Where to get LEDs rated for 1 to 1.5v

Question

I have tried to repair a solar flickering garden pagoda light. Unlike all the other types of that device I own, this one uses one AA NiMh battery rather than two. It’s LED needs replacing because one of its wires had broken where it entered the “glass” head.

I got some 5mm wide-angle straw hat (stubby length) yellow/amber LEDs from Lighthouse LEDs last year, but illness kept me away from the task until this week. The replacements from Lighthouse are 2v, but I guess I’d assumed that was a maximum rating. Unfortunately no, Lighthouse now informs me they need at least 2 volts to light up, which on reflection makes sense. It explains why the new lamp I soldered in wouldn’t light up... I initially thought I’d ruined the lamp with solder heat, and soldered in another using multiple heat sinks. It didn’t light either. Then I touched leads from a fresh AA battery to a lamp fresh out of the shipping bag, and it wouldn’t light. Ah ha moment... dumb me. My assumption that 2v was a max, because it was the lowest voltage lamp I could find, was ill advised. But being an LED neophyte I seem always ignorant of their technics, as numerous faulty Xmas LED strings lying comfoundingly unrepaired in a bag will attest.

I think the circuit board controls the flame-like flickering because the original lamp was a straw hat. But just in case, I also got from Lighthouse some standard 5mm self-flicker lamps at the same time. What I can’t figure out is how the original LED worked on a single AA battery. Lighthouse says I shouldn’t need a resistor in the circuit, but there is a coded red-green-red-silver one on the board. Perhaps that is to control the current coming in during midday from the solar panel? Regardless, surely the original LED must be a 1 volt lamp? But where can I purchase such an animal? Lighthouse claims to not be aware of any such device.

I guess I could try adding a second AA NiMh battery to bump the unit to 3v, but the box is only big enough for one and I don’t know but what the diodes and other CB components may not play well with 3 volts, nor if the panel has enough output potential to fully charge two NiMh AAs, even if they were only 600mAh ones.

Thanks in advance for any thoughts,

Joel

genebren · Accepted Answer

Joel, There are not many (if any) visible spectrum LEDs that will run directly from a single AA battery. But, most of the inexpensive solar/LED devices, using a single AA battery use a device (like cx2601) that has a voltage boost inside. I suspect that the 'green' component, with multiple stripes is an inductor (part of the boost circuit). I believe that if everything is working correctly, a normal LED should work fine. There are a lot of articles on the web concerning solar/LED lamps, that could help you identify the components on your LED driver board (and battery charger). Search on cx2601 or 'Solar LED hacks' for some interesting read (and potentially some help). Good luck! Gene

jw0752 · Answer

Hi Joel, I was set up to do some experiments this evening and so I decided to take apart a cheap dollar store outside LED light and check the voltages. I knew that they operate on one 1.25 volt NiMH battery so these LEDs have to be similar to what you are looking for. Because of how cheap they are I expected to find an LED that was running on 1.25 volts. This was not the case however. The unit contained a special 4 leaded IC YX8018 which in conjunction with the rest of the sparce components (1 inductor) produced square wave output to the LED of 3.72 volts. Here is a picture of the output as read across the LED. This is more than enough voltage to light a standard LED. The pulsing also serves a secondary purpose by extending the life of the battery since power is only applied to the LED for a percentage of the time. Incidentally the image on the scope is inverted as I have the leads of the scope reverse polarity on the board due to the difficulty in getting the large ground clip into place on the negative terminal. If your LED isn't lighting there is a pretty good chance that there is a problem in the circuit. John

hafcanadian · Answer

Thanks to all who’ve responded. The idea of an inductor that boosts voltage makes a great deal of sense. I can see where the AA battery would not work directly, and my suspicions that the 2v rating for the LED means just that and is not a “max” is confirmed. In my relatively simple mind I envision the inductor building voltage through resistance in a nanosecond, releasing it at 2 volts that activates the LED emission, and repeating the cycle producing the effect of the candle flicker. The self-flickering LEDs I “just-in-case” bought shouldn’t be necessary. I am aware of the polarity for the LED wires and was careful to correctly solder the black and white leads to them. My biggest inhibitor is my essential tremor. If you don’t have that malady yet, pray that you never do. It makes it exceedingly difficult to hold a soldering iron, solder, and wires steady enough to get a clean result anymore. I try using normally spring-closed tweezers and other heat sink devices to hold parts together, and brace my hands on things, and when workable an alligator clip/magnifier/wand holder stand, but it’s still a frustrating challenge. So forgive any sloppy-appearing workmanship in the pix. It would appear that the green device I thought was just a plain but large resistor is more so the inductor “coil”. Although it does have continuity across it, perhaps it’s not working to boost the voltage as intended. Or as you’ve suggested, one of the other components is faulty. There’s only 2 diodes, a 4-pronged transistor (I think), and whatever the flat round brown component is. The diodes have continuity in one direction okay, so I reckon it could be the transistor, which if what I’ve read is correct may be involved in the boost function? I’ll also use my Optivisor headband magnifier to more closely examine the board for a micro-cracked or slopped solder point. As per Gene’s idea and cx2601 reference, I checked this article: 1.5 VOLT WHITE AND UV LED DRIVE CIRCUIT . Its over my head in a lot of ways, but does offer insight that I can appreciate. Shabaz, Douglas, and Gough... all your input has also been enlightening, so I now understand a lot more than I did. It may come in handy in dealing with the other solar garden lanterns of various kinds in the future... a year doesn’t go by without at least one of them going belly up for some reason like corrosion, old batteries, spiders, or needing a new solar panel. Some flicker (the best ones) and some don’t. And to John, who helped this past week on another frustrating electrical repair project, another big thank you for the time and expertise you’ve contributed. I can take a couple more closeup pix of the cboard if anyone would like, appreciating any more analysis. They are the simplest and most inexpensive lanterns we have and I’ve been tempted to just chuck them, but their effect at night is the most realistic of all. It’s just that they have a history of not coming on in the evening after a day or two, which was frustrating, and why I tried disassembling one last year. I suspect the panel in full sun all day isn’t enough, despite needing to charge only one battery (all our others use two). Cheap lanterns or not, my curiosity and determination to learn and repair stuff perhaps has the better of me. Joel addendum: Okay, I always thought a capacitor was barrel shaped. Apparently the brown disk shaped component I referred to in a paragraph above is known as a disc capacitor. I see them in most circuit boards, but didn’t know what they were. I’ve known “barrel” capacitors to fail, so why not this disk version, esp. if it’s supposed to store the built up voltage, and isn’t.

Gough Lui · Answer

As mentioned earlier, this circuit from a solar light is a fairly common one and uses an inductor to boost the voltage to run standard white LEDs of about 3V forward voltage drop. Unfortunately, due to the nature of physics, to get LEDs for 1V to 1.5V would mean they would emit mostly in the infrared region, as the forward voltage drop is related to the P-N junction band gap and hence the energy of the photons emitted. For example, Deep UV LEDs often require 5.5-6.5V to run, whereas IR remote control LEDs can run at about 1.5V. For more information, you can see Wikipedia under Light-Emitting diode - Colours and Materials . - Gough

shabaz · Answer

For some reason that shade of green ("vintage kitchen green" is a popular color with inductors. Don't know why : ) Hi Joel, For the reasons Gene mentions, it is not possible to test the LED with the battery alone, i.e. it will only work when connected in the circuit.

dougw · Answer

Stanley makes red LEDs that work at 1.5 V. Maybe the circuit is boosting the voltage as it flickers. Try it with your 2V LEDs to see...

jw0752 · Answer

Hi Joel,

Here is the likely circuit for the 4 pin IC which I believe is a YX8018;

Here is the link to the Google Picture gallery for the YX8018:

https://www.google.com/search?q=YX8018&tbm=isch&tbo=u&source=univ&sa=X&ved=2ahUKEwih78z--IHeAhULiIMKHQzxDmMQsAR6BAgDEAE&biw=1920&bih=934#imgrc=_

If it isn't connections or a bad battery the IC is the likely culprit as Doug mentioned. The stage of rust on the leads pictured is not likely the cause of the failure.

John

Gough Lui · Answer

Forgot to mention - LEDs are polarised components. The amber colour requires less voltage than the white that you are replacing and should work. There's a good chance you may have hooked it up backwards, your solder joints aren't good, there's a broken wire somewhere or the battery contact is a bit duff. But replacing white LEDs in solar lamps with coloured LEDs generally works fine. On an LED, the flat side of the unit leads to the negative side, and the long leg is the positive side. You can try checking with a multimeter to ensure your polarity is correct as getting it wrong will mean no light, even if the voltage/current is sufficient. - Gough

14rhb · Answer

Hi Joel, An interesting post - I'm sure countless numbers of these, often cheap, solar lights end up in landfill/incineration but luckily not yours...so well done trying to repair them. Lots of good comments already re: the voltage boosting. A non-technical term for this type of setup is sometimes known as a Joule Thief . The phrase was coined to detail a circuit that extracts all the energy from a single cell battery...and the standard designs work well. Rod

dougw · Answer

Discoloration of the pins in itself does not mean it has failed, but it is one of the most likely components to fail in this circuit.

Since it has 4 pins, it is a more complex chip than a simple transistor.

I would first try re-soldering all the solder joints and closely inspect all wire connections.

Then try operating the circuit with a known good power source.

If that fails, you may need to start comparing what is happening on each pin of the chip with a what is happening on corresponding pins of a working circuit.

Gough Lui · Answer

Disk capacitors are generally ceramic type, owing to its material nature, it's almost like a solid layer of ceramic with conductive metal layer deposited on the ceramic dielectric to form the capacitor. The assembly is then dipped in some protective paint/coating. It's a simple construction that lends itself to fairly reliable low-capacitance value capacitors, although if moisture gets in, the conductive metal layers could (in theory) oxidise and stop acting properly as an electrode, with consequent reduction in capacitance value. The round "barrel" type is likely to be an aluminium electrolytic capacitor. These are generally much larger value capacitors formed by "rolling up" electrodes soaked in electrolyte, insulated with plastic film. This assembly has much larger area resulting in higher capacitance, which is then shoved into a metal can with a sealing rubber bung from which your two terminals protrude (for the radial type, anyhow). These capacitors typically get used in power supplies and often fail because of a number of reasons - namely that they deal with high amounts of ripple in power supplies (a stressful application), that they (like every non-ideal capacitor) have an equivalent series resistance (ESR), that the ripple current through the ESR causes the capacitor to heat up, and that the electrolytic capacitors are sensitive to heat as this causes the rate of evaporation of the electrolyte to increase which also increases the ESR (which forms a runaway feedback loop) to my knowledge. As a result, these capacitors often "cook" and blow their vents (low-quality electrolyte formulation doesn't help) or they end up drying out (sometimes due to poor seals around the rubber bung) which reduces their capacitance to non-useful levels. - Gough

genebren · Answer

Joel,

I am not sure that the pattern that John captured on his scope will explain the flicker LED effect. It is possible that his solar LED has a different chip than yours, but the pattern shown appears to be oscillating at a fixed pattern which is well above the eyes ability to see. It is oscillation pattern is likely only used to boost the voltage. Having said that, it would be possible to design a chip that modulated the switching pattern of the boost circuit to create a varying intensity on the LED (I just did not see any evidence of that in the switching rate of John's LED).

Issues that might be effecting your LED lanterns:

1) The batteries might not be able to hold a charge (or charge in the circuit). Try another battery, or an external power source. (Note: this simple devices do not follow manufactures specifications for battery charging. They simply act as a simple trickle charger. This may not allow the batteries to deliver the full charge available. They do not limit charge or protect against over voltage)

2) Along the line of the batteries, but there may be corrosion on the battery terminals that is causing a poor connection. Clean the battery connections.

3) The solar panel may not be functional. Try a fully charged battery.

4) The LED that you have used may have higher forward voltage than the circuit is designed to provide. In the lower current range of LEDs (<100mA), about 60% of those are rated below 3.5V. Have you tried to illuminate the LED with a power supply or multiple batteries (with a current limiting resistor) to see what the turn-on voltage is?

There could be other issues that are more difficult to diagnose (or guess) without further information.

Good luck in getting these lamps to light.

Gene

hafcanadian · Answer

Got distracted by other responsibilities for a few days, and haven’t tried yet to resolder anything, but found a minute to review the thread, and should note what just occurred to me is a missing component in the diagram John provided.

Rather than a light sensor via the panel as some units use, this one has a separate half inch round one imbedded in the lamp body opposite the solar panel. I haven’t had time to refresh my memory of the wiring, and don’t know exactly where in the diagram one would expect to see that light sensor. I do know it works though, as I had to cover it with electrical tape so my workbench lighting wouldn’t keep the circuit off while I was testing operation.

Whether faulty or taped over, it shouldn’t be keeping the circuit off should it? Also I don’t see the on/off switch in the diagram, so I reckon my circuit’s only a tad more complex.

Joel

Also I bent the IC up enough to read “SS550A” on its underside. Perhaps this diagram I found is closer?

https://goo.gl/images/KWBptN

Where to get LEDs rated for 1 to 1.5v

Top Replies