Hi!

My project is as follows:
Multiple LED stripes which are distributed over several rooms should be controlled. They either can be controlled from a phone, some buttons or they cycle through some colors without any input device.

I have thought of something like this:

A RaspberryPi as a "master" controller.
Some Arduinos directly connected to the LED stripes.

The Raspi should give instructions to the Arduino like "set color to <R, G, B>" over a wireless connection. I have thought of RF433.
But the problem is that there other inputs planned, which should also go over RF433. I imagine everything will clash and interfere which each other.

I have thought of 3 options:

1. The master Raspi acts as a coordinator: It sends each Arduino its new color value and then "asks" every input Arduino for new data, then waits for a response. So only one device is sending something at a time. Potential risk is something I can't judge about: If the RF433 can be fast enough for all this back and forth communication.

2. Somehow load all instructions onto the Arduinos over RF433 and then let them interpret them. Give some timing signal from the master raspi. I find this really wacky.

3. Use RF24 instead, which seems to have multiple "channels". This would easily allow for one "raspi -> arduino" and one "arduino -> raspi" connection.

Any advice? I am not familiar with RF24 and how good or bad it works with raspi and arduino.

Btw I have some libs: "pynrf24" for the raspi and "RF24" for Arduino

 

Take a look at ESP8266 and ESP32 modules in lieu of Arduino.

They are wireless, can run as a server or client, and can function standalone.

Tons of info on youtube, various sites, hackaday.io

https://www.esp8266.com/


Regards, Dana.

 

Thanks for the fast reply!

I couldn't find any relevant advantage with the ESPs. It's more expensive than an Arduino mini/nano clone plus an RF24 transceiver. I also don't have any existing wifi network at the projects location (but I think the ESP can setup one).

Is there anything against option 3?

 

Option 3 is perfectly acceptable. Note the ESP parts also have reduced
boards although never as cheap as a simple radio. My only reason for
recommendation is we often do not know how experienced/capable a
poster is for embedded work, and the ESP stuff has a relatively low
learning curve, number of users etc...

I am assuming your library has the protocol for implementing a net, etc...
otherwise writing that from scratch would be a pain.

Regards, Dana.

 

Ok, thanks for your advice!

IIUC it's the defacto standard lib for 2,4GHz, all batteries included.