I made this video in two parts while travelling, but now wish I had the other parts or some suitable test equipment to test my theory that it might have a design defect that could overcharge the lithium cell if the light is left in sunshine while switched off.
Update - I've run some tests and the unit does seem to lose control of the charge when the switch is off. Fortunately in most situations the light will be left switched on continuously after purchase, and will therefore have the charge control circuitry active.
The use of a processor controlled transistor for charge control is very similar to the circuitry of the skeleton shaver I reverse engineered a while ago. It makes me wonder if the chip is a microcontroller with a built in dedicated lithium charge system or if it's just relying on software and an analogue to Digital converter to measure the cell voltage.
My pondering about whether the chip could somehow use a single pin to measure cell voltage as well as control the transistor is possibly viable if the cell voltage and transistor base voltage were added. Technically speaking the chip could contain a simple zener or fixed voltage reference so the transistor turned off as the cell voltage reached a charged state.
Theoretically the base bias resistor is also going to have a hard time on a sunny day when the base is shunted, as the power dissipation would be just above its rating and the processor would also be shunting about 50mA on its control pin.
They may be over-economising on these designs.
If you enjoy these videos you can help support the channel with a dollar for coffee, cookies and random gadgets for disassembly at:- https://www.bigclive.com/coffee.htm
This also keeps the channel independent of YouTube's algorithm quirks, allowing it to be a bit more dangerous and naughty.
#ElectronicsCreators

Poundland's finest solar-powered cable light. This light acts as both a solar garden light in the sense that if I turn the lights off, well, I'll turn it off. right now. it lights up at very low intensity, but then you turn the light back on and watch your eyes.

the light's about to come back. Uh, it can also be used at much brighter intensities. by pulling this cord, it goes from off to low level and then high level and that's not too bad. Now the power source if you take this off here is a 18650 stuffed up the end and what? I've got right now to turn this off so you can actually see what's in there and what I've just done right now because uh, I got a top tip on how to remove this there I Have basically taken the sponge around the outside and look at this solar panel.

That's nice. Why? why is solar diameter 115 millimeters? Very nice solar panel and what we have inside here is Hotmail glued an LED PCB and there is the control PCB and the lithium cell. The control PCB will have probably well, I'd hope the means to regulate the charge from the solar panel, but also it's got just literally glued onto the end with a bit of wire. It's got a little micro switch here that toggles through the intensity settings.

That's quite neat. Okay, well you know what happens now. I'm going to flip the circuit board out and we shall reverse engineer it one moment. please.

Once again, we have a video that is in a slightly different location for the second part because I'm traveling with work and I have to say this is a something that I wish I did of the rest of this light because there's something very odd about the circuitry. I'll let you I'll let you leave your comments down below what you think it is and once I've described the situation whether you think this is a valid functional circuit, but anyway, the back of the circuit board is fairly boring. It's just some tracks and it's got an on off switch which disconnects the battery from the main positive reel. But there one thing it doesn't disconnect is there's a little track here which is part of the charging circuitry and that is a bit of a problem unless this chip is very magical indeed.

On the front, we have the mystery chip and we've got two transistors. One of the transistors is switching the LEDs via these two parallel resistors and the other transistor is controlling the output from the solar panel to the input of the lithium battery the Lithium cell. Um, and this transistor is normally turned on by the solar panels. I'll show you in the circuitry and it gets pulled off by this uh, microcontroller pulling this pin low and uh, that turns it off.

Perhaps peeled off was not the best at word to use there. There's a couple of unpopulated resistor positions. They are our voltage divider across the solar panel that's probably used as a dust dusk sensor for the their chip in here. and there's the pill core toggle switch with just a little micro switch and all its acting is a simple input to the microcontroller.
Okay, let's take a look at the schematic and I'll show you what I have found here, which is a bit odd. I'll Zoom down this right. So here's the sort of panel and it charges the Lithium cell via this transistor. The transistor is normally turned on by that 100 ohm resistor and as I mentioned earlier to, the only way it can really turn that off is for the microcontroller to pull the base.

This transistor law. Now here is the odd bit. The switch between the lithium ion cell and the rest of the circuitry is here, and if you turn that off, it disconnects, it turns this microcontroller off, and and theoretically then there's no way it can sense the battery voltage or actually control the charging. But there is a way it could theoretically do it if I was being inventive and it may well do that.

Who knows if some of these chips are very odd functions, but the LEDs are controlled. Here's the cluster of LEDs. There's two 3.6 ohm resistors giving approximately 1.8 ohms and a transistor, a g3y switching to the zero volt rail. Let's just Mark that zero volt and and it's got one key resistor between it and the microcontroller.

The microcontroller also has that toggle switch, the little pill cord switch, which is just basically it activism traction button, and there's a little decoupling capacitor across the mega controller for stability. Now, the only way I can see that if you get this switch off, that the microcontroller can actually stop this solar panel from charging this Lithium cell up to about 5 volts is if there's some bizarre circuitry in here that uses this connection here. If it made when open circuit, it could detect the open circuit solar voltage. Oh no, that's not going to work either.

This I was trying to invent ways this might work that it could measure what the solar panel was being pulled down to as it because it will be dumping all its current into this Lithium cell. Um, but I I came up with fancible ideas that it's basically a separate section of circuitry that monitors the voltage and then shunts it to the zero volt rail itself. But none of this works. I Think this is a design flaw.

that's why I really want the rest the circuitry. So I can basically hook the solar panel up again and just measure the voltage being sent out to the lithium cell. even maybe put a small lithium cell or a capacitor and output just to actually measure what voltage it charge is up to. But as it stands I don't see any way it can actually stop the charging of that lithium cell, which is strange.

Um, so if you have one of these I would recommend that you leave that little switch turned on all the time. Uh, just in case, especially if you're in very sunny areas, particularly outdoors in a perhaps a greenhouse or something like that where there's a lot of sunshine. because that at least means that, uh, hopefully the discharge at night time of the lithium cell is not. It's going to keep it a sensible level and not let it get overcharged.
I have to get the rest of the circuitry again, But um, I'll let you I like you? yeah? Leave them your own thoughts in the comments below as to how it could be doing this because uh, if it is actually monitoring in any way or if it is actually a design flaw, it's very strange. It would be an easy mistake to make if the people are just putting sort of Lego designs together, but there we have it. It's interesting and strangely exciting. Tour.


12 thoughts on “Perplexing solar light”
  1. Avataaar/Circle Created with python_avatars chatrkat says:

    Interesting little lamp. Never expected the circuitry would be that complicated, ๐Ÿค”

  2. Avataaar/Circle Created with python_avatars Bugkillahz says:

    18650 has a VCC which protects from over and under charging.

  3. Avataaar/Circle Created with python_avatars Dimitar Nikolov says:

    Definitely a design mistake. It could still work tho if the open circuit voltage of the panel is 5V and the MCU consumes 12mA (unlikely) trough the 100 Ohms resistor, that wold leave 4.8V on the base of the BJT and 4.2V at the emitter.

  4. Avataaar/Circle Created with python_avatars TheAdminFromHell says:

    All of my 18650 lamps use a protected battery, is this different?

  5. Avataaar/Circle Created with python_avatars Michรฆl Alan Baker says:

    Nope. I donโ€™t like it. ๐Ÿคจ

  6. Avataaar/Circle Created with python_avatars Stephen Belcher says:

    Mystery Chip ?

  7. Avataaar/Circle Created with python_avatars Julian Tyrrell says:

    The io pins should have ESD diodes to the power pins, so it may power the micro via the 100 ohm. As you say, measuring what it does would be the best test. Also check what the solar cell outputs with light intensity. The model I use is a fixed voltage with a variable output resistance that is inversely proportional to the light intensity.

  8. Avataaar/Circle Created with python_avatars Thomas says:

    Might the uC disable charging altogether when the switch is off, due to diodes in the uC body?

  9. Avataaar/Circle Created with python_avatars Bhaskarjyoti Gogoi says:

    Maybe one of those resistors is limiting the current to the battery? A cheap way of slightly mitigating the problem of overcharging and resulting degradation. Ignoring overcharging is a very common practice with all kinds of such cheaper electronics.

  10. Avataaar/Circle Created with python_avatars AA7YA says:

    The overall design of the lap is odd for an outdoor path lamp. When I saw the thumbnail for this video, I instantly thought this lamp was actually a small bedside table lamp because of the chunky base.

  11. Avataaar/Circle Created with python_avatars justpaulo says:

    I think itโ€™s a flaw but I also think that the bipolar transistor will introduce another 0.6 to 0.7 V voltage drop between the solar panel and the battery.

  12. Avataaar/Circle Created with python_avatars TheReal ST says:

    Can you just cut a trace and jumper to the other side of the switch?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.