Next Challenge for Low Power Drives

It has become an undisputed truth that brushless DC motors see an ever-increasing use in low power drives, such as pumps, fans, or small motion control, as they should - the advantages (such as efficiency, size, features) are strongly outweighing the disadvantages (electronic controls needed). Motors with brushes like the example pictured here are facing extinction! Engineers trying to implement this have been faced with some challenges. First of all, nobody wants to pay more - the expectations are such that the new features should effectively come for free. Looking at system cost level, a smaller motor (especially given current raw material prices) can compensate for higher electronics cost to some extent. Another challenge is the "traditional" way of implementing electronics, all on one PCB integrated with the host microcontroller, causing EMI issues and adding cabling complexity, since today in many cases sensors are used to detect the rotor position. Plus, the heat generation of the power electronics section (including a larger power supply) is at odds with the signal processing on the same board. The next challenge for low power drives certainly is to put the smart power module controlling the motor into the motor itself, effectively turning a dumb mass of metal into a smart, agile part of the system that then can be controlled with digital signals effectively. Easier said than done, however - as can be seen from existing solutions on the market, integrating power electronics with a motor is a piece of engineering art, given today's targets for power density, reliability, and performance. The first issue clearly is centred around power density. BLDC motors are already smaller than their predecessors of the same power rating, so the electronics to be added into the motor must be small as well, while at the same time meeting creepage and clearance norms, with more high-power signals that need to be routed, and providing good thermals. And, a small motor used e.g. in a refrigerator will see many thermal cycles in its lifetime, so significant reliability aspects must be considered. The second issue is centred around electromagnetic emissions. With 50Hz that was not a problem, but with modern switchmode electronics operating at 50kHz or higher (in the case of using MOSFETs for lower power drives), significant electromagnetic emissions are being generated. In many larger systems, like fans or pumps, this is not a big problem, since the motor case can also be used as a shield; these emissions can however cause malfunctions inside the system, forcing the designer to add larger design margins, particularly in the case of discrete solutions. The design margins needed are leading directly to the third issue, achieving a good efficiency. There is no point in moving to a "fancier" motor in many cases if the efficiency that can be achieved is not good. And if larger margins on the power devices, input capacitor or gate drive circuit are needed to yield stable and dependable performance over a larger load and component variation range, resulting in lower efficiency, the cost of operating this motor increases, and that is in nobody's interest. Using smart power modules in low power drives are a suitable way out of these three issues. Integrating the power devices with the driver circuits not only gives excellent alignment between the drivers and the switches (reducing the source of electromagnetic emissions, which are linked in many cases to fast switching), but also reduces the physical size (and hence the "antennae" for electromagnetic emission), and structures the high power and high voltage layout in a way that makes repeatability of the system performance much easier to achieve. This also helps to increase efficiency, by effectively controlling the power switches much tighter. Last but not least, a smart power module also removes a lot of the creepage and clearance issues from system design, since now many of these connections are internal to the module. And reliability is improved by testing all power devices contained in the module together, different to a discrete solution where the system FIT rate would correlate to the multiplied FIT rates of the individual components , a much higher value. Going forward, I think we will see many more lower power motors using smart power modules, integrated into the motor housing, enabling the next level of applications of this type of motor with ease and improved time to market.

www.fairchildsemi.com