> I don't know if it's legal to put a capacitor behind LM1117 regulator.
Oh gosh yes. By default you want at least a 0.1uF ceramic close to the output of a regulator, and the additional 100uF electrolytic is good. Then since it's a one-off and not worth worrying about, I'd put another 0.1uF close to the input of the regulator, and another 100uF across the battery. And if you're still having problems at that point, possibly a small resistor in series with the buzzer to limit its inrush current. You could also replace the mechanical buzzer with a piezo buzzer (or piezo speaker and generate tones with the avr). (PS if you're stocking up, you might as well opt for 1uF ceramics and "low ESR" electrolytics for these purposes)
Thanks for the comment! The one capacitor I put seems to be enough for my needs for now :) But it's great to hear what's the correct thing.
> I'd put another 0.1uF close to the input of the regulator, and another 100uF across the battery.
Interesting. I (naively) thought that a capacitor near a LiPo battery is illegal as you're not allowed to charge the battery directly. I thought when the battery power drops a capacitor will be doing exactly that. Can you explain why that's untrue?
Update: in other words: What's the difference between putting a capacitor near the battery and "charging"?
Well I'll just first disclaim that I don't have any design experience with Li-Po batteries, although reading about them I don't see any odd requirements. The guideline to not "charge them directly" is in contrast to say lead-acid batteries, which you can simply hook up to an appropriate voltage supply and let them sit indefinitely ("trickle charging").
To the extent that you need to worry about it, the voltage on the two will be the same. The battery power only drops as you're drawing current. The point of the capacitor is to smooth out changes to the load by supplying current "first". Your capacitor will never get "more" charged than the battery somehow. The main difference versus "charging" is that you have no external energy source.
From a slightly more advanced perspective, the reason for the capacitor is to smooth out the (possibly bidirectional!) transient loads on the battery. It's quite possible that your "inductive buzzer" (depending on how it's driven) is actually backfeeding a little current through the arduino and onto the supply rails. Your battery will never be overcharged this way because any such energy came from the battery itself, but smoothing out the spikes with a capacitor eases the stress on the battery. In fact, I'd be surprised if the arduino board didn't have a decent-sized capacitor immediately on its supply lines, and this is part of the reason.
This is limited electronics knowledge, plus conjecture. Buyer beware:
- Direct (unmanaged) LiPo charging is to be avoided because the cells react so violently to overcharging.
- Compared to a LiPo cell even a 100 uF capacitor contains essentially zero energy.
The capacitor won't overcharge the LiPo, because it would almost immediately expend the charge it stored
Oh gosh yes. By default you want at least a 0.1uF ceramic close to the output of a regulator, and the additional 100uF electrolytic is good. Then since it's a one-off and not worth worrying about, I'd put another 0.1uF close to the input of the regulator, and another 100uF across the battery. And if you're still having problems at that point, possibly a small resistor in series with the buzzer to limit its inrush current. You could also replace the mechanical buzzer with a piezo buzzer (or piezo speaker and generate tones with the avr). (PS if you're stocking up, you might as well opt for 1uF ceramics and "low ESR" electrolytics for these purposes)