No tags yet.
Years back, I was in charge of supplier and manufacturing quality for one of the large US-based companies designing and selling complex electromechanical home products. One of these products was a sophisticated vacuum cleaner, and we were about to order a large production run, with components coming from different suppliers. One of these components was a drive belt that provided fine control over the vacuum nozzle-head. The belt was transferring power from an electric motor to the nozzle-head’s brush-roll which was turning at high RPM. We decided to choose a new supplier that we haven't used before.
From our early prototype testing and from similar products in our portfolio, we knew that a precise and accurate length of the drive belt was critical. If the belt was too long, the tension would be too low and the belt would slip. This in turn would lead to poor nozzle performance, a noisy operation, and premature wear and tear. The mechanics of this is simple: a belt that is too long will be loose, and can slip against the gear teeth and over time, this can degrade the belt to the point that it doesn’t work at all. On the other hand, if the belt is too short, the tension would be too high causing too much stress on the gears. This too impacts long-term reliability, but also creates a high-pitch sound disturbing to the human ear.
As part of the design package delivered to this supplier, detailed drawings were sent, which included the required belt length and tolerances. The belt specifications were 150mm with a tolerance of ±10mm. To ensure they were producing the belt to spec, and since it was a new supplier, I insisted that we obtain a First Article Inspection (FAI): they needed to show us the first 30 samples for evaluation.
So they had sent us the samples, and the belts were within spec. We were ready to roll, right?