The ebay 10$ solar powered flashlight FM radio and mobile phone charger

On Wed, 29 Jan 2014 12:03:03 GMT, Jan Panteltje Gave us:

Hahahah... Jan tells a funny.

You should buy another, since tearing that one apart means you wont be able to prove it doesn't work. ;-)

So, describe this magic circuitry between the charging source (the solar output) and the "DC rail", and this LED indicator is on that rail as well?

On a sunny day (Wed, 29 Jan 2014 04:21:30 -0800) it happened DecadentLinuxUserNumeroUno wrote in :

It is perhaps something like this: -------------|>|--------------------------------------------- | | Schottky | | | | | [ ] --- red | | | |+ | 1M \ / charging | | radio | solar | --- LED | 0 / on / off 0 / flashlight panel | | | 0/ switch 0/ switch 4.35V oc | [ ] | | | | | | | | | | | |/c | FM radio voltage --- | --------------| NPN ---3.8 circuit converter | | |\/ e - Lipo | | LEDs | | | battery | | | ---------------------------------------------------------------------

As soon as the solar panel gives more than .7 V the NPN comes on and lights the 'charging' LED via the battery. So any room- or low level light will put the LED on, and discharge the battery, not charge it....

Fun, how big is the sloar cell? (~5cm X 5cm) seems like it should do more than 1mA.. but it'd need some 'smarts' to get to the right place on the I-V curve for max power. (and a DC-DC switching thingie to match the solar cell to the battery.) George H.

On a sunny day (Wed, 29 Jan 2014 07:52:06 -0800 (PST)) it happened George Herold wrote in :

Approx. 4x7 cm

This was inside in the window, with cloudy sky.

I thought it was neat they used a 4.35 V solar cell then a .15 V Schottky, leaves 4.2 V, exactly right for these small Lipos. That is the maximum allowed to charge a Lipo like that.

Not needed here, and would decrease performance due to conversion losses.

For the cellphone charger output it looks like they use a DC /DC converter from the battery to get to about 5.25V. I cannot imagine this thing to ever charge a cellphone directly from solar, but there is a switch for that, but have not looked at that part of the circuit. The manual talks about many many hours in Chinglish... is not very clear at all.

Hmm but if 4.35 is the open circuit voltage then the operating point voltage will be less.... (by maybe a volt or 2? I don't really know. Just guessing from numbers on digikey

I don't know about LiPo batteries either. (one might wonder if I know anything?) But won't they want different voltages depending on the state of charge?

George H.

On a sunny day (Wed, 29 Jan 2014 08:42:29 -0800 (PST)) it happened George Herold wrote in :

Yes, you are right, I remembered that after I wrote those words....

That sounds about right. I get only about 5% to 10% solar power through an IR and UV blocking window. It's not enough to drive a bright LED from small cells.

Hopefully there's a shunt regulator. I've read that 4.1V is correct for cases when the precise 4.2V charge cycle is not possible. The specs for the batteries I've purchased say that 4.2V and above will eventually lead to an explosion.

On a sunny day (Wed, 29 Jan 2014 16:49:36 GMT) it happened Jan Panteltje wrote in :

PS I have reconstructed the circuit diagram from the PCB of this solar powered thing, and it is a bit different than I thought.

The current that always drains the battery comes from a simple boost converter chip 2108A that is NOT the Analog Devices ADP2108, that is a step down, this seems to be a step up from the LiIon 3.8V nominal to the 5.5 V output connector to charge a cellphone. It is always in circuit, and always draws some current, even with power switch off. Does anybody from the picture of that 3 pin chip recognize the manufacturer logo:

Today we finally had some moments with sun, and I measured the solar module again: 6.12 V open circuit 22 mA short circuit measured inside, could well be one of teh modules in that link you gave, The module has GX74X45 written on it, could be size in mm.

I have been googling, it seems these Liions (lions?) can safely be used at that voltage, because the internal protection board it has (clearly visible, transparent foil wrapped around it) limits over voltage, under charge, and what not. I am just parroting what I found out in some forums. This same boost converter chip seems to surface in other solar USB chargers too. Just wonder if it can power my EM-411 60mA PS module from the Liion... Nice full bridge audio amp in it too... As mentioned earlier, apart from teh boost converter always sucking power, the 'charging' Led also draws power as soon as some light (artificial room light works) hits the solar panel. That makes -2 for charging. But it does provide some interesting parts. How about: Never put LEDs in parallel, here they had no problem either, and a 10 Ohm series resistor makes for an efficient ???? flashlight drive. not a bad deal, all those parts for 11$, The solar panels alone in your link are more expensive... not even counting shipping.

On a sunny day (Sun, 02 Feb 2014 12:24:10 GMT) it happened Jan Panteltje wrote in :

PS Way cool, that unknown converter chip 2108A makes 4.99 V from the Lipo (eh that is what it is) and indeed has not the slightest problem to power my EM-411 GPS module at about 33 mA (Lipo then delivers between 50 and 70 mA at 3.88 V; efficiency...). That chip seems a sort of bang 'on' bang 'off' flyback, I measure about 10 uS long pulses followed by wild varying delays, typical behavior of such a regulator at low loads.

Now for the amazing part, while looking for a place to place the GPS module in the solar case, I found one on top, but want one of the OLED modules to go there. I decided against all logic to try if the GPS module would keep lock on the sats while mounted _under_ the solar panel. And it does! I has sort of expected the silicon or whatever is in the solar panel to shield the RF, but it seems it has little or no effect, However I move it relative to the GPS antenna, the GPS still stays locked (and has fast lock-in times too). So this deal is getting better all the time.