I seem to have run into a problem with powering Arduino from a power bank (external USB battery). Arduino seems to be running fine for 30 s, then the power bank switches off. This happens with two power banks: Yoobao YB-659 (maximal discharge current 1 A) and Maxell MPC-R5200BK (maximal discharge current 2 A). My setup consists of an Arduino Uno, a data logging shield, a temperature/humidity sensor, and an LCD display (whithout the background lighting); all in all, it should be WAY beyond 1 A. Both power banks can successfully charge a cell phone, a e-book reader etc. What is going on?
My guess is that in addition to maximal discharge current, power banks also have some minimal discharge current, and shut off if the discharge current is below this threshold. They treated my Arduino as "leakage" and shut off. Does this make sense to anybody? Can somebody recommend a solution? Ideally, updating a firmware of the power bank...
My guess is that in addition to maximal discharge current, power banks also have some minimal discharge current, and shut off if the discharge current is below this threshold. They treated my Arduino as "leakage" and shut off.
Yep. That's exactly what the manual says - though not in so many words.
Well, the "power bank" contains a 3.6V lithium battery, so to put out regulated 5V, it contains a power up-converter. The power converter is not perfectly efficient so it uses some power to operate even if there is no load, and of course, you are running the indicator LEDs. To prevent the power converter itself and the LEDs from discharging the battery, it does not turn on at all until you press the "Power" button, and if it detects that it is not being used significantly for 5 seconds, it turns itself off to prevent discharge.
This particular arrangement works for my XiaoMi PowerBank (20A).
Total power drawn is roughly 3.5mAh (per hour), so its not too bad.
Assuming it runs at 5v (from the USB)... 0.0035Ah x 5v = 0.0175Wh
If you run it for 24 hours.. 0.0175Wh x 24 = 0.42Wh
Assuming the powerbank only holds 15A (x3.7v = 55Wh) ...
this circuit will finish the battery in about 4 months.
What's so strange about it? It's convenient (except for the auto power off). And there is nothing charger about a power bank, it's just a 5V voltage source. All the charging magic needs to be done in the device connected. That's why a power bank isn't called mobile charger And yes, the name USB charger / phone charger when you mean this is misleading / "wrong". It's simply a USB power source.
You don't need external circuit, you can use your Arduino itself for keeping the power bank running. Just put a 22R resistor between 5V of USB in and a pin (e.g. 1C) of ULN2803. Then supply signal to pin 1B of ULN2803 from Arduino for 250 ms in every 5 sec. This signal "sinks" the current provided by the power bank through the resistor. For me this configuration is working fine.
So the only external component you need is the resistor and a ULN2803. You can set the optimal timing of the signal in the sketch to find the minimal power consumption which still keeps the power bank running.
To keep it real simple, I added an LED & resistor to my circuit off to the side and it apparently draws enough to avoid the auto shut off problem. Obviously uses constant power so depending on your application this may or may not be optimal.
I've found a very simple solution, that works at least with my power bank (Silicon Power SP power S105, sold by Costco).
This power bank does not switch off when it is being charged (through the micro-usb plug).
So, I just make it believe it is being charged by supplying 5V to the power pin of the micro-usb cable... Since connecting directly the micro-usb input to the other usb output or the 5V pin of the arduino would draw a lot of current (charging the power bank with its own power output!), I connected a 27 ohm resistor in between (I've not tried with higher values for smaller current yet).
It worked: no switching off for a whole afternoon and night while running an arduino uno measuring a chinese calliper, and very little power wasted. On the other hand, I had no luck with the pulsed load solution with my power bank...