I was just reading latest news in the forum and stumbled over the “DIM D7” // Brown-Out Bug that happens especially on battery-driven Electron but - as far as I understand - also on Photon Devices. I am referring to this:
We (my company) just made the descision to use Electron as the main controller for industrial machines we produce. Since we are talking about a lot of machines we are currently a bit afraid of using the Electron so here are some questions:
Does this Bug still appear when NOT using a Battery but constantly switching on/off the device?
Are all Electrons affected by this bug? I have an Electron which rebooted after running low on battery. It still works!
Is this bug something that can be addressed by a fix? Or is it because low-voltage causes this? That would mean that this bug is not Particle.io specific but more STM specific and that this might also happen to other manufactures using STM!?
Really these questions are important for us because we have to make the next steps and there isn’t so much time left to choose wether to use Electron as the main controller or not …
Best I can tell is it’s triggered by low battery voltage. It can be avoided by running code that puts the Electron into sleep mode once it hit’s 20% SOC.
Not all Electrons are affected by this, only a small amount of cases have been reported. It’s never happened to me.
Powering the Electron ON and OFF with a good voltage should be perfectly fine.
When it does happen it looks like it clears the flash memory completely.
That’s what I have read so far.
Just don’t run your batteries down to the point the voltage drops out.
Hey @simmikolon – we’re currently in the middle of a deep firmware investigation into this issue. One of the ways that we’re mitigating this issue is by implementing CRC checks for all devices (including Photons, Electrons, and P1s) after operations that write to the flash memory to ensure that they “stick”.
To date, we have been unable to reliably reproduce the issue. The best data you can collect is data running in your target environment with the target firmware application. If that performs reliably, it is unlikely that the experience will deviate at scale.
In the meantime, look to upcoming firmware releases – we’ll be shipping improvements to device robustness and connection reliability in the next 1-2 releases.
No problem. If you experience any instability, be sure to let us know! In the meantime, we’ll be hard at work to ensure that all devices recover gracefully from events like this one.
