I have bought several cheap DC/DC Step Up Power Module, 2A convertors based on MT3608, In=2-24V, Out=5-28V.
Plain version as this, without USB connector:
Even there is many videos on Youtube shows they can burned up easily, I had no such experience with. It worked fine raising 3.6V from 3xAA NiMH Eneloop batteries. The converter need to power 5V MCU based device with the most 200mA consumption.
However, since this is switching regulator, I'm considered a bit about reverse overvoltage when shut down (and/or power up) device with NiMH batteries, since I doubt it have any protection. Even I can use simply schottky diode on input and output, I doubt this is enough to protect the battery and 5V device as well.
I would need advice how to optimally protect both, without too much additional circuitry. Thanks.
You’re over-thinking things. You certainly don’t need any battery protection and unless you’ve investigated the circuit with an oscilloscope and have seen there is over voltage on the output side at startup or shutdown, you’re trying to solve problems that don’t exist.
That is actually a good catch, it appears you looked at the schematic which I pasted below. If either cap fails there is the possibility of getting zapped. It does not much matter which diode you use, with batteries your biggest concern will be leakage. when the power disconnects the polarity of the inductor will reverse and come out. in reverse polarity, that is why we use clamp diodes instead of snubber circuits which are primarily used for AC because of there by directional capability. This response is to help you get started in solving your problem, not solve it for you.
Good Luck & Have Fun!
Gil
This kind of reasoning is responsible for dirty cheap and extremely badly designed products which may fail or catch fire at any time. Of course that anyone who know how inductor works can see there is a quite big problem here, not naive at all.
Yes, exactly. The first thing I did was to find schematic and add a schottky at input. Anyway, voltage at input and output is not under control when switch on/off, especially when capacitor fails and they will certainly fail if not properly rated. The main problem is that voltage spikes can be many times bigger until IC get control and it is uncontrolled on switching off, in which case none of capacitors will be able to survive long enough if rated voltage is ridiculously low, as I'm suppose it is regarding size of it. It is only matter of time when all thing will fail and it is question how much energy is possible to store in the coil and how much the voltage spike can be...
Another aspect of the question is that I was not certain about solution of the clamping diodes with HiMH batteries I have also though about as possible solution. Similarly with a zener and resistor.
With NiMH AA batteries, it is not advisable that voltage drop under 1V, thus monitoring input voltage and activating power-down sleep mode of the MCU is mandatory. I had in the past experience that some industrially designed devices from which I have forgot to remove Eneloop batteries, drained it to the level the batteries become negatively charged on plus terminal...
herbschwarz:
Why be concerned about voltage spikes across the battery?
It will just absorb that energy as it would a charging current!
All upper I have wrote is consequence of the NiMH charging spec. Aggressive charging, as well as applying excessive voltage on charging terminals will damage the cells. With Li-Ion, applying excessive voltage will be much violent - it will catch fire!
According to the spec., NiMH battery cannot absorb excessive voltage, excessive charging current, excessive discharging rate etc. without serious damage and shorting lifespan.
I see no point ruining one of the best NiMH batteries on the world (3rd gen Eneloope in this case). Instead of that it is necessary to protect it adequately as well as the device itself. With a lack of deeper knowledge in electronic, that is exactly reason for this thread.
