in my project I am using and Arduino Nano Every. I am concerned to run in current limitations for the devices I wish to connect. I am running a 12V power supply and have connected both GND pins to PSU GND and VIN to PSU 12V . Connected to the Arduino I have:
3.5" LCD drawing 150mA on the 5V pin, plus other small loads which I can't measure on the other digital pins (~200mA total I'd say);
DS3231 RTC, air T/H sensor, soil humidity sensor (power draw not measured, but estimated to ~100mA);
4 relays (70 mA on 5V + 10mA on I/O when on, 80mA on GND).
So total power draw is 200+100+80*4 = 630mA.
I have not managed to find any precise documentation regarding the current limits for the Nano Every, but I am aware that on the 328P Nano it's ~500mA.
With the 4 relays I need to switch 3 12V DC loads and a single 240V AC load. I thought about using solid state relays (less power draw) but I couldn't find any that worked for both AC and DC loads, so I am stuck to the electromechanical relays and their 80mA power draw.
My question is: how can I best solve this problem? Would it be doable to use an external 12V to 5V buck converter to power the relays only, as the ground would still be common with Arduino?
The 3 12V loads that I want to switch with the relays are <1A and share GND with Arduino as I run them from the same PSU. Should I use a different method to switch these 3 loads and keep the 240V AC on a relay?
For example I have been told to use MOSFETs to switch the 12V load, but I am not knowledgeable enough to make an educated decision. I would also like to keep complexity to a minimum so I'd rather prefer to use a 4-channel relay module (like this) with an additional power supply than use MOSFETs for 12V and a relay for 240V.
No Arduino board is usable as a power supply. Use a buck converter to drop the 12V to 5V for the Nano and the 5V peripherals. The grounds from the 12V supply, buck converter and Nano should be tied together.
Follow up: the DS3231 RTC, air T/H sensor and soil humidity sensor provide an analog signal to the IO pins and the LCD screen has its own parallel data bus. If I supply the modules with 5V from the buck converter, are the values going to get thrown off a bit (as the Arduino and buck 5V will be ever so slightly different)? Can the LCD or the Arduino IO pins be damaged because of the difference in voltage?
I'd just keep these 4 devices on the Arduino 5V pin (as they draw <500mA, less of the 5V pin current limit) and reroute all the other connections (relays) to the buck converter 5V, or am I getting it wrong?
I would connect the 5V from the buck converter to the 5V pin of the Nano. Then you can power everything from the 5V of the buck converter. That way every thing gets the same 5V.
I already had a setup similar to this (but with a Uno), but instead of a buck converter I was using a "standard" 240V to 5V Meanwell PSU. Sometimes the Arduino would just shut off and refuse to boot, even after disconnecting and reapplying power. I thought it was due to the direct 5V supply and I moved away from that design, as using it with the USB connection gave no problems whatsohever.
So am I safe with just the buck converter or should I put a capacitor/regulator to make the input current a bit more stable?
Most buck converters (and mobile phone chargers, another great 5V source) produce clean enough power for your Arduino to work properly. Those symptoms are more consistent with a poor power wire connection than generally unstable power supply.
If you want really clean power, use a Pi filter (two caps and an inductor). I've done this on projects where the power supply of the Arduino would also power some high-power devices such as motors and long programmable LED strips, where the Arduino's power supply was behind the Pi filter and the motors and LEDs connected directly to the power supply.
