Arduino Nano 33 IoT powered by Cr2032 coin cell battery?

I'm very new to Arduino and don't know much about electronics.
I'm building a device where I would like to translate the IMU information that the Nano 33 IoT is capable of, into midi information and send it to my laptop through Bluetooth. I made a prototype with the MPU6050 and Leonardo, which works, but it's wired and too big as I want it to be attached to a violin player's wrist. (That's why I think the Nano 33 IoT seems to be the best option since it has BTE and IMU) It would be ideal to have it on battery power, and am wondering if it could be powered with a single CR2032 @ 3v? (It doesn't need to last forever, it would only be used for an hour or so each use.)

Using bluetooth I belive that CR2032 will be emptied quite fast. Those cells are designed for very low current and long time.

Check the average and maximum current draw of your application.

CR2032 cells are recommended for very low power applications that do not draw more than about 15 mA, maximum (in short bursts).

If you do not exceed that, the expected CR2032 battery life in hours is roughly (235 mAh)/(average current draw in mA).

That much? I wouldn't have guessed that.

This paper describes tests on CR2032 under "high current" conditions and shows that short 15 mA bursts are not that harmful. https://e2echina.ti.com/cfs-file/__key/communityserver-discussions-components-files/104/7510.swra349-Coin-cells-and-peak-current-draw.pdf

The average current draw has a huge effect on battery capacity and lifetime, as documented here. https://www.dmcinfo.com/Portals/0/Blog%20Files/High%20pulse%20drain%20impact%20on%20CR2032%20coin%20cell%20battery%20capacity.pdf

Capture

I imagine that the OP's setup draws maybe 10 mA on average, so the lifetime estimate in the reply #3 above is not applicable.

1 Like

Thanks!
Having experience from traction batteries it looks like the same theory applies to "all" batteries, that the capacity gets a lot less for higher loads and so on. The less discharge current the more mAh to use. Even bursts of "high" current eats more than pulseTime * current.

Ah, okay. Thank you both so much. I guess this will not be fiesible then. I just looked at the datasheets for the onboard IMU and radio module. The IMU only draws 0.9 - 1.25 mA but the Radio module when used for Bluetooth out draws 130mA. (Just as Railroader mentioned in comment #2). Do either of you have any battery recommendations that could work? The goal is to basically have an accel/gyro that can be worn like a watch.

The accel/gyro is not a problem. The problem with your project is primarily the radio and secondarily the Arduino board, which has a bunch of power hungry peripherals that you don't need or want. The Nano 33 BLE sense would be a better choice.

That is fine for a crude prototyping, but for a workable watch, most people would use a low power processor, BLE for the radio (if you must have one) and a custom PCB.

Thanks for the advice. That board seems like a great option but unfortunately, the processer on the Nano BLE sense runs on nrf52 architectures which is not supported by the library I want to use (MIDIUSB.h - compatible with avr, sam, samd architectures).

I think you're correct that the best option is to have a wired prototype for communication and power, and to not use arduino for the final product.

Thanks so much for your help!