How far have we come with quality?

My first manufacturing job was back in 1973. It was at a dark and dingy shop that produced fittings on screw machines and turret lathes out of different materials, such as brass and copper, but I couldn’t even tell you what industry the parts were for.

The tolerances on most dimensions were about ±0.015 ". The parts were checked with GO/NO-GO gages for the threads and calipers for the other dimensions. The parts were only spot-checked during production and then shipped.

Several years later, I worked as an inspector for another company. A huge stepping stone at that facility was its use of NC mills and lathes. There, I had access to outside and inside micrometers, profilometers to check surface finishes, pi-tapes for checking diameters, an optical comparator and various other inspection equipment. The shop routinely achieved tolerances of ±0.005 " with an occasional ±0.002 ". Previously, that shop only had a final inspection program, but I implemented in-process inspections, which saved untold machining hours and significantly reduced scrap.

Eight years later, I was an applications engineer for a builder of multiple-axis turning centers that could machine to tenths and incorporated in-process inspection capabilities using touch probes. The primary customers were automotive and marine engine builders. At that time, we could also offer customers automated offline probe systems that checked dimensions and automatically updated cutter compensation on the turning center for whatever tool was wearing. There was even a broken tool detector.

Fast forward a few years and I’m employed at a job shop that purchased a large coordinate measuring machine to verify dies being machined. We thought the CMM would help improve part quality and in a way it did. But what it really showed was how close the parts were to being out of tolerance. Not just new parts, but previously made parts we thought were good. Don’t get me wrong, the parts weren’t bad—they just weren’t as good as we thought. Having the CMM forced us to make better parts.

Now I work for a shop that produces aircraft engine and power generation components. We have several large CMMs, a couple of portable CMMs and a variety of other inspection devices to support our quest for quality. Many of the machines have built-in touch probes to verify part quality. Not only do we have in-process and final inspection personnel, but also quality engineers who evaluate and document data for customers. We have to provide 100 percent inspection for many part dimensions, and some parts are so particular they require a cleanliness test.

As my experience shows, improvements in inspection equipment and techniques have evolved dramatically over the years.

However, we still have the human factor to contend with. Despite our best efforts, human error can still creep into any process. With fully automated machines, which are great for production runs, you can virtually eliminate bad parts. Once a machine is programmed and tooled correctly, it can run unattended and repeatedly make good parts.

When I was an inspector, there was a machinist, as good as he was, who occasionally made bad parts. He always had a good part on his bench ready for inspection, but when he made bad parts, he would bury them in the pile of completed parts, hoping to get them past the inspector. Fortunately, I caught them most of the time.

I don’t know if it’s a result of an individual’s poor work ethic, lack of training or what, but until that problem can be corrected, quality will never be a given. It will always be something to strive for. CTE