标签: windmills

MEMS-based designs can result to some strange and wonderful things. While thumbing through a recent issue of Wireless Design and Development (yes, it was print), I saw the story " Micro-Windmills: From Lab to Market " on MEMS devices which could be used to generate power from passing airflow, whether from the  convection currents of circuit heat, or general exposure to their surroundings, figure 1 . I did some further digging online and found more details on the work at the relevant press-release page of the University of Texas at Arlington, where much of the work is being done in conjunction with support and production from WinMEMS Technologies , a Taiwan-based company. Figure 1: The nickel-based MEMS micro-windmill can be bulk-manufactured on a waver using processes similar to conventional silicon-based ICs (courtesy University of Texas).   Building a MEMS windmill does seem pretty impressive, in part because you need a rotating joint, Figure 2 . While I have seen micro-gears and gear trains in experimental MEMS devices, and so we know this sort of mechanical structure can be fabricated, is still a real challenge. Figure 2: A microphotograph of the windmill shows its apparent simplicity and also its complexity (from WinMEMS Technologies Co,. Ltd.)   Several points in the story caught my attention. First, I assumed they used silicon as a base material, as nearly all MEMS devices do. It turns out that silicon doesn't work well here, because it is brittle despite its strength; instead, the team used nickel alloys. Also, at the tiny dimensions of these moving blades, the usually ignorable effects of factors such as moisture become an issue; in this case, the water molecules start to become glue-like and create static friction (stiction), so a special lubricant may be needed at the rotary joint Further, the article noted that there are very few development tools and established process techniques for fabricating nickel-based MEMS devices, so a lot of their progress was "learn as you go", based on their technical knowledge and dozen-plus years of experience with nickel alloys, etching, and plating. They wanted to make as much use as possible of the general process flow and wafer-scale manufacturing which silicon-based MEMS uses, to leverage the technology, production and cost curves. The work they have done on MEMS micro-windmills is absolutely impressive, no doubt of it. They envision a "field" of such micro-windmills implemented on a single die, which could produce "free" power -- the epitome of energy harvesting. I wonder about a few issues, though. I suspect that in order to get the attention of the broader audience, the press coverage touted applications such as recharging of cell phones, which is clearly a "hot button" angle to use. Yet while press releases talks about this micro-windmill as if the total energy-harvesting function is available, my closer reading of the reports indicates that they have the windmill part working, but there is no explanation of how actually they use it to generate electrical power from the turning motion. Is the next step placing micro-magnets on the blades and micro-coils right next to them? Other questions: what level of output power could you expect from these micro-windmills, assuming 100% conversion efficiency: is it microwatts, picowatts, femtowatts, or less? Since the power output and efficiency of windmills are each roughly proportional to the square of the air velocity, a micro-windmill might be have very low output at best, even before conversion losses. With all these tiny windmills in close proximity, there's also the well-known "shadowing" effect to consider: will the air flow to each be obstructed or turbulent, rather than laminar? However practical or realistic these micro-windmills will be, this is another dramatic example of how MEMS technology is opening up radically new approaches to implementation of existing devices as well as new concepts, from accelerometers to microphones, timing oscillators, and more. With silicon and even non-silicon MEMS-based components at one end of the design-resource continuum, and additive manufacturing/rapid prototyping at the other, there's lots of room for creativity. What new technologies do you see opening up radically new product-design opportunities?