I am powering a genuine Nano from a 5V 3A regulated power supply going into the nano's 5V and GND pins. I also have a PCA9685 running servos that I'd like to power from the same 5V regulated power supply.
I need to power the PCA through the header pins, rather than the terminal blocks (due to size constraints).
So here's what I'm thinking:
Nano powered from regulated supply to nano 5V pin and GND
PCA powered parallel to nano from regulated supply to V+ pin and GND
PCA logic powered from nano's 3V3 output to VCC
Is this healthy? I would think this would avoid any issues from powering USB at the same time as the power supply.
Does this bring up any issues with pull-up resistors with the PCA logic operating at a different voltage than the nano's? From the PCA manual: "This [VCC] is the logic power pin, connect this to the logic level you want to use for the PCA9685 output, should be 3 - 5V max! It's also used for the 10K pullups on SCL/SDA so unless you have your own pullups, have it match the microcontroller's logic level too!"
If I wanted to keep everything at 5V, is it a problem tying the V+ and VCC together on the PCA?
The PCA logic powered from nano's 3V3 output to VCC
Can you explain why, it will work just fine on 5V . The servos need 4.5 to 5V for the control signal. You are doing great. As far as the VCC and V+ both are 5V and can be powered directly from the power supply. Feel free to ask questions as you progress and when it is working let us know.
Here is why: I am trying to power the nano and PCA/servos off of a single 5V power supply input. I have the 5V+ going in to both the nano's 5V pin and the PCA's V+. I did not know if it would be harmful to also tie the VCC control power into this line, so I did some googling and found that the PCA logic could be 3-5V and decided to try this as an option.
Do you think I could tie them all together and keep everything at 5V? Would this be a better solution than dropping the PCA down to 3 if it means keeping the servos at a happier voltage?
OK, right, everything runs on 5 V, so you connect 5 V from the power supply to the 5 V supply of each part. You also connect the ground from the power supply to the ground on each part.
Now it is important when assembling something like this - actually any project - to run the power and ground from one part to another together as a pair, using twin "figure eight" style wire if it travels any distance. It is also equally important to run the data or control lines and ground from one part to another together as a pair from one part to the next so that means power, data and ground run as a group.
So in your case, power and ground run from your 5 V supply to the Nano, from where all the connections to your PCA9685 (four I believe as it is I²C) then run.
No, I take that back! The PCA9685 is the central part of your design, the 5 V supply runs to it, and
then power and data run back to the Nano.
Clearly you have to, though it might be appropriate to run separate wires (in the same bundle!) from the power supply to minimise effects from the servo current draw. The limitation is that the single ground wire is still common to the two.
And it is the "Vcc" which goes with the other wires to the Nano.
There is a concern about providing power to the "5V" pin on a UNO or Mega 2560 when it is connected to a PC via USB. This does not apply to the Nano.
We all do similar things especially when tired and in a hurry. The most important thing is that all the grounds be connected together. Look at the ground as your reference point and for a voltage to be valid it needs a reference point. That is why you typically connect the com, Gnd (or whatever) input of your multimeter to ground.