In Australia, you used to be able to get those lights for $2 each. It was worth it just for the color cycling LED. I once took a hand cranked torched (aimed at kids) from IKEA, and changed out the white LEDS for RGB cycling LEDs. The kid loved being able to wind up this light, point it at the roof and watch the colors shine onto the roof as they went to sleep. You could tell when they weren't asleep just yet because you'd hear the grinding of the tocch dynamo from the other room.
Decathlon stores have cute little keyring crank up lamps. I don't know how easy it is to take them apart for a mod like this, but if possible, they can bring the joy anywhere.
It looked like a happy fart cushion to bring smiles and happiness to those that crossed it's path and all it brought was sadness. BLOODY SADNESS!!! I do suspect a pen could rectify the dilemma tho!
If the capacitor was across the LED, it (the LED) would clamp the voltage at about 3-4 Volt and you couldn't store much electricity in the capacitor. However, in this arrangement the voltage of the capacitor is raised above that of the NiMH cell; when the field collapses in the inductor the voltage should still be high enough to provide some energy to the LED.
I think the relative voltage the LED sees is still roughly its own forward voltage drop, no? During the flyback pulses since the cap is tied to ground the voltage it theoretically will see will be the NiMH cell's from the positive rail + the LED forward voltage drop during the flyback pulses. But its still referenced to the positive rail when providing some current to the LED, if Im following it correctly. Could be wrong, tho since im not an EE. Still interesting none the less.
I recently built my own circuit boards for the yx8018 and then made a variant that had the extra diode and cap as seen in the datasheet. This was all because of your channel, i started building my own boards to repair lights in the yard with slightly more modular units, and they end up very inexpensive
Those two capacitor positions are equivalent, a.c.-wise, because the battery is an a.c. short circuit. Also, in the position they have it, the capacitor is running at a higher mean voltage, so because energy in a capacitor is 1/2 C V-squared, in that position it stores more electrical energy than across the LED itself.
Clive, a request - for future videos of this nature. You're great at explaining in these videos how these 'weird' circuits work but I (& I suspect many others) would appreciate more 'why' explanation - particularly of 'circuitry nonsense' - in this case the use of an inductor at all. I'm guessing that most of 'us' 'get' resistors, and most of us 'get' capacitors and diodes in simple use but I'm guessing the majority of us are clueless about the whole point og having an inductor in just about any circuit :) Maybe there's scope for you to do an INDUCTION course on inductors and their uses - if I right, one of the common uses is as a 'choke' (needs explaining) in fluorescent lighting fittings - although I understand this particular function - well, at least I think I do :) This may lead to a mini series where you explain things like transistors and other more exotic electronic components than the simple ones ?
Thank you for this video! Thanks to his help, I regenerated my RGB solar lamps. The electronics were corroded so I used PCB from old white solar lights. I glued larger solar cells on top and added a bigger battery. With that, they shine brightly all night!
I had a few solar "rock" lights that used that same green battery which never lasted. I did end up modding one to use a regular white LED and a NiCD battery from a bigger solar light. I thought these used rapid pulsing because I hooked a solar cell to a mini amp and put a solar light up to the cell and it made weird noises.
Assuming a battery to be a Thevenin equivalent "short-circuit", it hardly matters if you wire the "other end" of the capacitor to the -ve terminal of the battery or the +ve. Doing it *this way* likely reduces the effect of the parasitic-inductance of the "lower horizontal wire" (in your CliveCAD drawing at 6:45) - or maybe it was just easier to route on a single-sided PCB. Thought-experiment: When you are making an RC lowpass filter - it doesn't matter if you wire 'the other end" of the capacitor to Gnd or to Vcc - they both work equally well!
Great explaination. We have 200 LEDs on one solar panel. I supercharged it and found their was space for another AA in the S panel box...so added another 600ma batt in parallel and added another S panel minus the circuitry again in parallel with the existing S Panel and the lights stay on all night through to dawn even when it's been cloudy all day with no sun. Not bragging but so happy with my mod!! Great video
Nice video - that probably explains one of my old solar garden lights. It was a dual mode garden light, either white or color changing, and when set to color changing, it would stay red all of the time! Maybe the chip inside had failed due to a manufacturing defect. All of my other lights were fine.
This configuration is typical of our garden lights we have here in Perth Australia. Only difference is the battery. We tend to have AA 1.2v 600mah batteries in our lights sold here.
Clive, I've been watching youtube videos all way, once I got to your channel again this evening, I had to turn up my vol quite a bit to hear you. Just thought I'd letcha know; maybe normalize those tracks to 99 pct. or something clever.
As always, loved the video. If you did your videos in person up on a stage in a concert venue style, you would probably have all us geeks throwing our panties at you and be fighting off groupies back stage like a rock star. I was watching one of your other videos this week, and having watched a lot of your videos I don't remember which one it was, but how do you know when designing a circuit, whether or not to use an inrush resistor? I think that is what you called it but with my memory definitely may have been something else. It was a resistor that was in the beginning of the circuit, between the input voltage and the rest of the circuit, and seems like its reason for being was to protect the circuit from too large of a voltage or current.
An inrush resistor is often used in capacitive dropper circuits where connecting the circuit to power at the peak of the sinewave would result in a very high pulse of current. The resistor limits the peak current, but is a low enough value not to dissipate too much heat with normal current flow.
Hi Big Clive. I was wondering if there might be a greater use for the Dollar Store sorts of solar garden lights? I've often wondered if it would be worthwhile to combine the little solar panels into a greater array to perhaps have an 18650 sort of lithium cell powering brighter or maybe more bulbs - maybe a string of them? Seems like it would be a cool project but then maybe it's cheaper to get the larger components? Curious.
To have a single light than is bright? Using Li-ion cell? wrt Li-ion the charging is more complex and needs a higher voltage (than a Ni-MH), so would need PVs in series and parallel; easier if they are all the same. The thing you would be reusing is the PV - cheaper to buy a large PV.
Does this circuit turn off the light if the battery voltage gets low? A white LED in the most basic circuit over-discharges the NiMH cells in the night if there is not enough day light to charge the battery. I also have a couple of solar lights with RGB LED in a circuit as in the video but with a 5252F chip, and these lights turn off on low battery voltage, which helps to prevent battery degradation.
@@bigclivedotcom That's not exactly what I have experienced. I have about two dozens lights with an YX8018 chip + inductor + monochrome LED, and in all of them it seems that there is no lower limit to the battery voltage. I can observe how the LEDs slowly fade out as the battery gets low on charge. I have also measured the battery voltage to fall way below 1V. In the colored lights with 5252F chip and additional capacitor + diode, the light switches completely off on a low battery. As opposed to the monochrome lights I didn't have to replace a degraded battery in one of those. I am wondering if the difference is due to the chip (YX8018 vs. 5252F), the circuit around the chip (inductor vs. inductor+capacitor+diode), or the chip in the RGB LED.
I wonder if the chip has something like a MOSFET instead of a diode to charge the battery from the solar panel, to reduce the voltage drop. It would be easy enough to integrate into the chip, and that 0.6V savings might allow them to use one fewer solar cell in series, so it might be worth it.
Hey Clive, have you seen anything about the new Tool album release? In the physical CD version there's going to be a 4" HD display, a 2 watt speaker, a rechargable battery. It's going to be used to display what seems to be either concert footage, or their usual trippy visuals. It seems like something you might be interested in checking out. I'm not sure what's going to be running it all, but I'm going to assume a linux distro on something like a raspberry pi 0. Even if they're not your sort of music, it might be worth ordering to check the electronics out.
Very interesting. Local town Mitre 10 has some for $7. I have one of them that have failed - the one I have looks like glass dome which looks like it was "cracked" (not sure why they chose that for design) and I am thinking to take it to pieces to see if the battery cell is faulty, the LED has died or something else has happened. It could have also had some condensation get in there. Was working for a month or two then it just died.
Could the run time be extended by adding a resistor to the LED, or would that only be effective if a second diode and capacitor were added to form a voltage multiplier. please?
Run time is a function of the energy stored in the battery and the power consumption of the other bits. Putting a resistor in series with the LED reduces the power consumed by the LED - which will reduce its brightness - but the resistor will be consuming some power. Better to change the LED current by changing the inductor value. Same LED current versus brightness trade-off but no energy loss in a resistor.
@@geoffmorrison3648 By far the best results I've had to date is to have replaced the disc battery with a 600mAh Ni-MH 1.2V AAA and reduce the coil to 33uH, Crank up the power to 10.7V by addition of another diode and capacitor and add LED protection resistor of 390oHm. Fires up the colour changing LEDs for hours in those ornamental birds that have LEd's on their breasts even on an overcast day.
Quick question for the L.E.D King Are there any leds that have the same colour temp as a high pressure sodium bulb? HPS bulbs glow makes me hypnotised and i miss them.
The low pressure yellow lamps could be emulated with yellow LEDs, but for the golden glow of high pressure sodium you might need to get RGB tape and a colour controller.
Hey BC. Perhaps some type of analysis of tamper-proof alarm wiring (‘ end-of-line ‘), with its use of resistors ,might make for an interesting future video? ... just an idea. Enjoying your videos, good stuff.
If you place a small neodymium magnet next to the inductor it boosts the LED brightness considerably. but sadly the battery current increases to 15mA compared to 3mA. Superb circuit though!
thanks Clive, always wanted to know what I needed for rgb, and after putting in a"1000mAh" battery should last a few more hours, last month of winter here so should get a full night around October,
That color changing led might have an internal reverse diode (shunt) ...don't know why this is, but it really sucks in the dark, if installed backwards.
If you connect a colour changing LED backwards with no current limiting it will fail. The internal chip has a protection diode that will shunt the reversed supply and often fail short circuit in the process.
@@johncoops6897 My thoughts exactly, and one should always add the next highest resistor in the table, to the resistor which has been calculated. Most essential if doing as I've done by adding an additional diode and capacitor so as to form a voltage multiplier output at the LED.
I tried using them with AC mains. I found that using a nanofarad cap like in the previous video on the subject caused only red green and blue to cycle without any of the nice mixed colours. I used a bigger cap and it started to flicker as well. Using a puny cap gave the best results but still not as good as a battery. I suppose the bigger the cap you use, the closer to the peak voltage you get which may be the problem in this case.
If the current is too low the lower forward voltage of the red LED means that it pulls the voltage too low for the blue and green to light at the same time.
Hi Clive, here's a puzzler for you. Managed to 'acquire' quite a few Motorola XTN446 PMRs complete with the drop-in chargers, that had been disposed of. Now most of the NiCd batteries are fine (4.8v nominal) but a few show between 5.1v and 5.3V when measured with the multimeter. How can this be?? To my knowledge NiCd have a max output of 1.2v, so the pack which has 4 in series should be no more than the 4.8v stated. Has got me really confused. Thanks for any help, and thanks for your great videos.
Personally I'd put a 5.1v zener diode across the led too, just to protect it from over voltage. If the inductor value is low and the circuit is able to supply more current than the colour changing led needs, the voltage across the led will drift up until something in the led discharges it.
That would definitly be the case with a standard led, but the colour changing ones work a little differently. They have a driver chip in them and the current through the leds is set internally with a reference, rather than externally by a resistor or in this case by the amount of energy metered through the inductor on each switch pulse. The colour changing led will only take what it needs from the supply, probably about 5mA. If the power supply is pumping more than that (on average) into the capacitor, the voltage will drift up. Since the current is fixed (constant current driver built into rgb led), the voltage on the capacitor won't be regulated to the Vf of the led like it would be with a standard led. The driver chip in the led will have a maximum supply voltage it can tolerate, probably about 5.5V (or maybe 6V). Once the supply exceeds this, the driver will draw excessive current and discharge the cap back down, but repeatedly doing this will shorten the life of the led. The zener diode is a simple and safe way of regulating the supply voltage to the led and burning off any excess energy supplied by the inductor. Ideally there would be voltage feedback to the switching controller and it would change up the duty cycle of the PWM to maintain constant voltage, but this chip doesn't have that capability.
@@alexwood020589 ISWYM but the current still increases quite rapidly as voltage goes up, so they still provide some degree of self regulation. A zener is a pretty poor way to protect anything like this since a 5.1v zener will start conducting well below its rated voltage, shunting led current and further reducing efficiency/battery life. Additionally these leds almost never fail in these units, inevitably either the battery fails or the housing leaks water and the PCB corrodes to death.
so this isn't a BOOST converter right, since it never puts battery and inductor in series? Why not? So does having a smaller inductor increase the voltage delivered? What is the advantage of a boost converter compared to this?
It *is* a boost converter, just not in the way you'd think. It creates a higher voltage across the inductor as opposed to referenced to ground. When you pass a current through an inductor and suddenly stop the flow of current, the inductor will try to maintain the same level, which is causing an inductive kickback, raising the voltage. This effect is only limited by something clamping the voltage. This also why magnetic relays will often have a reverse diode across the coil. Otherwise the high voltage spike can cause arcs when the relay is disengaged. Why a lower inductance gives a higher current? When the on period starts, the current ramps up as the inductor builds up its magnetic field, since inductors resist changes in current. With a higher inductance, the current ramps up slowly, so when the supply is turned off, there's less current going through the inductor at that time, and less magnetic energy available for the inductive kickback effect.
@@Gameboygenius I get it is a *boost* converter, but the term BOOST CONVERTER is usually a device that switches an inductor in series with a power source at high frequency creating a higher voltage. The above system does not follow that topology. I wonder what the pros and cons are between this and a typical boost converter.
@@jovangrbic97 It is a boost converter, the switching pin is open-drain, the inductor is switched to ground at 200kHz: ez.analog.com/university-program/b/blogs/posts/hacking-an-led-solar-garden-light. You can redraw the circuit so that it looks just like the circuit in en.wikipedia.org/wiki/Boost_converter
I agree, this is much less efficient than a conventional boost topology since it has to generate the full led Vf across the inductor rather than Vf-Vbattery. I suspect the problem would be leakage through the led when the switcher is off, even though the battery voltage isn't high enough to light the led some small current would flow.
The capacitor is equally valid where they have it and where you say it should make sense. So long as one end is connected to the signal to be decoupled and the other to a low impedance DC rail then it fulfills its job,
In their arrangement, the capacitor is essentially siphoning current from the flyback pulses which then keep the LED running when the inductor is being switched by pin Lx. This is more effective because the capacitor will tend to help average out the pulses.
So when you say frequencies and you’re talking about the NPN does that mean how many times it is switched on and off per second? Say 2k Hz is 2,000 times per second?
I was trying to change out my 120v AC cold white LED Christmas lights, with slow flashing RGB's but they just stay red. Do I just need a capacitor for smoothing or something more? The lights seem to be broken into two sets in the string and plastic covered resistor in the line that gets slightly warm when left on.
I've converted strings to RGB in the past and needed to add a capacitor across the string (even a low value suppression capacitor may work), or made a new power supply.
I tried this mod because I wanted to fit a flickering candle effect LED. Strangely, it only flickers if I DO fit the diode (Schottky) but NOT the capacitor! Any other combination and it just stays permenantly on.
hey, partner, i bought some cheap $1 solar lights from the dollar store and I try to put the multi-color changing bulb in and it only stayed one color? what parts do I need to make them change? thanks.
It needs a diode and capacitor. The output is pulsed and that keeps resetting the color changing chip. The capacitor goes across the LED and the diode interrupts the positive track to the LED and it's capacitor, with the band pointing to the LED.
@@bigclivedotcom I have some crystal oscillators of various values would those work? Also thanks for replying I love watching your videos, especially about LEDs, my wife loves to light things up all the time.
Maybe they put the cap there cuz the negative rail was easier to get to when the were routing the board. But in that case why didn't they just put the cap right next to the led?
So the inductor is being used as a swtich to disconnect the LED side of the circuit by polarity? Clever. And somehow, at just 6 components, still over-built. ;)
The inductor is being used as an energy store. This allows the voltage to be boosted at the expense of a lower current to the LED. The “switch” is in the chip.
@@Mark1024MAK Hm. Guess I still don't understand the circuit. I'd like to see a sine wave or something. So it ostensibly slows the inrush of current by storing a portion of it as a magneticive field, then when the power drops off, the field begins to collapse, starts moving electrons, and by the time we're starting the forward push of the DC again, we're adding the stored voltage of the field? You know... my problem is that there just aren't good teachers. People either speak in simplest terms to the uninformed, or assume knowledge and fire over the heads of the half-informed and act like you should already /get this./ Where are the folks speaking to the just-knowledgeable-to-be-dangerous group I find myself in?
roninpawn - In simple terms... Current flows to the chip (YX8018). It switches on its internal transistor ‘switch’ (connected between pins LX and GND). This then allows a DC current to flow from the battery to the inductor through the chips switching transistor and back to the battery via the GND pin. This current builds up a magnet field in the inductor. After a time, the chip switches off its internal switching transistor. Inductors don’t like change, so in an attempt to keep the status-quo, the collapsing magnetic field induces voltage into the copper wire (coils) of the inductor to try to maintain a current flow. As the inductor is now a voltage source, the polarity swaps over (compared to when it was receiving voltage). The resulting current can only flow via the LED. And the LED will not conduct until the voltage across it gets to its turn on voltage. Until a current starts to flow, the collapsing magnetic field will cause the voltage across the inductor terminals to rise rapidly, going far above the original supply voltage. Once the LED conducts, the energy that was stored in the magnetic field is converted to current that powers the LED. After a period of time, the energy is exhausted, so the chip turns its internal transistor switch back on and the cycle repeats. Because it operates at a relatively fast rate, the flicker from the LED is not visible to the human eye.
I would love to understand these circuits but the change in electrical potential between components is quite awkward to understand as they are not simple constants. They change in relation to each other and with time. Does anyone know of software that could be used to replicate such a circuit.
Love your videos of questionable devices. I might suggest the baby stroller fans that use 18650 cells and recharge with a usb cable. They are wildly popular and I can only wonder if min discharge and max charge voltages are obeyed. I use one while mowing the lawn with a cordless mower and love it, but I would hate to imagine how many are charging in nurseries!
I've found they only work with a capacitor too, but it also depends how the led is connected to battery, could then be switched to an on state with a set 'darkness' level
I've got one I built with 2 supercaps. It works even during Feburary when it's like 0 degrees out. Obviously less bright than a NiMH powered one though, since there's less capacity.
Clive will you make impromptu teardown vids on your toolbox inside the workarea at the tattoo ? possibly while the other guys/girls there are being a pest :-)
Easy fix to prevent that, remove the switch and bridge the connections so the light is permanently switched on, so no chance for any rust problems shorting things out... :)
@@xplosivpoison yocto-farts would have to be strange quarks or something. I think an amoeba would only be an atto-fart if they scale linearly. Viruss would be like 300 zepto-farts
seems like the capacitor is used to store a charge with a higher voltage than the battery voltage. i also find it strange that they didn't put the capacitor on the LED leads aswell.. I guess the possible advantage is that the LED and capacitor combination are set between two (somewhat) known and constant voltages : the battery positive voltage and the ground. Maybe this avoids further noise production?... (which would be surprising, if we take into account the usual attention to design details of those kind of cheap gadgets)
wait... could is be that the capacitor dumps some charge back to the NiMh battery (through the LED) once the inductor did its job?..... if so, it's a clever way to salvage some of the spent energy which would otherwise be "wasted" (even though it would probably be a tiny ammount in this case)
@@schluderjupp Yes, i agree. That is indeed the job of the capacitor, and what clive has said about the LED requiring power to keep information stored to know which is the next color to shine. But i was wondering about the placement of that capacitor. It could have worked even if the capacitor was in parallel with the LED, but it wasn't the case. when the capacitor is charged and supplies energy to the LED, it seems that the battery is in series, and that the capacitor should store a voltage which is greater than that of the battery. So it would seem that the design hits 2 birds with one stone : it powers the LED, AND the current that goes through the LED returns to the battery. It won't be perfect and run forever since there will be some power loss through that process, but the small amount which is being sent back to the battery successfully may lengthen the operation time (although i'm not sure by how much).
Damonack The internal circuitry of the chip keeps current from flowing back to the battery. Current only flows to the led when the chip disconnects LX from ground. When LX is connected to ground, it charges the Inductor from the battery . So the Inductor never charges the battery.
@@schluderjupp ... Yes charging the inductor happens when LX is pulled to ground by the chip. When LX is released, the current from the inductor passes through the diode to the capacitor. The only way the LED can be powered by the capacitor is if the capacitor's voltage is greater than the battery. Follow the diagram : the LED's Anode goes to the positive side of the capaciror, the LED's cathode is connected to the battery's positive side, and both the battery and the capacitor's negative side are tied to ground. The circuit is made, and the only way for the LED to be powered is if the capacitor's voltage is greater than the battery's voltage. Then the current does flow into the battery's positive side when the LED is powered and the battery is not charging the inductor.
I wonder if this would work just by scaling up the panel and the battery by a factor of ten or a hundred. Would be able to run something far more useful than an LED practically for free then.
That's rather intriguing how you increase the current through the LED by *lowering* the inductor value. Would have thought bigger inductors would store more energy, therefore more current? Perhaps it's because the larger value is actually de-tuning the circuit in some way?
This is actually very simple. Larger inductor will have lower current, if energized in same time with same amount of voltage. This is probably Ton=constant chip and Toff=needed for CrCCM. And that explains everything.
Not strange at all, if you look at the math in en.wikipedia.org/wiki/Boost_converter#Continuous_mode, in particular, the equation for Ioff. @560uH, Ioff=.5×(200kHz)^−1/560uH × 2.5V=11mA, @47uH, Ioff=.5×(200kHz)^−1/47uH × 2.5V=132mA. I haven't accounted for a bunch of other stuff here, so my numbers don't match the datasheet, but the math works out close enough.
Quite simply a larger inductor resists current change more than a smaller one, so for a given pulse length on the output of the IC a smaller inductor will reach a higher peak current. A second order effect is that lower value inductors of the same size tend to have lower series resistance (fewer turns of wire, often thicker wire).
It allows the use of a single 1.2V cell to drive a 3V LED. That saves cost and also prevents cell damage in stacks of cells when the first to go flat gets reverse charged.
Those little chips are great! I use them to power tiny high voltage transformers directly. They get a little warm doing it, but who really cares about that.
I've been tinkering with a similar solar light with a similar YX8058 chip and the same button cell (1.2v 40mAh). While waiting on some higher capacity batteries to replace the button cell, I managed to blow up my first button cell by charging with too high current, then too high voltage. My reaction was... OH COOL! Perhaps I should secure my explosion containment pie dish sooner than later. 😆
Also handy for Joule Thief circuits, which since getting a bag of 100 (actually 101, cos china can't count!) BC547s, I've made a few of, one of which is annoying as it sounds like a CRT flyback, but that was my first attempt, I'm still learning though... :D
datasheet www.solidfiles.com/v/NaLamYWagdBWZ there is couple of those datasheets, I'm not sure, maybe is there more then one company making this chip also from what i know there is no official English datasheet for this
I built my first computer with RGB lighting today, Unfortunately Red=Green and Green=Red, blue is OK. The wiring is in a molded connector so no swapping the wire/pins around. Oh well :)
@@bigclivedotcom it have an atex headtorch and led lenser head torch that both dont appear to work as well as they used to I have multiple old lights etc that I could bring along to the tattoo and let you see what may be of interest to you there just lying around collecting dust now a days
Although the 1N4008 might actually do the job, all these are mains rectifiers and pretty terrible for high frequency switching applications. Use a cheap Schottky!
Most likely it would. Again, a flashing LED uses a small integrated circuit to cycle power on and off. If the power to that integrated circuit is being cycled, the chip won’t flash. Try it. Let us know.
And when tested using a multimeter switched to the resistance range, inductors / chokes have a much lower resistance that does not match the colour code.
Wow... I got a solar lamp that had a 80mAh button cell and it only lasts 4hrs using white led. Eventually I had to mod in another cell in parallel. Surprised they only supplied 40mAh in a color changing unit.
tensazero By changing the Inductor you can raise or lower the current to the led and make it run longer/dimmer (or brighter/shorter). A larger battery often doesn't help because the solar panel can't deliver enough energy to totally charge the battery. Usually a larger Inductor value will lower the current to the led. Just try it out. 😀
Susan Amber Bruce As big as you like, but you have to remain within the switching current limit, so a big battery may take as impractically long time to charge.
In Australia, you used to be able to get those lights for $2 each. It was worth it just for the color cycling LED. I once took a hand cranked torched (aimed at kids) from IKEA, and changed out the white LEDS for RGB cycling LEDs. The kid loved being able to wind up this light, point it at the roof and watch the colors shine onto the roof as they went to sleep. You could tell when they weren't asleep just yet because you'd hear the grinding of the tocch dynamo from the other room.
That's actually a kool idea, I like that.
Decathlon stores have cute little keyring crank up lamps. I don't know how easy it is to take them apart for a mod like this, but if possible, they can bring the joy anywhere.
Really nice idea
Don't aim torches at kids, hand-cranked or otherwise.
@@Anvilshock
?
I was disappointed uptil just after @2:04 that cap looked likeit had a smiley face, then clive revealed the true let down
The famous electronic happy pill. Proven to smooth out the jitters.
It looked like a happy fart cushion to bring smiles and happiness to those that crossed it's path and all it brought was sadness. BLOODY SADNESS!!!
I do suspect a pen could rectify the dilemma tho!
If the capacitor was across the LED, it (the LED) would clamp the voltage at about 3-4 Volt and you couldn't store much electricity in the capacitor. However, in this arrangement the voltage of the capacitor is raised above that of the NiMH cell; when the field collapses in the inductor the voltage should still be high enough to provide some energy to the LED.
I think the relative voltage the LED sees is still roughly its own forward voltage drop, no? During the flyback pulses since the cap is tied to ground the voltage it theoretically will see will be the NiMH cell's from the positive rail + the LED forward voltage drop during the flyback pulses. But its still referenced to the positive rail when providing some current to the LED, if Im following it correctly. Could be wrong, tho since im not an EE. Still interesting none the less.
It can be worse, actually. th-cam.com/video/2MQyQUkwmMk/w-d-xo.html
Thank you for the tip Clive - I put an RGB LED in a solar light once and it was stuck on red, couldn't figure out why.
#metoo
@@mahlapropyzm9180 You haven't earned the right to use that hashtag :P
I recently built my own circuit boards for the yx8018 and then made a variant that had the extra diode and cap as seen in the datasheet. This was all because of your channel, i started building my own boards to repair lights in the yard with slightly more modular units, and they end up very inexpensive
Why is the capacitor looking so happy? 2:00 ;-)
Because it could get to the other side. 😁
Those two capacitor positions are equivalent, a.c.-wise, because the battery is an a.c. short circuit.
Also, in the position they have it, the capacitor is running at a higher mean voltage, so because energy in a capacitor is 1/2 C V-squared, in that position it stores more electrical energy than across the LED itself.
If they are equivalent, why does the circtuit behave differently then? Ya walnut.
Clive, a request - for future videos of this nature. You're great at explaining in these videos how these 'weird' circuits work but I (& I suspect many others) would appreciate more 'why' explanation - particularly of 'circuitry nonsense' - in this case the use of an inductor at all. I'm guessing that most of 'us' 'get' resistors, and most of us 'get' capacitors and diodes in simple use but I'm guessing the majority of us are clueless about the whole point og having an inductor in just about any circuit :)
Maybe there's scope for you to do an INDUCTION course on inductors and their uses - if I right, one of the common uses is as a 'choke' (needs explaining) in fluorescent lighting fittings - although I understand this particular function - well, at least I think I do :) This may lead to a mini series where you explain things like transistors and other more exotic electronic components than the simple ones ?
Thank you for this video! Thanks to his help, I regenerated my RGB solar lamps. The electronics were corroded so I used PCB from old white solar lights. I glued larger solar cells on top and added a bigger battery. With that, they shine brightly all night!
I had a few solar "rock" lights that used that same green battery which never lasted. I did end up modding one to use a regular white LED and a NiCD battery from a bigger solar light.
I thought these used rapid pulsing because I hooked a solar cell to a mini amp and put a solar light up to the cell and it made weird noises.
Assuming a battery to be a Thevenin equivalent "short-circuit", it hardly matters if you wire the "other end" of the capacitor to the -ve terminal of the battery or the +ve. Doing it *this way* likely reduces the effect of the parasitic-inductance of the "lower horizontal wire" (in your CliveCAD drawing at 6:45) - or maybe it was just easier to route on a single-sided PCB.
Thought-experiment: When you are making an RC lowpass filter - it doesn't matter if you wire 'the other end" of the capacitor to Gnd or to Vcc - they both work equally well!
Just done that little mod and works a treat, thanks Clive.
Great explaination. We have 200 LEDs on one solar panel. I supercharged it and found their was space for another AA in the S panel box...so added another 600ma batt in parallel and added another S panel minus the circuitry again in parallel with the existing S Panel and the lights stay on all night through to dawn even when it's been cloudy all day with no sun. Not bragging but so happy with my mod!! Great video
Nice video - that probably explains one of my old solar garden lights. It was a dual mode garden light, either white or color changing, and when set to color changing, it would stay red all of the time! Maybe the chip inside had failed due to a manufacturing defect. All of my other lights were fine.
This configuration is typical of our garden lights we have here in Perth Australia.
Only difference is the battery.
We tend to have AA 1.2v 600mah batteries in our lights sold here.
Ty for explaining that never could figure that one out
Thank you! I always thought it was low voltage stopping the rgb change.
That happens on the 'specially designed' ones. It gets stuck on red as the battery drains, then starts oscillating, then shuts down.
Love the Mega Zoom, Clive. Well done.
Honestly I'd have thought that's just a resistor there. LOL. World's smallest inductor / autotransformer / flyback. Very neat little circuit!
Clive, I've been watching youtube videos all way, once I got to your channel again this evening, I had to turn up my vol quite a bit to hear you. Just thought I'd letcha know; maybe normalize those tracks to 99 pct. or something clever.
Most channels are super quiet this week. It makes me suspect it's a TH-cam thing.
Great video Big Clive
Thanks Clive. Now I know what those 4 pin chips are, I've had some kicking around but couldn't find the information for them. 👍
As always, loved the video. If you did your videos in person up on a stage in a concert venue style, you would probably have all us geeks throwing our panties at you and be fighting off groupies back stage like a rock star.
I was watching one of your other videos this week, and having watched a lot of your videos I don't remember which one it was, but how do you know when designing a circuit, whether or not to use an inrush resistor? I think that is what you called it but with my memory definitely may have been something else. It was a resistor that was in the beginning of the circuit, between the input voltage and the rest of the circuit, and seems like its reason for being was to protect the circuit from too large of a voltage or current.
An inrush resistor is often used in capacitive dropper circuits where connecting the circuit to power at the peak of the sinewave would result in a very high pulse of current. The resistor limits the peak current, but is a low enough value not to dissipate too much heat with normal current flow.
Thanks for the Information.Are you using a Noise Gate on your TH-cam sound?
Hi Big Clive. I was wondering if there might be a greater use for the Dollar Store sorts of solar garden lights? I've often wondered if it would be worthwhile to combine the little solar panels into a greater array to perhaps have an 18650 sort of lithium cell powering brighter or maybe more bulbs - maybe a string of them?
Seems like it would be a cool project but then maybe it's cheaper to get the larger components? Curious.
To have a single light than is bright? Using Li-ion cell?
wrt Li-ion the charging is more complex and needs a higher voltage (than a Ni-MH), so would need PVs in series and parallel; easier if they are all the same. The thing you would be reusing is the PV - cheaper to buy a large PV.
Does this circuit turn off the light if the battery voltage gets low? A white LED in the most basic circuit over-discharges the NiMH cells in the night if there is not enough day light to charge the battery. I also have a couple of solar lights with RGB LED in a circuit as in the video but with a 5252F chip, and these lights turn off on low battery voltage, which helps to prevent battery degradation.
Many of these chips do seem to turn off the output if the battery voltage drops to around 1V or so.
@@bigclivedotcom That's not exactly what I have experienced. I have about two dozens lights with an YX8018 chip + inductor + monochrome LED, and in all of them it seems that there is no lower limit to the battery voltage. I can observe how the LEDs slowly fade out as the battery gets low on charge. I have also measured the battery voltage to fall way below 1V.
In the colored lights with 5252F chip and additional capacitor + diode, the light switches completely off on a low battery. As opposed to the monochrome lights I didn't have to replace a degraded battery in one of those. I am wondering if the difference is due to the chip (YX8018 vs. 5252F), the circuit around the chip (inductor vs. inductor+capacitor+diode), or the chip in the RGB LED.
I wonder if the chip has something like a MOSFET instead of a diode to charge the battery from the solar panel, to reduce the voltage drop. It would be easy enough to integrate into the chip, and that 0.6V savings might allow them to use one fewer solar cell in series, so it might be worth it.
it's not worth it, that chip is produced as cheep as possible
Hey Clive, have you seen anything about the new Tool album release? In the physical CD version there's going to be a 4" HD display, a 2 watt speaker, a rechargable battery. It's going to be used to display what seems to be either concert footage, or their usual trippy visuals. It seems like something you might be interested in checking out. I'm not sure what's going to be running it all, but I'm going to assume a linux distro on something like a raspberry pi 0. Even if they're not your sort of music, it might be worth ordering to check the electronics out.
Very interesting. Local town Mitre 10 has some for $7. I have one of them that have failed - the one I have looks like glass dome which looks like it was "cracked" (not sure why they chose that for design) and I am thinking to take it to pieces to see if the battery cell is faulty, the LED has died or something else has happened. It could have also had some condensation get in there. Was working for a month or two then it just died.
Could the run time be extended by adding a resistor to the LED, or would that only be effective if a second diode and capacitor were added to form a voltage multiplier. please?
It does seem to extend run time slightly if two LEDs are wired in series, but the best way to increase run time is by changing the inductor.
Run time is a function of the energy stored in the battery and the power consumption of the other bits. Putting a resistor in series with the LED reduces the power consumed by the LED - which will reduce its brightness - but the resistor will be consuming some power. Better to change the LED current by changing the inductor value. Same LED current versus brightness trade-off but no energy loss in a resistor.
@@geoffmorrison3648 By far the best results I've had to date is to have replaced the disc battery with a 600mAh Ni-MH 1.2V AAA and reduce the coil to 33uH, Crank up the power to 10.7V by addition of another diode and capacitor and add LED protection resistor of 390oHm. Fires up the colour changing LEDs for hours in those ornamental birds that have LEd's on their breasts even on an overcast day.
Fantastic zoom on this camera!
Quick question for the L.E.D King Are there any leds that have the same colour temp as a high pressure sodium bulb? HPS bulbs glow makes me hypnotised and i miss them.
The low pressure yellow lamps could be emulated with yellow LEDs, but for the golden glow of high pressure sodium you might need to get RGB tape and a colour controller.
Hey BC. Perhaps some type of analysis of tamper-proof alarm wiring (‘ end-of-line ‘), with its use of resistors ,might make for an interesting future video? ... just an idea.
Enjoying your videos, good stuff.
I used a 20 uF Capacitor in a breadboard circuit like this and got 10 Volts across it, (no load) but not much current.
If you place a small neodymium magnet next to the inductor it boosts the LED brightness considerably. but sadly the battery current increases to 15mA compared to 3mA. Superb circuit though!
thanks Clive, always wanted to know what I needed for rgb, and after putting in a"1000mAh" battery should last a few more hours, last month of winter here so should get a full night around October,
High capacity batteries usually don't do much as charge is limited by the size of the solar panel, i.e
even 600mA ones often don't get fully charged.
That color changing led might have an internal reverse diode (shunt)
...don't know why this is, but it really sucks in the dark, if installed backwards.
If you connect a colour changing LED backwards with no current limiting it will fail. The internal chip has a protection diode that will shunt the reversed supply and often fail short circuit in the process.
@@johncoops6897 My thoughts exactly, and one should always add the next highest resistor in the table, to the resistor which has been calculated. Most essential if doing as I've done by adding an additional diode and capacitor so as to form a voltage multiplier output at the LED.
I tried using them with AC mains. I found that using a nanofarad cap like in the previous video on the subject caused only red green and blue to cycle without any of the nice mixed colours. I used a bigger cap and it started to flicker as well. Using a puny cap gave the best results but still not as good as a battery. I suppose the bigger the cap you use, the closer to the peak voltage you get which may be the problem in this case.
If the current is too low the lower forward voltage of the red LED means that it pulls the voltage too low for the blue and green to light at the same time.
Beautiful soldering!
what volt ? with short ,with that coil connect parallel with led RGB diode ? ?
Hi Clive, here's a puzzler for you. Managed to 'acquire' quite a few Motorola XTN446 PMRs complete with the drop-in chargers, that had been disposed of. Now most of the NiCd batteries are fine (4.8v nominal) but a few show between 5.1v and 5.3V when measured with the multimeter. How can this be?? To my knowledge NiCd have a max output of 1.2v, so the pack which has 4 in series should be no more than the 4.8v stated. Has got me really confused. Thanks for any help, and thanks for your great videos.
The fully charged voltage of a NiMh cell can float up to about 1.5V, but drops back to 1.2V quickly under load.
@@bigclivedotcom Thanks Clive!!!
Personally I'd put a 5.1v zener diode across the led too, just to protect it from over voltage. If the inductor value is low and the circuit is able to supply more current than the colour changing led needs, the voltage across the led will drift up until something in the led discharges it.
It's really not needed, these things can only deliver a small current so the led does a fine job of clamping the voltage to it's own Vf.
That would definitly be the case with a standard led, but the colour changing ones work a little differently. They have a driver chip in them and the current through the leds is set internally with a reference, rather than externally by a resistor or in this case by the amount of energy metered through the inductor on each switch pulse.
The colour changing led will only take what it needs from the supply, probably about 5mA. If the power supply is pumping more than that (on average) into the capacitor, the voltage will drift up. Since the current is fixed (constant current driver built into rgb led), the voltage on the capacitor won't be regulated to the Vf of the led like it would be with a standard led.
The driver chip in the led will have a maximum supply voltage it can tolerate, probably about 5.5V (or maybe 6V). Once the supply exceeds this, the driver will draw excessive current and discharge the cap back down, but repeatedly doing this will shorten the life of the led. The zener diode is a simple and safe way of regulating the supply voltage to the led and burning off any excess energy supplied by the inductor.
Ideally there would be voltage feedback to the switching controller and it would change up the duty cycle of the PWM to maintain constant voltage, but this chip doesn't have that capability.
@@alexwood020589 ISWYM but the current still increases quite rapidly as voltage goes up, so they still provide some degree of self regulation. A zener is a pretty poor way to protect anything like this since a 5.1v zener will start conducting well below its rated voltage, shunting led current and further reducing efficiency/battery life. Additionally these leds almost never fail in these units, inevitably either the battery fails or the housing leaks water and the PCB corrodes to death.
so this isn't a BOOST converter right, since it never puts battery and inductor in series? Why not? So does having a smaller inductor increase the voltage delivered? What is the advantage of a boost converter compared to this?
It *is* a boost converter, just not in the way you'd think. It creates a higher voltage across the inductor as opposed to referenced to ground. When you pass a current through an inductor and suddenly stop the flow of current, the inductor will try to maintain the same level, which is causing an inductive kickback, raising the voltage. This effect is only limited by something clamping the voltage. This also why magnetic relays will often have a reverse diode across the coil. Otherwise the high voltage spike can cause arcs when the relay is disengaged.
Why a lower inductance gives a higher current? When the on period starts, the current ramps up as the inductor builds up its magnetic field, since inductors resist changes in current. With a higher inductance, the current ramps up slowly, so when the supply is turned off, there's less current going through the inductor at that time, and less magnetic energy available for the inductive kickback effect.
@@Gameboygenius I get it is a *boost* converter, but the term BOOST CONVERTER is usually a device that switches an inductor in series with a power source at high frequency creating a higher voltage. The above system does not follow that topology. I wonder what the pros and cons are between this and a typical boost converter.
@@jovangrbic97 It is a boost converter, the switching pin is open-drain, the inductor is switched to ground at 200kHz: ez.analog.com/university-program/b/blogs/posts/hacking-an-led-solar-garden-light. You can redraw the circuit so that it looks just like the circuit in en.wikipedia.org/wiki/Boost_converter
I agree, this is much less efficient than a conventional boost topology since it has to generate the full led Vf across the inductor rather than Vf-Vbattery. I suspect the problem would be leakage through the led when the switcher is off, even though the battery voltage isn't high enough to light the led some small current would flow.
@@ferrumignis but when off the PV is making power so wouldn't be an issue?
Very good informative video
The capacitor is equally valid where they have it and where you say it should make sense. So long as one end is connected to the signal to be decoupled and the other to a low impedance DC rail then it fulfills its job,
Where would I find these little color changing bulbs? I have solar wind chimes that have a couple lights out.
eBay 5mm slow RGB LED
In their arrangement, the capacitor is essentially siphoning current from the flyback pulses which then keep the LED running when the inductor is being switched by pin Lx. This is more effective because the capacitor will tend to help average out the pulses.
How do you know when it is an inductor? To me it looked like a normal resister.
Cool, thanks for the breakdown and explanation.
So when you say frequencies and you’re talking about the NPN does that mean how many times it is switched on and off per second? Say 2k Hz is 2,000 times per second?
Yes. These chips often operate at 100's of thousands of Hertz.
bigclivedotcom 😮😮😮😮
Great explanation!
I was trying to change out my 120v AC cold white LED Christmas lights, with slow flashing RGB's but they just stay red. Do I just need a capacitor for smoothing or something more? The lights seem to be broken into two sets in the string and plastic covered resistor in the line that gets slightly warm when left on.
I've converted strings to RGB in the past and needed to add a capacitor across the string (even a low value suppression capacitor may work), or made a new power supply.
I tried this mod because I wanted to fit a flickering candle effect LED. Strangely, it only flickers if I DO fit the diode (Schottky) but NOT the capacitor! Any other combination and it just stays permenantly on.
In my experience I have found that just the 1n4148 in series with the anode is all you need, i've never bothered with the cap.
Thanks for the information.
hey, partner, i bought some cheap $1 solar lights from the dollar store and I try to put the multi-color changing bulb in and it only stayed one color? what parts do I need to make them change? thanks.
It needs a diode and capacitor. The output is pulsed and that keeps resetting the color changing chip. The capacitor goes across the LED and the diode interrupts the positive track to the LED and it's capacitor, with the band pointing to the LED.
@@bigclivedotcom awesome 👌 thank you sir! Can you tell me the size of the capacitor and diode thank you love all your videos.
@@jimmysegovia2688 1N4148 diode and a 100nF capacitor.
@@bigclivedotcom awesome partner thank you so much for sharing.
Just a question, I have two color changing LEDs in a project but they become out of sync shortly after powering them on. How can I keep them in sync?
Each has its own oscillator. You'd need to blink then out briefly as they reached peak red to restart then.
@@bigclivedotcom I have some crystal oscillators of various values would those work? Also thanks for replying I love watching your videos, especially about LEDs, my wife loves to light things up all the time.
Maybe they put the cap there cuz the negative rail was easier to get to when the were routing the board. But in that case why didn't they just put the cap right next to the led?
did anyone else see a smiling face in the rust colored round part?
For a while, I enjoyed the ceramic capacitor looking like it had a smiley clown face from the side
Most illuminating, thank you ;-)
So the inductor is being used as a swtich to disconnect the LED side of the circuit by polarity?
Clever. And somehow, at just 6 components, still over-built. ;)
The inductor is being used as an energy store. This allows the voltage to be boosted at the expense of a lower current to the LED. The “switch” is in the chip.
@@Mark1024MAK Hm. Guess I still don't understand the circuit. I'd like to see a sine wave or something. So it ostensibly slows the inrush of current by storing a portion of it as a magneticive field, then when the power drops off, the field begins to collapse, starts moving electrons, and by the time we're starting the forward push of the DC again, we're adding the stored voltage of the field? You know... my problem is that there just aren't good teachers. People either speak in simplest terms to the uninformed, or assume knowledge and fire over the heads of the half-informed and act like you should already /get this./ Where are the folks speaking to the just-knowledgeable-to-be-dangerous group I find myself in?
roninpawn - In simple terms...
Current flows to the chip (YX8018). It switches on its internal transistor ‘switch’ (connected between pins LX and GND).
This then allows a DC current to flow from the battery to the inductor through the chips switching transistor and back to the battery via the GND pin. This current builds up a magnet field in the inductor.
After a time, the chip switches off its internal switching transistor.
Inductors don’t like change, so in an attempt to keep the status-quo, the collapsing magnetic field induces voltage into the copper wire (coils) of the inductor to try to maintain a current flow. As the inductor is now a voltage source, the polarity swaps over (compared to when it was receiving voltage). The resulting current can only flow via the LED. And the LED will not conduct until the voltage across it gets to its turn on voltage. Until a current starts to flow, the collapsing magnetic field will cause the voltage across the inductor terminals to rise rapidly, going far above the original supply voltage.
Once the LED conducts, the energy that was stored in the magnetic field is converted to current that powers the LED. After a period of time, the energy is exhausted, so the chip turns its internal transistor switch back on and the cycle repeats. Because it operates at a relatively fast rate, the flicker from the LED is not visible to the human eye.
I would love to understand these circuits but the change in electrical potential between components is quite awkward to understand
as they are not simple constants. They change in relation to each other and with time.
Does anyone know of software that could be used to replicate such a circuit.
SPICE
Love your videos of questionable devices. I might suggest the baby stroller fans that use 18650 cells and recharge with a usb cable. They are wildly popular and I can only wonder if min discharge and max charge voltages are obeyed. I use one while mowing the lawn with a cordless mower and love it, but I would hate to imagine how many are charging in nurseries!
I've found they only work with a capacitor too, but it also depends how the led is connected to battery, could then be switched to an on state with a set 'darkness' level
I've got one I built with 2 supercaps. It works even during Feburary when it's like 0 degrees out. Obviously less bright than a NiMH powered one though, since there's less capacity.
Clive will you make impromptu teardown vids on your toolbox inside the workarea at the tattoo ? possibly while the other guys/girls there are being a pest :-)
Too bad that the switch always seems to short out after a while in these kind of solar lamps, often taking the chip with it.
Easy fix to prevent that, remove the switch and bridge the connections so the light is permanently switched on, so no chance for any rust problems shorting things out... :)
Thank you
Nano Fart. The smallest of farts.
Femto Farts -- I think only a baby mosquito makes those -- are 1,000,000 times smaller. :3
@@xplosivpoison yocto-farts would have to be strange quarks or something. I think an amoeba would only be an atto-fart if they scale linearly. Viruss would be like 300 zepto-farts
@Mai Mariarti Small, silent, deadly
Interesting as always...
seems like the capacitor is used to store a charge with a higher voltage than the battery voltage. i also find it strange that they didn't put the capacitor on the LED leads aswell.. I guess the possible advantage is that the LED and capacitor combination are set between two (somewhat) known and constant voltages : the battery positive voltage and the ground. Maybe this avoids further noise production?... (which would be surprising, if we take into account the usual attention to design details of those kind of cheap gadgets)
wait... could is be that the capacitor dumps some charge back to the NiMh battery (through the LED) once the inductor did its job?..... if so, it's a clever way to salvage some of the spent energy which would otherwise be "wasted" (even though it would probably be a tiny ammount in this case)
Damonack nope and nope....the capacitor is just an energy reservoir to keep the chip in the led powered until the next switching cycle.
@@schluderjupp Yes, i agree. That is indeed the job of the capacitor, and what clive has said about the LED requiring power to keep information stored to know which is the next color to shine. But i was wondering about the placement of that capacitor. It could have worked even if the capacitor was in parallel with the LED, but it wasn't the case. when the capacitor is charged and supplies energy to the LED, it seems that the battery is in series, and that the capacitor should store a voltage which is greater than that of the battery. So it would seem that the design hits 2 birds with one stone : it powers the LED, AND the current that goes through the LED returns to the battery. It won't be perfect and run forever since there will be some power loss through that process, but the small amount which is being sent back to the battery successfully may lengthen the operation time (although i'm not sure by how much).
Damonack The internal circuitry of the chip keeps current from flowing back to the battery. Current only flows to the led when the chip disconnects LX from ground. When LX is connected to ground, it charges the Inductor from the battery . So the Inductor never charges the battery.
@@schluderjupp ... Yes charging the inductor happens when LX is pulled to ground by the chip. When LX is released, the current from the inductor passes through the diode to the capacitor. The only way the LED can be powered by the capacitor is if the capacitor's voltage is greater than the battery. Follow the diagram : the LED's Anode goes to the positive side of the capaciror, the LED's cathode is connected to the battery's positive side, and both the battery and the capacitor's negative side are tied to ground. The circuit is made, and the only way for the LED to be powered is if the capacitor's voltage is greater than the battery's voltage. Then the current does flow into the battery's positive side when the LED is powered and the battery is not charging the inductor.
I wonder if this would work just by scaling up the panel and the battery by a factor of ten or a hundred.
Would be able to run something far more useful than an LED practically for free then.
That's rather intriguing how you increase the current through the LED by *lowering* the inductor value. Would have thought bigger inductors would store more energy, therefore more current? Perhaps it's because the larger value is actually de-tuning the circuit in some way?
This is actually very simple. Larger inductor will have lower current, if energized in same time with same amount of voltage. This is probably Ton=constant chip and Toff=needed for CrCCM. And that explains everything.
Yes, i thought that was strange as well.
Not strange at all, if you look at the math in en.wikipedia.org/wiki/Boost_converter#Continuous_mode, in particular, the equation for Ioff. @560uH, Ioff=.5×(200kHz)^−1/560uH × 2.5V=11mA, @47uH, Ioff=.5×(200kHz)^−1/47uH × 2.5V=132mA. I haven't accounted for a bunch of other stuff here, so my numbers don't match the datasheet, but the math works out close enough.
Quite simply a larger inductor resists current change more than a smaller one, so for a given pulse length on the output of the IC a smaller inductor will reach a higher peak current. A second order effect is that lower value inductors of the same size tend to have lower series resistance (fewer turns of wire, often thicker wire).
Why do they pulse like that though? Why not switch to using the battery?
It allows the use of a single 1.2V cell to drive a 3V LED. That saves cost and also prevents cell damage in stacks of cells when the first to go flat gets reverse charged.
Those little chips are great! I use them to power tiny high voltage transformers directly. They get a little warm doing it, but who really cares about that.
I suppose you could use the chip's output to drive a transistor for higher current.
Easily, I was just impressed they could do it alone. 2 part mini HV supply, That's about as simple as it gets.
Don't think anyone uses the YX8108 for solar lights anymore, The lights I've bought & made have the newer QX5252.
I've been tinkering with a similar solar light with a similar YX8058 chip and the same button cell (1.2v 40mAh). While waiting on some higher capacity batteries to replace the button cell, I managed to blow up my first button cell by charging with too high current, then too high voltage. My reaction was... OH COOL! Perhaps I should secure my explosion containment pie dish sooner than later. 😆
If you re-draw your schematic, you'll see it looks exactly like the schematic for a boost converter: en.wikipedia.org/wiki/Boost_converter
JW said CE stands for "cromulent electronics". The CE is stamped on lots of stuff and also means caveat emptor. (buyer beware).
I like the later explanation! Who is JW? Only JW I know is Jehovah's Witness.
John Ward
@@templebrown7179 Mister Ward does deadpan delivery to perfection. Search above for video ID: 0FtJn1oD0Fw
Also handy for Joule Thief circuits, which since getting a bag of 100 (actually 101, cos china can't count!) BC547s, I've made a few of, one of which is annoying as it sounds like a CRT flyback, but that was my first attempt, I'm still learning though... :D
It's only now that I've noticed that your background has changed…
(Edit: Thanks for the explanations!)
It's a seasonal change. The Big Clive migrates at this time of year to the Edinburgh area.
He's doing the Edinburgh tattoo at this time of year. I am sure we'll get some videos on it soon
Thanks for the explanations!
I do like pound shop electronics for how elegant and simple they are.
datasheet www.solidfiles.com/v/NaLamYWagdBWZ
there is couple of those datasheets, I'm not sure, maybe is there more then one company making this chip
also from what i know there is no official English datasheet for this
Too lazy to search but i'm sure many noticed that the led has cathode on the plus rail in the schematic
It does and that is how it should be. Listen more carefully to Clive's explanation.
I made a YX8018 circuit with a transformer like in a joule thief to get 6-7 volts my battery is a AAA 1.2v 200mAh
Is this not a jewel thief circuit? Just with a diode and an inductor.
I built my first computer with RGB lighting today, Unfortunately Red=Green and Green=Red, blue is OK. The wiring is in a molded connector so no swapping the wire/pins around. Oh well :)
Most RGB software will have settings that can correct it
In the software you have to pick the right group, ie rgb gbr grb etc. This is the same with any of the LED's that use a controller box. Atb nick
Hello big Clive it's that first aider from the tattoo here I have an old light that would be cool for you to have a look at
Hi Nathan. What sort of light is it?
@@bigclivedotcom it have an atex headtorch and led lenser head torch that both dont appear to work as well as they used to I have multiple old lights etc that I could bring along to the tattoo and let you see what may be of interest to you there just lying around collecting dust now a days
Although the 1N4008 might actually do the job, all these are mains rectifiers and pretty terrible for high frequency switching applications. Use a cheap Schottky!
It looks like a 1N4148 fast switching signal diode, these are much faster then mains rectifier diodes like the 1N400x series.
@@Mark1024MAK Must misheard then, usually I'm listening to the videos as podcasts, there is something about Clive's voice.
rain drowned the sound. never realised before just how much of your video goes through my subwoofer...
Dolby seems to treat my voice as a bass track.
@@bigclivedotcom Should you ever start to stutter i take you on as dubstep artist.
Why does the current increase when inductance decreases? Not very intuitive.
A lower value inductor can "eat" more energy when the chip's switch is turned on for a fixed time.
Clive I wonder if this would fix the problem I had with a idea I wanted a red flashing L.E.D in a solar light but it did not work. would it?
Most likely it would. Again, a flashing LED uses a small integrated circuit to cycle power on and off. If the power to that integrated circuit is being cycled, the chip won’t flash. Try it. Let us know.
Novice here please excuse me. How did you know that was an inductor and not a resistor? Looks a lot like a resistor.Thanks
Very similar looking components. The colour is sometimes a clue, but the biggest clue is the application and position within a circuit.
@@bigclivedotcom thank you
And when tested using a multimeter switched to the resistance range, inductors / chokes have a much lower resistance that does not match the colour code.
@@Mark1024MAK thanks Mark. I was pretty sure that would be the case. I was wondering if there was a way by just looking at it.
the same as every other shitty solar step-up light. good info. I'm actually swapping rgb to warm white as i watch but notes taken
Had this issue.
Wow... I got a solar lamp that had a 80mAh button cell and it only lasts 4hrs using white led. Eventually I had to mod in another cell in parallel. Surprised they only supplied 40mAh in a color changing unit.
tensazero By changing the Inductor you can raise or lower the current to the led and make it run longer/dimmer (or brighter/shorter). A larger battery often doesn't help because the solar panel can't deliver enough energy to totally charge the battery. Usually a larger Inductor value will lower the current to the led. Just try it out. 😀
Hum! I wonder how big a battery this little chip can charge.
Susan Amber Bruce As big as you like, but you have to remain within the switching current limit, so a big battery may take as impractically long time to charge.
Yes, I'll admit it. I often buy lots of Chinese dirt cheap trinkets like this (shipped free!) just to strip the parts out.
Where can I buy these? I just want something like the part shown in this video, not a complete outdoor ornament.
It's cheaper to buy it as part of a small light.
if the capacitor would've been across the LED wouldn't the inductor be unnecessary ? EDIT: I totally forgot it was NI-MH not Li-ION :facepalm
The inductor is necessary. Cannot get voltage step up without it. LEDs need more than the 1.2 volts from a single NiMh cell.
oops I forgot it was NIMH not LI-ION. Thanks for awakening me :) I ofcourse know that 1.2 needs to be boosted for a LED