I have a Raspberry Pi 4 with a Navio2 and a Power Module. I use a 4S Lipo Battery (I tested several batteries, which I usually use for flying other drones. So the batteries are fine) at the Power Module. Recently, I noticed that the Pi4 always reports “Under voltage detected” and throttles the CPU to 600MHz. I dont have any hardware connected to the Navio2 currently and only a camera to the Pi4.
It seems like the voltage drops as soon as the Pi4 draws a bit more current. I guess this shouldnt be the case. Any idea what might be wrong? Or is the power module faulty?
A good quality 2.5A power supply can be used if downstream USB peripherals consume less than 500mA in total.
I guess the power module is not up for the task. You could try to plug a big capacitor(I use 1F+) into the 5V pin of the RPi, this might help to buffer voltage drops.
Another solution might be to use the 5V port of my power distribution board in parallel (it also only delivers max 2A). Can I connect my 5V at the same time to the PI 5V GPIO pin while the Navio power module is connected to the Navio?
You should not connect multiple BECs to the same load. They will work “against” each other and heat up and eventually fail. It sometimes takes only seconds before one BEC starts to smoke…
You could use a seperate more powerful BEC and remove the +5V wires from the power module cable.
I just thought the condensator is just another power source as well as another battery which are plugged-in in parallel. Why doesnt that work with such 2 power sources (navio power module and my power distribution board)?
I am no electrical engineer either. The way I understood it (except from seeing magic smoke being released from a ESCs BEC right after pluging in the lipo) is that BECs, like everything else, have their tolerances. So even two identical BECs will not have the same voltage output. Then there are two types of BECs, linear and switching. Mixing the two types is bad, but even two (or more) BECs of the same type may result in failure. That is because the BEC with the lower voltage output essentially becomes a load, being supplied by the BEC with the higher voltage. Due to the low internal resistance of the BEC circuit, the current flow can become high enough to destroy the BEC with the lower voltage. This is even worse for switching BECs, because they work by switching the supply voltage on and off really fast to create a resulting lower voltage. Two of them will never be in sync, so one BECs switches off, the other on and the voltage will be fed to the switched off BEC. Perhaps there is an electrical engineer around, who can explain it better (or correct me ).
The capacitor is just a passive element. It is charged by the BEC until it is full. If the RPi draws more current than the power module can handle, the voltage of the PM will drop below the voltage of the capacitor and the capacitor will discharge, this eases the load on the PM and the voltage will rise again. This will happen really fast of course and the voltage will stabilize at some value, as long as the power draw of the RPi stays constant.
Two batteries in parallel of the same voltage are no problem either. If one battery has a slightly higher/lower voltage, they will level out quickly. If you connect a charged and a discharged battery interesting things will happen, especially with lipos.
If you replace the power module, you will not get battery voltage and current telemetry anymore. I just read the technical specifications of the Navio2 again. You can connect more than one power source, because the servorail, micro/usb-c port and the power module are seperated by diodes. You can leave the power module in and have the UBEC as the main power source for the RPi.
I would use the GPIO pins. That would be the same as the USB-port right?
Could I then also use my old power distribution board, which delivers 2 A, together with the Navio Power Module to get the necessary current? Or should one board deliver all the necessary current?
I would use a power source that can handle the load by itself, but after we know (again, my bad…) that there are diodes in place, you could try two lower power sources.
I now found the time to install the new UBEC on the GPIO pins. Its kind of working, but it doesnt provide nearly enough power. The raspberry pi 4 is constantly in low-voltage mode. Is it even possible to put enough Amps through the GPIO pins? My multi-meter says I have 4.87V at the GPIO pin. Is that already too low?
Or do I have to craft some USB-C plug?
I haven’t tried to power Raspberry from GPIO pins, so I can hardly be of much help here. However, 4.87V may indeed not be enough. As far as I know, the voltage for Raspberry should not go below 4.9v.
At this moment, I can hardly conclude what could cause RPi to show such a warning since we have never encountered such issues before. However, I have checked that this warning might indicate that the power supply is inadequate. If you could conduct a couple of tests, that might help me understand where the root of the issue lies.
Is it possible to disconnect Navio2 with RPi from other parts of your copter and connect it to the power source only? Does the warning appear again?
).