This Sherlock Ohms posting is a great example of the danger of assuming. He asked for 10 feet of cable, he was given a length of cable. But he didn't check to make sure he actually received 10 feet of cable.
Trust, but Verify. This example belongs in a collection of wise tales and parables handed to each new measurement apprentice. When it comes to measurements, folks should not be offended when their data is double-checked but instead remind the receiver to re-check their measurements and calculations. Imagine how much unnecessary effort and research was avoided by checking the initial value...
I'm not too sure how you get around this one. One company I consult for has work instructions AND a check list AND employee training AND there are still errors made relative to measurements. I will say this, we use a software package called MiniTab to perform Six Sigma calculations. With this software package, it is fairly easy to discover measurement errors. We always try to run a sample size of at least 30 pieces PER shift. Sometimes when we are checking against critical-to-quality specifications, we collect samples every hour for analysis. Another important factor, we always measure and calculate for gage R&R and factor that into analyzing the final data.
Many years ago a VP of engineering I knew had a sign on his office wall: "Measure twice, cut once." I memorized that one, and it's been helpful for avoiding mistakes like this one. Of course, it's easier to observe this rule when it's just you doing the measuring both times. I like the "Trust, but verify" maxim. That maxim completes the first one.
The problem came from assuming that the directions had been accurately followed. I had a similar problem when a new tech grabbed a stack of incomplete circuit boards from a stack and then was challenging me that my design was no good because the boards didn't work. I wound up posting a general memo that "designs will not perform as required unless they are built as designed." The lab manager challenged me as to what it was about, and I explained to him that it meant that the circuit boards would not work correctly unless all of the parts were installed. He agreed that was a correct assertion, and cautioned that tech to only use circuit boards that had passed the final inspection. The non-calibrated boards seldom met the requirements, and the incomplete boards never met them.
William: Your story really touches a raw nerve with me. I have lost count of how many times the "New Guy" finds errors where none exist merely because he wanted to impress somebody higher up with just how much he knows. Not all new guys, but a significant number.
At the Design News webinar on June 27, learn all about aluminum extrusion: designing the right shape so it costs the least, is simplest to manufacture, and best fits the application's structural requirements.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This radio show will show what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.