in the post. How is anyone to translate that to, say, wrench torque?

After polling my colleagues – to mostly blank stares, I hit the web as a last resort, and found four calculator pages!

Well…no surprise – these pages behave differently, and don’t all accept the same input parameters. The last two seemed best suited. With input parameters of:

  • 3 mm thread pitch
  • 22 cm lever (wrench)
  • 3 cm diameter
  • 41 kN maximum load

the calculators returned 89 N, and 25 N·m (or 25 N·m / 0.22 m = 114 N). Considering the two calcs used a different subset of input parameters, those two results are pretty close and provide some confidence I’m in the ballpark. If we take 98 N as our result, that translates to 10 kg of force on my wrench (98 N / 9.8 m·s-2).

The final problem then is measuring the force on my wrench. I do have a socket-torque wrench, but that wasn’t going to help. What I needed was some sort of scale. Something like:


I also ordered another one, just to be safe, but it’s yet to arrive. My main concern was that the weight wouldn’t display until it was stable, and that obviously wasn’t going to happen here. Fortunately, the reading is continuously updated.

Pulling about 31 N (3.17 kg) on my wrench.

Of course, the question of accuracy also arose. The scale is 0.6% low compared to my Friden postal scale at 5 kg. Not too shabby.

There’s always fear of unknown unknowns, and of a lack of assumed knowledge, when stretching beyond one’s sphere. Did I get it right? I think so, to a reasonable degree of accuracy, and the worst that will happen if I didn’t is a broken jack (hopefully no shrapnel as it fails). What experiences have you had outside of your comfort zone?


Related articles:


Michael Dunn is Editor in Chief at EDN with several decades of electronic design experience in various areas.

Leave a comment