New to arduino, please bear with my ignorance. I've done some general net research, figured why not ask the experts in the forum.
Scope of project: Will be running off 12v 3S lipo, IR sensor, solenoid, multiple relays or mosfests for power efficiency and RF RX/TX will be involved in this project. Need this to run off of a battery and last for extended period of time up to a week for example off 2200-4400mah capacity.
Questions: Looking for the most power efficient arduino hardware, I believe it is the mini Pro 3.3v or 5v option. Sadly on the ardunio website, the mini Pro is retired older legacy device, but I suspect it will work with the regular IDE and i can load similar libraries as needed?
I don't need wifi or BT, not having the USB-to-Serial chip saves power so that is a huge plus. I'm not sure if there is a more modern board which offers the same power savings benefit and is not retired?
If there is a better option than mini pro, i'm open to suggestions?
One question going back to efficiency is the 5v vs 3.3v option, using a buck converter to bring down the lipo 3S voltage to 3.3v or 5v as i understand is much more efficient than the unregulated "RAW" pin which the mini pro will convert is also a problem on fully charged 3S lipo which is >12V. Looking for confirmation the inefficiency compared to regulated 3.3V or 5V from a buck converter. I'm going to guess the consensus will be to use a quality buck converter?
I have a great quality converter in mind but its output is 1.5A, I have not been able to find much information on the idle current draw of the mini pro. I believe each i/o pin output is 40mA but nothing about static load consumption?
The current consumption depends on several factors such as: operating voltage, operating frequency, loading of I/O pins, switching rate of I/O pins and code being executed. However frequency and voltage are the predominant factors.
The pro mini uses the 328P microcontroller and the data sheet shows the the active supply current at 8MHz and 3.3V is only around 3mA. Power consumption decreases with decreasing frequency and voltage. Of course judicious use of sleep can greatly reduce overall power consumption.
You need to do the calculations but sometimes a linear regulator can be as efficient as a switching regulator.
What proportion of the time will you have high power users like relays and solenoids activated? If it's a lot of the time, the Arduino hardware you pick will be irrelevant compared with them.
If on the other hand the application does nothing for long periods of time and you can put the whole thing into deep sleep, then the the MKRWAN 1310 has a deep sleep current of about 110 uA, That's micro amps not milli. I think this is the lowest of any Arduino product.
Again, depending on your power usage profile and the voltage requirements of the accessories, it may be an advantage to reconfigure the power source so it is say 3.7v and use a boost converter if a higher voltage is required for the other components. That way, the MCU can sleep without needing voltage conversion, periodically wake and switch on the boost converter to power the accessories, then resume sleeping.
I agree with @6v6gt: If you don't need to power any 12V devices, then this is a bad choice of battery since the components you mention need much lower voltages and the conversion will be inefficient how ever you do it.
If you do need to power 12V devices, but the circuit & Arduino will be in a low power sleep state for long periods, then a linear regulator may turn out to be more efficient than a switching regulator.
Generally, Linear regulators are less efficient than switching regulators. But at very low power levels, the reverse is often true, because linear regulators can often have a much lower quiescent current than switching regulators.
The best choice for your project will depend on the current consumption during sleep, the current consumption during normal operation, the percentage of time spent in sleep mode, and the efficiency and quiescent current of the linear and switching regulators you consider using.
Sorry, but it's one of those "it depends" answers!
I assume what you meant to say was "at least a week", because "up to a week" means that it should not last for more than a week, which would be a strange requirement!
What is "this" super-secret device you will not divulge in two topics and sixteen posts?
I believe that is absolute maximum per pin. You do not want to operate at absolute maximum. The microcontroller should not be a substitute for a controlled power supply.
As you have heard/read before, post drawing of the project. From that you can discover devices to use (and not use), then it's onto the minimal sketch to make each device work, and the finishing sketch to make everything work together.
I've run a string of 100 x WS2812 off 7.5Ah batteries for two weeks between charges. I suspect your hardware will consume much more than a few WS2812, from a battery less than half the size... so you might be looking at a day of use.
The 328P will sleep at something less than 1uA. But to get the Pro Mini down to that level, you would have to remove the power-indicator LED, and the on-board linear regulator. I routinely do that with the 8MHz version, and power directly from a lithium rechargeable battery.
Pro Mini clones can be found on Aliexpress and Ebay, or even Amazon.
sterretje, yes I was looking at sparkfun since they are a adruino dealer thanks
jim-p agree on operating conditions, what I'm unfamiliar with is the lowpower library and how that will work holistically with this project, something to test on my end but the 3.3v with the mods to remove the regulator and led is the plan moving forward
sciroccotorc high power users will be minimal so the standby current is key on this project and i use solenoid as an example to generalize the project, the actual project itself is part of a larger proprietary design which can't be shared.
6v6gt Boost converter is not as efficient as a buck converter by nature and the ancillary hardware requires lower voltage same as mini Pro so it would be required regardless of arduino state (sleep or active idle) but good point based on power profile it can be advantageous.
PaulRBon On switching regulators vs linear I agree, in my case switch is far more efficient for the use case. As for runtime yes up to a week and longer.
xfpd i provided the reason why already regarding the project. As for the battery capacity it should be more than sufficient and the max current per pin noted, the intent is not to drive high power with mcu obviously.
ShermanP that is exactly the plan to remove both and use the buck to drive this and other sensor. This link provides good testing results for anyone interested Making An Ultra Low Power Arduino Pro - The DIY Life
