LED block inside

Hello. I hope this is not too novice question. IKEA got stuck with piles of garden solar lighting (their product no 21777) which they sold for 1€ each.

Looking at the back (large solar panel, 2000 mAh NiMH battery, switch etc.) I took seven home (I wish I had taken 20...). I Took one apart to play with it and found a 3-LED block which gets 1.2 V and produces very strong light. What am I missing?

The block does not appear to have any electronics on it apart of the three parallel LEDs. How does the LED work on 1.2 V?

The product IKEA 21777

battery and module-1 Battery and module-2

Hello again. I think that SamGibson's answer (with the Scope wave pic) is the right one. The battery is normal size as can be seen in the pic (it also say on it HR6/AA) and the module does not have any electronics on it. BUT the 1.2V DC I measured seem to be wrong (I use basic digital multimeter). as an experiment, I decided to connect the module outside, directly to the battery poles. Nothing !! but when I connected to the circuit it works fine. Changing the setting on the multimenter to AC shows 2.5V. As it is a basic multimeter, I assume that this is also very not accurate, but the the basic experiment shows that there is some sort of boosting circuit on the main board to drive those leds. (Now I just need to figure out why this does not drive my 5mm, flashing blue, LED...) Many thanks to all that help and contributed. All the best.

  • 1
    \$\begingroup\$ What's the PCB in the background of the upper photo? Give us a clear view of all the connections. \$\endgroup\$
    – Transistor
    Sep 1 '19 at 12:51
  • \$\begingroup\$ Have you measured the voltage at the LEDs? \$\endgroup\$
    – Reinderien
    Sep 1 '19 at 13:00
  • 1
    \$\begingroup\$ The LEDs appear to be connected to the little PCB in the background. That may have a simple boost converter on to raise the voltage from 1.4 to over 3V to drive those white LEDs. \$\endgroup\$
    – JRE
    Sep 1 '19 at 13:25
  • 1
    \$\begingroup\$ A boost converter may not even look like what you might expect. It might look like a simple transistor, and the associated inductor might look like a resistor. \$\endgroup\$
    – JRE
    Sep 1 '19 at 13:29
  • 2
    \$\begingroup\$ It's almost certainly something like the YX8018 in each light. It might be a chip-on-board or something else. But it's there. \$\endgroup\$
    – jonk
    Sep 1 '19 at 21:10

The block does not appear to have any electronics on it apart of the three parallel LEDs

There will be electronics, since you can't light a white LED directly with the 1.2V from a rechargeable cell - around 3V is needed to light a typical low-power white LED with a low current (a higher voltage is needed for higher-power white LEDs). This has been covered in previous questions on the topic.

As mentioned in comments, you can expect to find a small boost converter as part of a solar cell LED controller - it might look like a 4 pin transistor with an external inductor, or it might be under a black epoxy "blob" (COB), again with an external inductor.

One example of such a component (there are many different types you could find in your solar light, so this is just one example) is the YX8050 - a typical solar light uses the schematic 1-2 below from that datasheet:

snippet from YX8050 datasheet

On my bench, I just captured this LED drive waveform from a YX8050 with a single white LED in a solar garden light (battery voltage = 1.2 V):

LED drive waveform from YX8050

As you can see, the voltage to the LED reaches 3V with a duty-cycle of around 40% at a frequency of 788kHz. So the LED is flashing, but it's flashing so quickly that it appears to be lit continuously, due to the persistence of vision.

If you try to measure the DC voltage across the LEDs with a multimeter, you will measure a much lower voltage (I measured 1.2V), even though the LED is being driven with 3V pulses. If you could use an oscilloscope to measure the LED drive waveform on your LEDs, you would see something similar to the waveform above.

  • \$\begingroup\$ 788 KHz? I bet you could detect it with a broadcast band AM radio... Something with a BFO would be better but I bet you could hear a distinction with the battery in vs removed to disable it. Of course the askers may not be at that frequency. \$\endgroup\$ Sep 3 '19 at 18:58
  • \$\begingroup\$ @ChrisStratton - "Of course the askers may not be at that frequency." Agreed. I also saw that the frequency changed with varying battery voltage. So although I saw that frequency during my specific test, I would not be surprised to see different oscillation frequencies even with other YX8050-based lights, let alone those based on other similar ICs. \$\endgroup\$
    – SamGibson
    Sep 3 '19 at 19:19
  • \$\begingroup\$ Thank you very much Sam Gibson. your reply seem to be right on target !! (sorry Im still struggling with the format of the questions/answers/comments) \$\endgroup\$
    – AR57
    Sep 5 '19 at 11:44

LED's have a forward voltage, this can be tweaked by the manufacturer for different purposes, generally below 1 V for visible light is hard, but possible to buy (likely infrared or red with a phosphor to bring it up to visible)

The battery has a certain amount of internal resistance, being rechargeable generally means it is lower than non rechargeable batteries, it is likely each LED can handle enough current without damage to essentially have the battery's internal resistance act as a current limiting resistor

This is assuming that green circuit board out of frame on your picture doesn't have some small LED constant current driver.

  • 1
    \$\begingroup\$ How can a phosphor increase the frequency of light? I think this only works the other way around, unless you have some kind of nonlinear optical effect. So I doubt you get visible light from an IR LED with phosphor... \$\endgroup\$
    – jusaca
    Sep 1 '19 at 15:03
  • 1
    \$\begingroup\$ Please, please point me in the direction of a white LED that runs from 1.2V \$\endgroup\$
    – Finbarr
    Sep 1 '19 at 16:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.