Power control, cheap and good
When my basic plug-in heating pad stopped working after a few years and many hours of use, I was about to throw it out when I realized it perhaps deserved a better "sendoff." So I opened the controller just to see what was inside; I rationalized that I at least owed it that token of respect. Once again, I was reminded of how far and how effective basic power control has come, and for so little money.
This pad has three heat settings and a 2-hour automatic shutoff timer so you don't inadvertently leave it on for too long. (I am pretty sure the unit won't overheat even at 100% use; the shut-off is really an "energy conscious" feature from what I can see.) All the electronic circuitry is on a small PC board (see photos).
And here's the impressive part: you can get all this for under $20 at a local store. When you think about it, that means the price to the store is likely somewhere between $10 and $15, so the BOM and manufacturing costs must even lower, probably in the $5 to $10 range. Yet that modest cost has to cover the loaded PC board including a few MOSFETs and a basic microcontroller, plastic enclosure, an AC cord and the wires between the controller and the heating pad, and the heating pad unit itself. That's a lot of stuff for short money.
I have conflicted feelings about this situation. On one side, I am proud and impressed with what the aggregation of so many engineering disciplines - component, process, design, fabrication, and assembly - has achieved to enable this relatively modest but still challenging product. On the other side, I am once again discouraged: because we have made it look so easy, and the average consumer thinks it's all no big deal; they "expect" this sort of modest miracle to happen and are not at all impressed nor appreciative.
We have seen this over and over as we produce amazing, technology-packed devices, which took years of work, struggle, and uncountable costs, and which sell for a few dollars - yet people still moan, "it costs too much, doesn't do enough, etc." I also know that complaining about this is not going to help as long as Hollywood-type celebrities are given more attention, respect, and stature than scientists and engineers, so I will let it go, for now.
Still, the low cost and decent performance of this now-defunct heating pad and its controller made me wonder: could a clever engineer who needs a basic heater or power control adapt the controller from this $20 pad, perhaps with a different heating-element configuration? After all, what you have here is a fairly decent open-loop control block. If that's all you need, you're all set. And if you need some precision in the temperature value, could you then add a basic feedback loop using a sensor and have closed-loop control on the cheap?
Have you ever taken a standard, low-cost consumer device and adapted it for your project, to save development time or cost? Was that a good idea, or was it only what seemed like a good idea at first but actually more headache than it was worth, in the final analysis?