Finding a suitable and reliable power supply for a project seems trivial to most. The battery selection is typically one of the common types available at stores. However, if you have a project that needs to run on battery, that choice will influence its overall use.
The first question to ask is, what are the components of the project. These can range from discrete components to boards or modules requiring varying supply voltages. As the number of components increase, so does the complexity and requirement for power management devices, such as regulators or switching power supplies. Collectively, they are the project load. This is a measurement of the work required to operate effectively.
One example of a project that uses 2 devices, the first is a ESP-32 Camera module (max 310mA @ 5V). the next is a GP-20U7 GPS receiver (max 40mA @ 3.3V). Since we have two different operating voltages, there will need to be a third device, a voltage regulator to bring the 5 volts down to 3.3 volts for the GPS receiver. Our power source will likely not be a perfect 5 volt source, so a forth device will be needed to provide the initial 5 volts. These 2 additional voltage regulators will require power to operate.
Kevin Darrah demonstrates the use of different regulators and how their efficiency varies.
It should be noted that his results are in par with the data-sheet spec the manufacturers provide. It can be hard for those outside the trade to nail down a precise number when referencing data-sheets from various vendors. This is because it requires a careful eye and scrutiny. Having test equipment is useful to provide validation. However, the data-sheets should lead and the test equipment should follow to validate.
Andreas Spiess demonstrates different sensors that can be used to measure load current.
These are simple low cost devices that can serve the purpose of validation when a bench environment isn’t practical.
As I pointed out, data-sheets are a fundamental resource. This is also the case with batteries. Lets take a popular AA battery, https://data.energizer.com/PDFs/E91.pdf. The data-sheet provides details most of us weren’t aware of, the 10 year shelf life, or the discharge curve based on load. From the specs, if we have a load that draws 250mA, we can expect the battery to operate until it reaches a low voltage of .8 volts at or after 8 hours.
With the data-sheet information, a measured test to validate will likely substantiate the findings in the data-sheet. Try it for yourself, connect this battery to a circuit with a 6 ohm resistor and you should expect it to maintain a voltage above .8 volts for at least 8 hours. Of course as the battery voltage begins to drop, so will the current draw over the fixed load, resulting in a bit longer run time. Some data-sheets lack details, so testing should be done after determining the expected behavior from the data-sheet. That testing should give more resolution, not establish the expected behavior, remember to test later, not first.
In this link, Rui Santos and Sara Santos demonstrate how to use solar panels and a rechargeable lithium battery circuit to power a micro controller that also monitors the battery level, https://randomnerdtutorials.com/power-esp32-esp8266-solar-panels-battery-level-monitoring/. Some may suggest that these data-sheets are a great way to help fall asleep. I won’t fault you for having reluctance in clicking one of these links, “tp4056.com” hint hint.
At any rate, this brings us back to the devil in the details. If you want a project that operates as you expect, you will need to stay within the spec.
Further information related to this topic.