Need guidance on battery selection for small wearable device

I need help in understanding the factors that affect my battery selection for a device. I'd like to make this generic so that it is useful for others but I will use what I am designing for examples and clarity.

The goal is to design a small battery powered device worn by a child with a wrist whos surface area on top is 40mm by 20mm with a height of around 5mm.

I believe that I need to know the voltage required by the unit, the amount of time that the power will be provided, and how long I want the battery to last in order to figure out the battery I need.

The chip specs have this:

• Electrical Characteristics
• Output Power:
• Max. Power: 21 dBm
• Consumption @ LTE Cat M1 (Typical):
• Power Saving Mode: 3.2 μA
• Sleep Mode:
• 1.61 mA @ DRX = 1.28 s
• 0.61 mA @ e-I-DRX = 81.92 s
• Idle Mode:
• 19.6 mA @ DRX = 1.28 s
• 19 mA @ e-I-DRX = 81.92 s
• Active Mode: 228 mA @ 21dBm, GNSS off
• Consumption @ LTE Cat NB1 (Typical):
• Power Saving Mode: 3.2 μA
• Sleep State:
• 1.54 mA @ DRX = 1.28 s
• 0.66 mA @ e-I-DRX = 81.92 s
• Idle State:
• 15.8 mA @ DRX = 1.28 s
• 15.3 mA @ e-I-DRX = 81.92 s
• Active Mode:
• 165 mA @ 21dBm, GNSS off

The device will need to connect to LTE-cat M1 or NB-iot, GNSS or gps to provide updates for location and geofencing.

I have two main points I am struggling with.

1. I would like an update every 15 seconds or so if the devices physical position has changed. Assuming the device is in use 25% of the time (6 hours/day) I am not sure how to calculate the power requirements per day that I could then use to say "I need a battery with voltage X that has Y amp hours to last Z hours."

2. Does gnss off mean this does not account for positioning data? if so, how do I account for that?

I have a design docs for the chip I am connecting too and I am only linking to it in case there is something I have missed here that helps. https://www.quectel.com/wp-content/uploads/pdfupload/Quectel_BG77_LPWA_Specification_V1.4.pdf

https://datasheet.octopart.com/BG77-Quectel-datasheet-139004296.pdf

• You're taking on an ambitious challenge. Those power consumption figures are for the cellular function only, so GNSS (e.g. GPS) will take even more. Even if the entire volume (40 x 20 x 5 mm) was filled with a lithium polymer battery, it would only provide about 300 mAh typically. You need to know the duty cycle of the cellular connection to do the calculation. If you don't understand that, I don't think you should tackle this difficult project without help. Commented Aug 2, 2021 at 15:55
• Thank you @MarkLeavitt for the response. Do you have any direction on where I can go to learn this, (I would accept that as an answer)? Also, knowing that this is an ambitious goal, If I were to look for someone for help am I looking for an electrical engineer? Commented Aug 2, 2021 at 17:23
• My advice would be to build it first and then try to tweak it to maximize battery life once it's built. Most of the modifications you'll need to make will be in software. The hypothetical power consumption figures are only relevant if your design is highly optimized.
– Drew
Commented Aug 2, 2021 at 18:19
• The GPS, combined with the 15 second updates mean that your power usage is never going to get low enough for uA to matter, electrically that simplifies things since you don't need to worry about leakage currents and the like (except for the sitting on the shelf for a year use case).
– Drew
Commented Aug 2, 2021 at 18:21

You're taking on an ambitious challenge. Those power consumption figures are for the cellular function only, so GNSS (e.g. GPS) will take even more. Even if the entire volume (40 x 20 x 5 mm) were filled with a lithium polymer battery, it would only provide 300-500 mAh. You would need to know the duty cycle of the cellular connection to do the calculation, and that actually depends on the cell network, signal strength, and device movement.

Rather than jumping ahead into chip and battery selection, start higher up in the process. Does this device already exist? Yes -- search for "GPS Tracker" and add "child" or "kids" and you'll find plenty of existing products. Some are wrist-worn, others are pendants.

The next-easiest approach is to use ready-made modules or development kits. These will not be miniaturized, but they will often include antennas (saving you that design) as well as example firmware. Now you can measure the actual current draw and duty cycle and begin to estimate the battery capacity needed and battery life.

If for some reason you want to design your own fully-miniaturized device from scratch, you'll need a very multi-skilled EE, or more likely a team of them. There's electrical design, PCB design and assembly, RF and antenna design, firmware, cloud server software, and app creation. Don't forget mechanical design and engineering, molding or 3D printing, FCC certification...and by the way, it's hard to find chips these days!