3 Design Best Practices to Assure a Long Battery LifeFollow article
If you have already been following Qoitech and our work for a while, you probably came across our three best practices for long battery life - or the three most critical steps to get the best out of your Internet of Things device, or any battery-driven device. In my opinion, it really can’t be said enough and I want to tell you why.
Battery life is the most critical thing. Full stop. Not only is it essential to get the best out of your device but if unreliable and unpredictable can actually break your product or even your business. It might sound a bit dramatic but I think we’ve all found ourselves in a situation once where we swore over a dead device with no explanation whatsoever.
Designing energy-efficient devices is a mindset and needs to be practiced and nurtured on a daily basis. And by that I mean looking beyond your datasheets, digging a bit deeper, and to have all the insights and knowledge you need about the hardware and software that make up your product.
In this article, I want to share with you my/our three best practices that will help you and your team to get into this energy-efficient mindset.
#1 Know your device
Every device has its unique energy profile and reacts in a specific way in different scenarios. Measuring repeatedly is the key here. Snapshots here and there won’t bring you very far unless you measure continuously - for changes in hardware and software, for different battery voltages, for different locations or environments.
Let’s say your IoT product (using LoRaWAN technology) is supposed to be placed relatively far from the gateway or base station. The long-distance will force your device to go into a lower data rate which means that it will consume more energy for each transmission compared to a smaller distance and a higher data rate, see Fig 1. So, just placing your device differently will quickly lead to a very different energy profile without you even making changes to neither hardware nor software.
Figure 1 Otii Arc measurement of LoRaWAN power profiles for two different spreading factors
#2 Know your energy source
Once you know your device and its energy profile you want to choose the right battery. This is the moment where you should say goodbye to data sheets or at least look beyond their general statements. Of course, you need to know a battery’s behavior when it comes to self-discharge but also (and more importantly) its behavior in the context of your device’s specific energy profile - its peak current, duty cycles, and sleep modes. So, profiling is crucial for evaluating and choosing the right battery for your device and application.
For example: let’s assume your device’s energy profile has quite high peaks but you still want to go with a coin cell battery because the form factor of your device doesn’t allow for much bigger options. Even though coin cell batteries are not the ideal choice for high peaks, there might be batteries that work very well. (We actually compared many different brands with surprising results. If you’re interested, you can read about it here). You want to know how different batteries and brands perform with the high peaks, but also how they perform in different environments (high and low temperatures) and how they handle different data rates. This is the only way to find the right battery.
Figure 2 Difference in performance of two batteries of the same type and size but from two different brands
#3 Emulate your energy source
Now that you have an energy profile for your device and the profile of your battery, you want to combine these and emulate your device’s performance. Iterate the used capacities until it stops working. This is the only way to see how much you can actually get out of your battery. Knowing your battery’s profile is necessary, yet only emulating is sufficient to really know (not estimate) the actual capacity that you can use and therefore, your actual battery lifetime.
Look at the table below. A data sheet can tell you that Battery 1 has 610 mAh. Profiling the battery will show you that it is actually 224 mAh and emulating it with your device will tell you that it is in fact less than 100 mAh. Or in other words, the battery life of almost 10 years as stated in the datasheet can turn very quickly into 1.6 years - not even 20%.
Table 1 - Battery Data Comparison
So, here they are - my three best practices for long battery life. Measure, measure, and measure. Make sure you know your device/application, understand your energy source and look at the interplay between your device and its energy source. A “low power mindset” together with the right tools is the best way to avoid panic and firefighting and will help you exploit your product’s full potential.
One last thing I’d like to add: Involve everybody in the stack. Battery life is not only a challenge for a few hardware specialists in the lab. On the contrary, it is something that concerns the whole team and is for sure a system-level challenge. So, involve your software developers and app designer, and don’t forget to measure every day.
For more information go to qoitech.com or reach out to me directly.