These issues are more prevalent if you're trying to make high-power LED circuits or some RGB circuit with accurate representation of colors.
Always thought LEDs are my friend but they made me lose some hair in last years.
These issues are more prevalent if you're trying to make high-power LED circuits or some RGB circuit with accurate representation of colors.
Always thought LEDs are my friend but they made me lose some hair in last years.
I got lighting because i was handy with electronics, and that sent me down a rabbit hole of the vagaries of current limiting circuits, buck/boost converters, color dependent forward voltage, temperature management, cabling and dmx control. All while staying on a modest budget and maintaining sufficient safety and durability that a bunch of highschoolers can build it and set it up and tear it down in a timed set change at high school gyms and theaters across the country.
Very interesting experience, highly recommended and i concur, slapping a resistor on it to limit current is focusing on a small and somewhat straightforward portion of the LED spectrum (har har). That said, it’s still useful for folks that just want to slap a light on something.
But at least the article mentioned it,
> It's important that there are n LEDs (connected in series) in each of the m parallel branches of the circuit and that the LEDs all have the same Vled and Iled. Otherwise, all bets are off.
If you're using LEDs for illumination, then do some math.
Also, as a general rule don't put LEDs in parallel unless you really know what you're doing with matching I-V curves.
LEDs are different one from another, so the one with a lower voltage drop will have more current flowing trough it. A lot more, and it will heat more.
Luckily, heating it up will make its internal resistance higher, so all LEDs will end up with the same current. It will shine brighter though, and fail earlier, especially if it heats a lot.
What happens after it fails? If it fails open, which is usually the case IIRC, more current flows trough the other LEDs, making them fail earlier. If it fails with a short, the other LEDs turn off, and the resistor dissipates all the current (with a higher voltage drop, so more current). If you're lucky, it stays at that. If not, you can burn the resistor, and/or the power supply due to increased current.
Always use a single resistor for each LED, except in very specific cases (like putting them in series).
Don't put diodes in parallel.
Did you maybe reply to the wrong comment?
> The oversimplification for me is not talking about PWM
Omitting PWM isn't very important IMO; showing diodes in parallel like this sounds like a critical omission.
How you do that calculation is entirely different when you use PWM.
Hence, I felt like it was an oversimplification around calculating current limiting resistor values.
Also, there are some cases where LEDs have to be in parallel, for practical purposes. Like LED matrix displays, or things meant to be inexpensive and not durable, like holiday lights.
Holiday lights generally aren't in parallel, as driving in series makes for more efficiency (higher voltage) and easier wiring. Once strings are long enough they're probably put directly in parallel, but that's just cheapos being cheap and relying on statistics.
Many of them do, especially the outdoor single color ones, where 1 led per pixel isn't bright enough.
>Holiday lights generally aren't in parallel
Granted, "holiday lights" is a broad category, but there are many that are obviously in parallel. Two 1.5 volt batteries couldn't power many LEDs in a serial string. And many of the AC strings are parallel strings of serial connected LEDS as well. Like: https://midimagic.sgc-hosting.com/chrsleds.htm
There are always going to be exceptions (like your 2 cell battery example, or "LED throwies"), but we're talking about the common scenarios. In general I think talking about putting LEDs in parallel is just confusing for anyone that's reading an article to know what size resistor to use.
Yes, they use a lot of ICs that are either 3.3v or 5v as the driver to the LEDs.
However, resistors are usually a lot cheaper than LEDs. Each LED comes with its own resistor on cheap LED strips from china, sometimes resistors are even integrated in the LED packaging.
As for PWM -- do you have any source on this? I guess you just have to take into account the capacitance of the LED?
Well, mostly just "duty cycle" related. If you're pulsing the LED half the time (50% duty cycle), for example, you can pulse with twice the current you would normally use if it weren't pulsed. Then you control brightness by pulsing for a lower percentage of the time.
I would love to figure out how to make an arbitrary transfer function shape using a pot as input, and some combination of transistors/op-amps to keep the nice analog behavior.
That being said, if you're not on battery power, we're talking negligible energy. So, put a current source in there and stop giving me a seizure with your flicker.