The problem of moisture absorbtion is often nonlinear, which is why just baking the deteriorated units would not result in another five years of service. What probably happened is that after a bit of exposure the more moisture resistant portion deteriorated and the more porous material was exposed, at which point the process went much faster. So baking and re-coating could have been a solution, but the cost would probably be similar to installing a new antenna. Sometimes repairs are completely possible but not economical to do. Given what I know about high labor costs on railroads, I would presume that was the case here.
Yes, it was good detective work, although great detective work would have been to stop and think for a while and come with a suspect at the beginning.
If the parts absorbed moisture, why wouldn't baking them drive the moisture out and give you another 5 years of operation? Obviously the design was faulty, but cycling the devices periodically through a drying operation could potentially save having to replace them unless they were very inexpensive and the cost of handling exceeded the replacement cost.
PTFE (DuPont "Teflon") is hydrophobic (water repelling). PET (material of the aforementioned housing) is hygroscopic (water absorbing). I know long term testing is sometimes sacrificed in order to get product to market. In this case the product worked for 5 years before failure. I would imagine, realistically, that when the problem was discovered, the solution would be to change the material to something better suited for long term outdoor use.
Sometimes the more expensive material is the cheaper (or should I say, more economical) and better choice in the long run.
I think it would have been a combination of incomplete specification & manufacturing (Moulding process & cleanness). The materials used were those specified by the original European design. The manufacture of this product was taken over by another company shortly after so there was no further trend analysis data to see if the problem continued to other batch lots.
Are they robots or androids? We're not exactly sure. Each talking, gesturing Geminoid looks exactly like a real individual, starting with their creator, professor Hiroshi Ishiguro of Osaka University in Japan.
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 discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.