Energy harvesting is not an all-or-nothing proposition

Energy Harvesting Sensors
Misperceptions about energy harvesting abound. An EH solution need only provide a fraction of peak system power requirements, even intermittently. (Getty Images)

Since the very first time I powered up a big computer server in 2004 working for HP (now HPE), I was blown away (literally) by the deafening wall of hot air that blew in my face.  In that naïve moment, I wondered why all the engineering effort and discussion around me was focused on how to get rid of this “waste” product.

Ironically, it turned out more energy was spent on mitigating waste heat and the associated overhead than consumed by the load itself.  But what really bothered me was everyone’s infatuation with treating this energy like garbage that we must figure out how to get rid of.  After all, this was an energy source!

Fast forward to 2014 when I established the power electronics consulting firm PowerRox and was able to focus my attention more on areas of interest.  (Side note for those out there considering independent consulting: if you cannot enjoy the consulting work unpaid more than what you do otherwise, this may not be the path for you. )  At this time I got to dive head first into the world of energy harvesting (EH) and learn that we can capture and reuse every manner of energy afforded to us by the physical world. 

This research quickly gave way to giving talks to try and educate folks on the topic as well help bring the supporting technologies (and eventually, ensuing ecosystem) closer to the mainstream.  In particular was a hot, new topic about the Internet of Things (IoT) and the millions or even billions (now I will even argue 1 T) doohickeys that sat at the sweet spot for EH because of their [typically] very low power consumption.

After a couple years and couple dozen talks (mostly in and around the Silicon Valley area), it became apparent there were two major misperceptions about this emerging technology space: First, EH was an academic lab experiment not fit for production (particularly high-volume) applications. Second, EH produced negligible power for doing “real” stuff. 

While neither of these misperceptions were very astute (even at that time), they were, nonetheless, very common opinions of the key stakeholders needed to embrace and adopt EH technologies to enable successful deployments.  For one, there existed a production ecosystem at that time for many key components (such as EH-optimized power management ICs or PMICs), including multiple, top-tier semiconductor companies, that had already been around for more than 10 years way back then!  The technical readiness levels (TRL) and manufacturing readiness levels (MRL) for EH tech are not the focus of this article, however.  Now we’ll focus on the second concern: negligible power.

One of the most pragmatic and eye-opening concepts (regardless of its simplicity) to internalize in the world of EH is that it is not an all-or-nothing kind of solution.  In other words, one should not assume EH solutions have no value in their system if they cannot practically design an EH-based energy source that is equivalent to the energy source it is intended to supplant, whether that be a battery or some offline (i.e. – plugged into the wall) source. 

There are many applications in which supplemental power has a whole lot of value.  Start with the quintessential example of all things in the battery-powered world, the smartphone in your pocket.  You may not keep your phone perpetually topped off today by adding a PV solar panel to harvest light energy (indoor or outdoor) or a thermoelectric generator to take advantage of the temperature differential between your body and the ambient air, but what if you could extend battery life by 20 percent?  Might this be the difference between making it to the end of the day without plugging in?  Energy storage technology does not come close to following Moore’s Law. Battery capacity (energy density) only doubles roughly every decade. Even though Apple or Samsung may release the latest and greatest like annual clockwork, your battery is not advancing at that rate.

Perhaps you are thinking that is well and good for really low power systems, but EH may still be negligible (or better put, not financially-justified) for higher power systems.  Here is a good opportunity to say that any large system can be broken down into smaller, lower-power pieces.  EH may not power a server in the data center, but it could provide the supplemental, auxiliary power a big system may need to maintain a heartbeat in a sleep mode and monitor for wake signals. 

Today, that tiny auxiliary power rail is adding a great deal of cost, space, and inefficiency to a much larger power supply optimized to deliver kilowatts of power.  What about a drone or unmanned aerial vehicle (UAV)?  Might those extra few minutes of flight time be a difference-maker in your application?

The best approach here is to think long and hard about your application and what the true, system budgetary power requirements are.  In general, system power budgets are already heavily overestimated due to many layers of margin that tend to be added to an already flawed, sum-of-maxima approach to simply adding absolute max power draw values from the datasheets of key system loads.

Energy storage is the true hero in this story that does not receive nearly enough attention.  Your EH solution need only provide a fraction of peak system power requirements, even intermittently, with a well-designed storage solution for addressing the peaks while keeping the overall infrastructure sized to something closer to the (much lower) steady-state average.

A recent EU publication from the EnABLES EU ‘power IoT’ project gives good insight into real-life examples as shown in the figure below.

energy harvesting chart


 EH sweet spot for IoT edge devices / battery life extension impact. (Source: courtesy of EnABLES)


In the Power Sources Manufacturers Association (PSMA) Energy Harvesting Committee (EHC), we recognized these gaps and took an iterative approach to collecting these compartmentalized constituents of an emerging technology area. We assembled them in a meaningful way that not only connects these critical stakeholders but provides the appropriate environments/mediums to allow the ecosystem to flourish and penetrate the mainstream.

The EHC is composed of many members of both IEEE Power Electronics Society (PELS) and PSMA organizations so it is ideally-suited to tackle these multidimensional challenges.

Another major, recent achievement of the PSMA EHC is the recent release of a white paper addressing “Energy Harvesting for a Green Internet of Things” as compiled by an expert team of many of the world’s leading experts in their respective fields.  This effort was led by Thomas Becker of Thobecore and Michalis Kiziroglou of Imperial College London.

For more information or to download this white paper, please visit the PSMA EHC’s website or reach out to the author of this article.

Brian Zahnstecher is a senior member of IEEE, Chair of IEEE SFBAC Power Electronics Society (PELS), Chair of IEEE PELS Technical Committee 7 (TC7)si and sits on the Power Sources Manufacturers Association (PSMA) Board of Directors. He is also co-founder & Co-chair of PSMA Reliability Committee, Co-chair of PSMA Energy Harvesting Committee, and is Principal of PowerRox.  He Co-chairs IEEE Future Directions (formerly 5G) Initiative webinar series and is the founding Co-chair of IEEE 5G Roadmap Energy Efficiency Working Group and has lectured on this topic at major industry conferences.  He previously held positions in power electronics with Emerson Network Power, Cisco, and Hewlett-Packard with nearly 20 years of industry experience.  He holds Master’s and Bachelor’s degrees from Worcester Polytechnic Institute.

Editor’s Note: Sensors Converge  features a  pre-conference symposium, “Energy Harvesting Enters the Mainstream to Power the IoT” on Tuesday (Sept. 21) at  9 am PDT ).  A one hour tutorial, “The Path to 1T Sensors Before 2025: An Exploratory Workshop focused on Energy Consumption” happens on  Wednesday Sept. 22 at 2:45 pm PDT.  The program is free online and in-person in San Jose with advance registration.