Hello, I am working on a project where I am running a machine learning program constantly on a Nano 33 BLE Sense, that is connected to a Nano Every on a datalogging shield (Deek Robot 8105) to save to a micro SD card and run an RTC.
I have been developing and testing the boards with USB battery banks, but because I'm trying to consume as little power as possible for extended in-field deployments, I am running into issues with pulling insufficient current for the bank to stay awake.
I also think it would be better (and cheaper) to change the setup to being battery powered. However, I want to be cautious of running into electrical issues, especially since the boards run at different voltages.
Here is a rough schematic of the setup and I was hoping for guidance if anybody had experience with similar projects or if there are any obvious issues. My biggest concern is if there are any issues with powering both nanos from one battery source by shifting the 3.3v source from the nano every through the level shifter into the VIN pin of the 33 BLE Sense. Not pictured in the diagram is the datalogging shield attached to the nano every and 6 AA battery pack attached to the terminal block. Thank you
Sounds like a great idea but do not do it. The micro draws current which will cause the output to drop below its needed voltage. The unit also does not draw a consistent amount of power so it is possible it could allow the voltage to go to high damaging the Arduino.
Many people use a Buck Converter, they are in the mid 90% range to drop voltage. They will stay stable at varying loads and are not that expensive. Personally I use SEPIC (Burk/Boost) converters. YES running two or more microprocessors on the same set of batteries is OK if properly done.
Before powering the units check the voltage range of the pin you plan on supplying power to. Vin has a voltage range with a minimum and maximum. Going out of this range can damage the Arduino or cause it to not operate properly.
That thing between the two Arduino boards is for logic signals, not voltage.
You can get USB battery packs that do not stop delivering current when you don't draw enough.
On the other hand, or is it really the same, you can use a briefly applied load to keep the pack awake, it would depend on the pack and perhaps need experiments.
It might be something like demanding 30 mA for 1 second every 20 seconds. This can be done with software and a few components, or built stand-alone - the 555 timer chip can be configured to make such a current use profile.
I've done it less artfully merely by figuring out just how little current can still be enough. Wasteful, yes.
As a side note, the level shifter works well to communicate between the two boards using tx/rx.
With regards to why I am using two boards, through my limited background and a lot of time, I couldn't get the 33 BLE Sense to work with the data logging shield. It's been a while since I decided I couldn't use one board, but the best understanding of what's going on is that the libraries for the RTC and SD modules on the shield aren't compatible with the nRF52840 on the BLE Sense. Or maybe that the shield is meant to run at 5v? Either way, I got it to work using a nano every and just moved forward with that.
If anyone has any suggestions or links to similiar projects or ways I can troubleshoot this, that would be amazing.
