I was fresh out of university and worked as a development chemist at an industrial printing company. One of my first projects was to put in to service a thermogravimetric analysis (TGA) instrument. Basically, it’s an analytical balance with a platinum crucible on top that can be slowly heated to a very high temperature. As the crucible is heated, the weight is monitored over long periods of time.
As part of my testing, I would leave the instrument to run overnight with nothing in it, to monitor the baseline. These tests showed some interesting but confusing results. The baseline had a very periodic bounce of a few milligrams – not massive but outside the tolerable range to put the instrument in to service.
I went to my boss with the results and campaigned for a real balance table; the table it currently sat on was a glorified library table, not one designed for an analytical balance. This was shot down as it was too expensive, and anyway, the signal was so periodic, it had to be electronic, I was told. So after calling in the instrument service rep, who gave the instrument a clean bill of health, I was back to scratching my head.
It just so happened that some months before, I had been watching a TV show on the history of skyscrapers. On the show, they talked about how, as buildings got taller and larger they tended to sway at their own harmonic frequency. This was well below our ability to perceive at 2-5 Hertz but was nonetheless, a known phenomenon. I studied the instrument baseline again and low and behold, it came in at a steady 3-4 Hz. I proposed this cause to my boss and again, petitioned for a blance table to dampen out the oscillation with the same result. It was out of the realm of my boss’s experience and therefore, quickly dismissed.
So I developed my own solution. Raiding the laboratory supply closet, I found 4 large rubber bungs, and a trip to a building supply store got me a large concrete paving slab. From these, I created a rudimentary balance table. Much to my boss’s amazement, this home built rig cut out most of the oscillation in the baseline for around $10. Discussions with the head of the maintenance group led to a probable explanation: although the building was no sky scraper, it did have a very large air conditioning stack at one end that had caused oscillation problems in the past.
So the moral of the story is: when trying to solve difficult problems that seem to have no logical cause, do not be too quick to discount the unlikely. Just because it falls outside your sphere of experience, does not mean it is impossible –just unlikely.