Raspberry Pi Stack Exchange is a question and answer site for users and developers of hardware and software for Raspberry Pi. It only takes a minute to sign up.
Sign up to join this community
I'm working on a project with several sensors needing various 3.3v and 5v power. I'd like to have my sensors running for at least 24 hours before needing to recharge everything. I did the calculations:
6 Volts 12 Amps discharge current
To get over 24 hours I would need at least a 400Ah battery. Perhaps there is a better way to power this? What is typically done for large robots with a lot of sensors?
Your approach is generally sound in that you are undertaking a power budget early in your design. Wrt your question ("What is typically done ...?"), I think the answer is that one typically evaluates different approaches, and different technologies to determine the tradeoffs between the candidate designs. Tradeoffs might include cost, weight, performance, etc. There are typically requirements to be considered as well; e.g. The sensor system must operate for X hours without re-charging the batteries.
One thing I noted when reading your comment was your statement, "... a single D battery is about 10Ah ...". If you're talking about a D-cell battery, you should know that the internal resistance of most batteries will increase when they discharge more current. This has the effect of reducing the actual amount of energy you can draw from the battery (Energy being equal to power over time).
I won't go into details here, but you can ask other detailed questions if you like. My point is only that you cannot correctly assume that a battery under discharge will maintain its rated voltage over the entire period of discharge. The website "Battery University" is usually a good resource for understanding the behavior of various battery chemistries.
Hope that helps.
I-squared R losses due to the higher currents in the batteries themselves and the wiring are lower when the voltage is higher... this is where the tradeoff study is useful. Another consideration is avoiding deep discharge whenever possible - deep discharges reduce the lifetime of your batteries.
Please re-evaluate your power budget. Perhaps measure the actual typical current.
You mentioned 9 MQ-2 sensors, 0.8W each. This makes 7.2W considering the "maximum" values from the datasheet. The actual average consumption will likely be lower - the heating element will draw the most power during initial heat-up. You may also consider if you need to run the sensors continuously - although for the gas sensors the heat-up time will be a limiting factor.
Even considering the Pi (let's assume 5V*2A = 10W, although it most certainly won't draw 2A average), I'm not sure how you arrived at the 12A current - for a 6V battery, 12A discharge current produces 72W (the actual useful power will be a bit less considering the efficiency of a step-down converter).
A 2.4kWh battery is somewhere between an e-bike and electric motorcycle battery. This is where I would look if I really needed this kind of capacity.
I haven't run the numbers, but in my case for example I have a system (Pi 4 + 7" HDMI screen) which will consume 7W (@ 5V) continuously. It will run for 7 hours on 6S li-po (3 in series and 2 in parallel each).
Even in the worst case (when a circuit associated to it consumes 72W for 5 seconds every 2 minutes) the batteries will last 3 hours.
400Ah seems to me an overly large battery for your case. Will you really need all those 12A continuously? If so, can't you by chance switch them on an off as you use them?
Just quick, this is a time to add power efficiency to your design process. It's a whole different mindset from functionality.
Can you get more efficient heaters, better insulation?
Does everything need to run all the time?
Maybe this is a time for external I/O hardware to offload the power requirement
The Pi is wonderful, but it's not a low power device. Is there a lower power device? (zeroW is marginally better but needs some work to get GPIO, iirc)
Do you need the gas sensors to be running all the time? Implement a relay that turns on the sensor, waits for them to heat up, take a reading, and shut them down for another ~60 seconds.
Even having to warm the heater, should use less power overall than leaving the heaters hot.
You need to decide how often the gas sensors need to be polled. There's no point polling every second or faster, but do you need 10 seconds, 1 minute, 2 mins, etc.
Another option is to use a dedicated PSU that can take 12V and supply the 5V and 3.3V needed. Something like a Pico PSU should be able to take voltage from your 12V battery and supply the load.
If finances and the physical environment permit, you could extend the runtime of your 12V battery source using a solar setup. A small 12V "trickle charger" may give a small time extension, or a large-enough panel could run the project indefinitely.
