Take a look at some of the clever systems for staff has developed to evaluate everything from elliptical trainers to kitchen drawers.
I have a fantasy job. It’s not that I don’t like what I do but realized recently that there’s something better. No, it’s not doing quality control at the Godiva factory or captaining a snorkeling boat so that I can spend my days out on the water in the Caribbean. No, I want to work in the Consumer Reports test labs. Those folks have the best time.
Now, I’m not talking about the ones who test drive lawnmowers for hours to see which best cut the grass or repeatedly chop ice in blenders to see whether they can survive several hours of punishment. (Then again, the blender test might not be so bad, if you could add a little bit of margarita mix). No, I don't want to supervise hummus taste tests or evaluate mosquito repellent. I want to work on projects where they find ways to break stuff.
I swear, I had a happy childhood.
Seriously, though, look at their protocol for testing door locks. It includes an automated sledgehammer that slams down to try to break the lock and a ram that plows into the lock to mimic a burglar kicking the door in. And these guys get paid to do this. What a racket.
Okay, I'll admit, the job has its challenges. Mind you, some of the testing protocols are pretty straightforward—installing thermocouples throughout refrigerators to monitor heat, for example, or performing a drip test on coupons of painted metal to have been exposed to the elements for weeks to evaluate how well car wax protected the finish. Those represent good basic science, but nothing interesting or unusual. How about their cabinet test system, though? They fill the drawers with gravel and then use a robotic unit to open and close them—30,000 times over a day and a half. Another ram mimics the effects of kicking a door closed. And here my mom always yelled at me when I did that.
Testing strollers by taking them through an obstacle course is a worthy effort, but not particularly exciting. In the crib testing video, though, they bring out the hardware to see just how easy it is to tear the thing apart.
What I find most intriguing about this type of testing is the process of evolving the best way to quantify performance. In the case of evaluating garbage disposals, the team took months to develop their test system. “The Digester” features transparent pipes, nested filters, and a roll-tap machine to discover just how effectively covers disposals can grind up bones. To test elliptical exercise machines, they evolved a system that combines hydraulic pistoning and mechanical arms to monitor the consistency of resistance presented by the machine.
Based on what I’ve seen, the Consumer Reports gang develops some pretty clever and interesting systems to quantify performance for a surprising array of products. I can think of all sorts of easy-to-solve test cases—opening and closing windows and doors, testing TV brightness, pressing the buttons on a phone, etc. How about trickier tasks like testing running shoes, though? I have visions of a Rube Goldberg-esque machine featuring shoes mounted soles out along the circumference of a wheel set atop a treadmill to mimic running a few hundred miles over the course of a month.
What do you think would be the hardest product to test in a quantifiable fashion? How would you go about solving the problem?
These are mechanical test analogous to testing electronics through the temperature and humidity range, using shake tables as well as EMI, static and discharge tests after the usual bench tests for test waveforms and power supply variations.
The area where current electronic design testing falls flat on its face is the testing of firmware and software. Reliance is given to Beta testing by prospective customers, not exactly ideal.