Batteryless Microcontrollers - The future of IoT
Everything from farming to national defence is now being bolstered by smart technology enabled by the Internet of Things.
Why has IoT managed to penetrate so far and wide you ask?
It’s really simple. Think about the world before the world wide web - remember how difficult it used to be to coordinate a hang-out with your friends, book tickets or find your way somewhere? The internet allowed for essential information to be distributed essentially seamlessly at practically no cost to the end-user.
Processes became more streamlined and orders of magnitude less resource-demanding. Everything from buying a car to finding a romantic partner could then be done using this immense framework - even if tinder was around in the 1980s, the basic technological framework to enable such a possibility was already being put in place.
The story with IoT is very similar in many ways - IoT is best understood as an evolution of the internet. Built on robust and highly sophisticated systems that were largely already in place before its inception, IoT technology has found employment in a staggeringly diverse range of applications, across thousands of sectors.
But, what is the internet of things?
We all know and love the internet, right? But what is the internet, though? It’s essentially a group of computing devices that are connected together to form a network such that information can be freely shared and accessed by those connected to it.
The internet of things is no different. Over the years, our silicon chips got tinier and smarter, until we were able to fit them into practically anything - when we started producing chips that could essentially mount a tiny, fully-autonomous computer onto a single integrated circuit, the foundation stone for the IoT was well and truly laid.
In its essence, IoT aims to connect all the objects around us to the internet. By bringing internet connectivity to all objects and not just traditional computing devices such as smartphones and computers, the true power of the internet can be unleashed.
Wireless sensors have made digital intelligence into a commodity that can be customised and purpose-tuned to be integrated into a variety of everyday objects across various settings and use cases.
When an object is implanted with sensors and a microcontroller that is able to send and receive packets of data from the internet, the object suddenly has a whole new range of functionality. This principle has been used to good effect in most consumer-facing applications of IoT technology, such as fitness trackers and smart appliances.
You are able to control your smart television or smart air conditioning unit with your smartphone because the appliance is fitted with a microcontroller that allows it to interact with the internet.
What is a microcontroller, you ask?
Well that’s simple - a microcontroller, often abbreviated as MCU, is a self-contained computer that is mounted on a single integrated circuit. What makes a microcontroller (MCU) different from your laptop is that microcontrollers are often designed to perform one dedicated computing task.
Any smart device that you see around you right now is using an MCU. Microcontrollers are the building blocks of any IoT network, facilitating the collection and distribution of data. Today, IoT companies are on a veritable rampage, fitting all manner of objects with IoT microcontrollers, trying to entice customers with the added functionality and wow factor.
However, a strange conundrum has faced IoT technology right from its nascent years - the battery problem.
The problem with battery-powered IoT
About ten years ago, IBM stirred up a whole lot of buzz in the tech world by making a nutty-sounding prediction that there would be 1 trillion connected devices in the world by 2015. Scoffing onlookers would be proved right as 2015 rolled around, and we were yet to hit even 100 billion.
Around the same time, a couple of enterprising MIT graduates named Calhoun and Wentzloff realised the obvious problem that was staring them right in the face - a trillion sensor world would essentially mean having to provide a trillion batteries.
This is a massive problem for a number of reasons. Firstly, the economics simply don’t make any sense - after all, the big value-add that IoT technology was bringing to the table was range - IoT would bring the power of the internet to far-flung corners of the world, throwing open a vast spread of area to the incisive information-gathering power of IoT sensors. This meant that for the first time, the incredible power of the internet would make a difference in places like construction sites, factories, large farms, lumbering operations, mining operations etc. - the core ideas being decentralisation and enormous coverage.
IoT networks rely on sensors to collect data which is then distributed across the network to be processed and analysed. If batteries are used to sustain these sensors, that would mean a severe restriction on the number of sensors that can be used - which is then tantamount defeating the whole point of IoT in the first place.
Imagine a large-scale agricultural operation - the major value add that IoT brings to such a business is coverage - By deploying a large number of sensors spread across the entire operation, the owner can access highly actionable insights. This proposition very quickly becomes a less compelling one when the business has to account for the recurring cost of maintaining and replacing hundreds or even thousands of batteries every year.
A robust mid-sized smart factory typically employs about 10,000 IoT nodes that are deployed in various discrete locations checking on a variety of parameters such as air quality, humidity, temperature etc. The costs of maintaining and replacing the batteries on all these sensors would be astronomical. They would very easily outweigh any potential savings such a system would bring to the table.
Even if batteries last as long as they claim to, which obviously can’t happen nearly all the time - there is the issue of replacement and maintenance - facilities would still have to fork out sizable chunks of cash in addition to valuable man-hours for maintenance.
Again, not very feasible.
It is estimated that a trillion sensor world would make for 274 million battery replacements every single day! And that is assuming that every battery deployed in every IoT device would reach the claimed mark of 10 years.
Manufacturers and industry insiders very quickly realised that powering microcontrollers using batteries simply made zero economic sense, robbing potential clients of any upside that an initial investment could bring to their operations.
The need for a potentially batteryless power solution for the burgeoning internet of things was starting to be recognised.
It gets worse
The economic and logistical infeasibility of battery-powered microcontrollers is just one part of the puzzle - batteries come with another sore pain point - unsustainability.
Most batteries use rare minerals that are predominantly found in small pockets of the world. Lithium and Cobalt are minerals that are widely used in modern batteries, and both of them are mined and distributed by an industry that is highly contentious for its humanitarian and environmental negligence.
Lithium mining is highly destructive to the surrounding environment, often rendering it inhospitable to life. Cobalt, on the other hand, is mired in ethical dubiousness. Over 90% of the world’s Cobalt deposits are localised to a single country - the Democratic Republic Of Congo, where egregiously underpaid child miners risk their lives and health in appalling and inhuman working conditions.
Disposing of these batteries presents another dire challenge to us - When batteries or Li-ion battery-powered devices are discarded into landfills, they tend to leach corrosive chemicals such as Cadmium, Nickel, Lead and Mercury into the soil and water table, which can wreak havoc on the health and well-being of entire communities, gradually poisoning the very environment we live in.
An EU-funded study estimated that over 78 million IoT batteries will be discarded every single day by the year 2025! Take a second to wrap your head around that number - that’s 28.4 billion batteries every year! There is no way life on this planet can thrive for much longer if we reach anywhere close to that figure in actuality.
These are just a few of the major pitfalls that come with relying on batteries to power IoT nodes - simply put, they are nowhere near being a sensible power solution for IoT devices.
It was starting to get clear as daylight for everyone involved that the battery problem was to be the defining challenge that needed to be overcome if IoT was to truly take off. There was simply no other way we could build a connected world with a trillion connected devices unless we came up with an answer to the battery problem.
Initial efforts to counter this problem were directed towards building better batteries that would outlast the devices they were meant to power - it was typical for IoT devices to have life spans of up to 10 years while the batteries that were powering them would usually run out in 2.
A seminal paper on the matter, published by EnABLES, an EU-funded project, called for smarter product design, low-power IoT devices and self-recharging batteries - the message was loud and clear - the existing framework for powering IoT devices was simply untenable.
As the years rolled along, sensors and semiconductor devices became unbelievably efficient and cheap to manufacture - modern sensors were extremely reliable and consumed minuscule amounts of energy. The emergence of these ultra-low power semiconductor devices shone a massive beacon of hope to everyone that still believed in the gargantuan potential of IoT technology in crafting a better future.
Batteries improved, too - necessity drove innovation, and the market saw a lot of cheap, robust batteries that were a marked improvement on their forerunners - for example, an innovation that stirred up quite a lot of buzz was a solution that used steam traps in industrial settings to recharge IIoT batteries.
Most of the IoT industry woke up to the fact that batteries, no matter how good they got, were nothing more than a stopgap solution, at best. It was becoming clear that the future of IoT sensors and microcontrollers needed to be batteryless.
After all, even rechargeable batteries that are extremely reliable and last a long time are still going to pose a number of pretty insurmountable challenges. In any industrial-sized rig, the biggest problem would be that they would require manual maintenance - which simply doesn’t make any sense if you factor in the fact that there are typically thousands of these sensors monitoring various kinds of parameters in these setups - that would make such a setup prohibitively expensive in terms of running costs.
Secondly, batteries, again, no matter how long they lasted, needed to be disposed of at some point in their life cycles - the more rechargeable they are, the better, of course. But if the future of IoT were to be entirely battery-powered, that would account for hundreds of millions of batteries piling up in landfills and ocean beds over a period of time. Any battery-based power solution for IoT microcontrollers was going to be a big-time environmental fiasco. The need was for an entirely self-powered IoT system.
Batteries running out after a finite amount of time would also pose logistical and operational problems - points that would surely not be overlooked by any potential industrial client. If critical sensors placed in locations of high strategic importance suddenly experience failure, it can lead to costly gaps in mission-critical situations, which again defeats the essential purpose of IoT.
Painfully for manufacturers, batteries also add significant heft to any IoT design, which adds a further layer of constraints to a sensor’s functionality. This additional bulk presents a design challenge which can often be painful to contend with, considering the typically diminutive sizes of IoT devices.
In light of these challenges, it became imperative that IoT find an alternate power solution if it were to remain relevant in the near future.
It is against the backdrop of the aforementioned problems that the need to develop batteryless microcontrollers assumed paramount importance. Energy harvesting technologies were emerging thick and fast and were making a very strong case to eliminate the battery from IoT microcontrollers altogether.
What is energy harvesting you ask?
Energy harvesting, power scavenging and ambient power are all terms that refer to the same thing - at any given moment in time, our surroundings are pulsating with millions of intersecting waves of energy. This energy can be in a number of different forms - electromagnetic energy in the form of radiofrequency waves, heat energy, vibrational energy etc.
Energy harvesting refers to any solution designed to convert ambient energy into usable packets of electrical energy that are then used to power sensors and IoT nodes. Considering that the typical IoT microcontroller only needs a handful of millivolts to work, innovative players in the IoT space started looking towards these technologies as a potential power solution that would solve all their battery woes.
A number of energy harvesting technologies based on various kinds of materials and power sources - RF energy harvesting, piezoelectric energy harvesting, and triboelectric energy harvesting emerged as leading contenders in the category.
Cut to a few years ago, and energy harvesting-based startups were securing hundreds of millions of dollars in investments. The market had firmly caught on to the battery problem and were pouring tons of cash into energy harvesting in a bid to accelerate the growth of the standard and bring it to market as soon as possible.
Today, the self-powered microcontroller is no longer a pipe dream - we have a handful of batteryless MCU options on the market today. Without the restrictions imposed by batteries, these MCUs have majorly disrupted the IoT space with their near-endless scope for applications.
Applications of batteryless IoT
Unlike battery-powered microcontrollers, self-powered ones are 100% untethered - this means they can go essentially wherever one wants them to go. Batteryless IoT is finding a wide range of applications in the world around us. In the following section, we highlight a few notable applications where uptake of batteryless IoT technologies has been prolific.
IoMT or the Internet of Medical Things, is possibly one of the most exciting verticals within the IoT space at the moment. The implications for the healthcare industry are huge - most countries in the developed world are having to allocate enormous proportions of their annual spending for healthcare. Nevertheless, thanks to a growing global population, poor lifestyle choices and other factors, our healthcare challenges are growing by the day.
Long story short, batteryless MCUs make for a compelling proposition in the IoMT space - with biomedical sensor research reaching a feverish crescendo the world over, solutions like continuous monitoring, fall detection, automated EMRs etc. have transitioned from being promising ideas to feasible real-world applications.
Self-powered MCUs are being used in a number of savvy IoT-based innovations such as smart bandages, smart hospital beds, biometric tracking devices etc.
2. Smart cities
Smart cities were touted as being the most promising IoT vertical in the early days. Over the years, although there have been some famous examples of cities embracing connected tech, smart cities have not quite stirred up as much interest as they were expected to.
The battery problem was a crucial impediment for smart cities simply because of the scale entailed - smart cities, probably more than any other application, had a dire need for a viable batteryless solution. With inexpensive wireless sensor networks and self-powered microcontrollers, this really exciting IoT application can truly take off worldwide.
3. Smart home
This is another IoT use-case that wasn’t quite able to generate as much interest among consumers as expected. This is a vertical that is especially promising for batteryless sensor tech.
As such, today’s smart home user has to deal with multiple providers and manufacturers offering various solutions that work with only a handful of things at home. Adding an additional pain point of having to replace batteries is simply not great business - so it’s no surprise that manufacturers are looking towards batteryless MCUs for solutions.
4. Waste management
With the worldwide population hitting the 8 billion mark earlier this year, there has been widespread apprehension about the various challenges that await us as a global collective in the near future.
As more and more people start breaking out into the middle class and moving into urban areas, our already severe waste management problem is only going to worsen.
IoT-based solutions are being embraced wholeheartedly in order to combat our growing waste problem - real-time analytics and tracking add an impressive amount of rigour and efficiency to the process of managing, sorting, distributing and disposing of our wastes.
Batteryless microcontrollers are being used in a number of ways in waste management, such as landfill alert systems, effluent level sensors etc. A waste management solution powered by batteries is ironic and untenable. Batteries would add to the problem that’s being solved in the first place.
Advantages of batteryless Microcontrollers
Organisations stand to benefit greatly by opting for batteryless microcontrollers over their battery-powered counterparts. Firstly, batteryless MCUs slash maintenance costs significantly - they have fewer running parts and don’t need periodic maintenance that battery-based IoT systems require. Over the lifecycle of an IoT system, this adds up to some substantial savings over battery-powered offerings.
By eliminating the battery and its accompanying components, Batteryless MCUs not only help to cut costs in the Bill of Materials (BOM), but they also optimize real estate on your circuit board. This makes them an economical choice for any project. Typically, a switch to self-powered microcontrollers allows for savings of up to 80% on the total bill of materials (BOM) cost.
But here’s the real advantage that gives batteryless microcontrollers their winning edge - they are significantly more eco-friendly - they have a lower carbon footprint and have lesser impact on the environment compared to batteries which are often problematically sourced and account for huge amounts of toxic waste being generated.
This is by far the most compelling advantage that batteryless IoT brings to the table - as we move ahead in these scary and precarious times, it is starting to become glaringly obvious that our margin for error as a civilisation is minuscule - we really are at a knife’s edge in terms of potential ecological disasters. The need to embrace any step that leads us in the direction of sustainability and low-impact development is severe and urgent.
Sustainability isn’t optional
Public awareness of issues that pertain to all of us as a collective is at an all-time high - today, more people than ever before are well-informed about ecological issues and are reflecting this awareness in their buying choices.
In the past, companies often embraced green causes as a PR ploy and possibly even as part of a greenwashing strategy to divert public attention. Today, it’s proving to be really hard in this age of unlimited information access for companies to get away with anything short of the real deal.
Today’s businesses simply won’t get away with eye-washing the public - some of the biggest companies in the world are catching up to this fact. They understand that millennial consumers are extremely particular about the brands they associate with, often demonstrating that they would rather pay more to associate with the right brands.
Samsung, for instance, has stepped up its sustainability game, announcing a slew of ecologically-minded initiatives. Similarly, IKEA garnered a lot of public attention for its ban on all non-rechargeable batteries across all their outlets worldwide.
While these are encouraging signs, there is still a lot to be done. Energy harvesting technology or batteryless sensors won’t stop erratic monsoons, wildfires, mass extinctions or soil degradation. The challenge of making sure that the world is still suitable for life to thrive in the near future is a daunting and massive one. But here’s the thing. We have to start somewhere and get better through iterative improvement. After all, that’s exactly what IoT systems do. Batteryless IoT is a step in that direction - a trend that truly inspires hope and stops millions of coin cells and small batteries from ending up where they shouldn’t.