LED shining with 0.5mA?

Hello,

I tried to regulate the brightness of a green LED and noticed that it doesnt matter if I put 200, 300 or 1000Ohm so I connected a 10kOhm resistor and see what it does then, and its still shining(even though less bright).

This doesn't really make sense to me since to my knowledge the digital outputs give 5V which would give me a current of 0.5mA. Are there ulta low current LED's or what the hell is going on?

Im using an UNO, and yes everything is connected correctly.

An LED will shine with just about any current. Just brighter with more current. Whether you can see it depends on the ambient light level and the sensitivity of your eyes. An LED will light up (dimly) through the pull-up resistor of an Arduino pin (20-30kΩ).

But with a 10k Resistor the forward voltage couldn't be reached could it (which is normally around 2V)? Or is that also much lower than one might think?

Fennec:
But with a 10k Resistor the forward voltage couldn't be reached could it (which is normally around 2V)? Or is that also much lower than one might think?

I don't know why you think that or what you might think, if I did I would be able to explain. I'll try anyway, but if this doesn't make sense to you please explain your thinking behind that statement. An LED has some forward voltage, which depends on the colour and some manufacturing details. For sake of argument suppose it is 2V. That means that if you pass some current through it then 2V will appear across its terminals. As long as you don't exceed the maximum current the LED can take then it will always be (about) 2V. For a 5V supply the rest of the voltage, in this case 3V, appears across the series resistor. In the case of a 10k Ohm resistor the current will be 3V / 10000 = 0.3mA.

My experience is that green LEDs do not need much current to shine. I'm currently designing a box with a few LEDs (status indicator) which will be used in darkness.

Since I did not want any of the LEDs to be too bright, I check using different resistors for green, red, blue, yellow and pink LEDs. I ended up using the following resistor values:

Green, 27k
Blue, 6k8
Pink, 7k5
Yellow, 1k8
Red, 12k

But with a 10k Resistor the forward voltage couldn't be reached could it

The voltage drop across a resistor depends on the current flowing through it.

Vdrop = I R

The voltage drop across an LED behaves very differently, but you can assume it is roughly constant.

PerryBebbington:
I don't know why you think that or what you might think, if I did I would be able to explain. I'll try anyway, but if this doesn't make sense to you please explain your thinking behind that statement. An LED...

I was replying to your post when it struck me. So the 2V drop across the LED either way and the only thing that the resistor controls is the amount of current that flows, I was thinking completly backwards.

Even though I calculated tons of diode circuits in tests and exercises, when it comes to actually building a circuit in real life it feels like you forgot everything you learned theory wise. Really helps deepen your understanding when you connect theory and practice.

Thanks a lot

Remember, everyone's visual color acuity is not the same as yours. What may seem dim to you may be bright to someone else.

Paul

Very much summarised:
Red LEDs are just darned efficient by themselves, so need little current to be really bright.
Human eyes are super sensitive to green light, making those LEDs appear very bright while they’re really not.
All other colours suffer from less efficiency (the lower the wavelength the less efficient) plus lower sensitivity of your eyes.

That sounds right to me. If you wire it up with 1k, and with it powered, touch a 300 ohm across the 1k resistor, you should see the brightness change. Or wire us two otherwise identical LEDs with 1k and 300 ohm resistors.

I dont think that would be noticable if you were fully turning it off between changing the resistor - the eye/brain isn't very good at picking out brightness when you cant fo a side by side comparison.

I normally default 1k resistors.

wvmarle:
Red LEDs are just darned efficient by themselves, so need little current to be really bright.
Human eyes are super sensitive to green light, making those LEDs appear very bright while they're really not.
All other colours suffer from less efficiency (the lower the wavelength the less efficient) plus lower sensitivity of your eyes.

Interesting and quick Google confirms the eye is most sensitive to green light. However for some reason "white" lights prefers blue light - I believe it is due to the best cost/lumen ratio for the blue LEDs. Do you know why it is if not due to the eye sensitivity?

LED brightness is approximately linear with current up to the maximum rated current for the LED (or a bit above.) For a modern LED rated at 10000mCd or more @ 20mA, you're still looking at 250mCd of output, which is a lot more than "normal" LEDs output "back in the old days."

Modern white LEDs use a phosphor - that's why they have this opaque yellowish patch instead of a piece of clear plastic.

Old white LEDs are RGB, don't know why they have that blue hue.

Maybe I am confused somehow but this is problem with the modern white LED lights. The cheapest and most readily available are the "cold white". For some reason unknown to me "cold" means more blue light (radiation of a black body should have relatively more red light than a hotter one).
When someone goes for cost they will get the cold white LED light which produce unpleasant light. It is only inconvenience until you get into a field where colours are important. For example during a surgery most tissues are of some shade of pink. When you get lights that have mostly blue light they are blindingly bright when shining elsewhere but you still see nearly nothing in the wound.

Also blue is the preferred colour for i.e. indication diodes. Backlight of displays have a lot of blue light to get a nice looking parameters (and forcing you to activate some "eye saving" feature that reduces the blue light) and so on.

For some reason unknown to me "cold" means more blue light

I assume (I don't know for sure) because of ice being blue so the association is with cold ice.

It's because we perceive reds and oranges as "warm" and white and blue as "cold". orange, yellow light is the color of warm fires, and is emotionally comforting ("warm"), while bluish whites are the color of the moonlight, of light reflecting off snow, etc, and is emotionally cold.

Anyway - the reason RGB generates an awful white is that you get three narrow band sources, each around our eye's peak sensitivity for that color. But there's nothing inbetween, so many colors look wrong (as the object might reflect those intermediate wavelengths not present with RGB led to different degrees than the three wavelengths that are, and that contributes to your perceptions of the colors), and your eye can tell that there's something not right about the light.

One of the reason that even warm white LEDs often don't look right is the shape of the emission curve - there's a spike at blue (blue from the blue that pumps the phosphor making it through), then a chasm, then the emission of the phosphor picks up and covers green and red with a relatively smooth emission curve.

Smajdalf:
Also blue is the preferred colour for i.e. indication diodes.

I thought that is still red. The first colour to be readily available in LEDs. At least it's my default for things like power LEDs, or other places where the colour has no obvious meaning, and it just matters whether it's on or off.

White LEDs have come a long way over the past few years. Not so long ago we had the choice of red + blue or "full spectrum" LEDs for growing lights, the latter looking a lot better but still having a strong purple hue. RGB LEDs may have worked as well but just a few years ago were just too expensive in comparison. Modern day white LEDs work great for growing, and look a lot better, which is important for home use (not so much in greenhouses of course).

For warm and cold, maybe it's simply because water taps are always marked that way. Red for the hot, blue for the cold.

Fennec:
But with a 10k Resistor the forward voltage couldn't be reached could it (which is normally around 2V)? Or is that also much lower than one might think?

LEDs (and semiconductor diodes in general) are non-linear devices, that is, the voltage and current are not linearly related. Thus "normally around 2V" applies only at the typical rated current.

Extended discussion here: http://lednique.com/current-voltage-relationships/iv-curves/

I also generally default red or orange LEDs for power and indication, particularly at 3.3v (for the boards I sell, I finally settled on a consistent LED choice for LED_BUILTIN - red for boards with no reg, orange for 3.3v (both of these have forward voltage around 2v), white for 5v (forward voltage around 3v) - paired with 1k resistors, this keeps the forward voltage well below operating voltage, while still keeping the current no more than a couple of mA at the chip's maximum operating voltage of 5.5v; I considered using green for the 5v ones, or blue - but the white is more distinctive; branding)

I think people often use blue because it looks 'futuristic', and green because they look brightest for a given amount of current, since the eye is most sensitive to green. I hate green and blue power LEDs - i put blobs of black acrylic paint over them. I used to be much worse at sleeping when there was stray light, and had even the dim red and orange power lights blocked out in my bedroom. Which is also my workshop, hence tons of powered devices.

For growing, if you are optimizing for growth efficiency (ie, yield per kWh of electricity) you want the red and blue grow lights. Chlorophyll has two absorption peaks, one around 440nm (royal blue) and one around 660nm (deep red). You know how plants look green? That's because the leaves reflect green light, instead of absorbing and using it. The whiteish grow lights are more aesthetic, which matters if your plants are in the open in your home, as in the indoor/hydroponic grown herbs and veggies that may double as decorative elements - but the plants would be just as happy with just the deep red / royal blue (which would be cheaper to run for given amount of plant growth, since you're not wasting power to produce green light that they can't use).

I'm surprised that chart shows white as a higher forward voltage than blue - maybe that's the difference between the 450-460nm blue (which our eyes are more sensitive to) vs the 440nm royal blue that is used to pump the phosphors? Because like, as we've been over, a white LED is just a royal blue LED with a blob of phosphor on top.

DrAzzy:
For growing, if you are optimizing for growth efficiency (ie, yield per kWh of electricity) you want the red and blue grow lights. Chlorophyll has two absorption peaks, one around 440nm (royal blue) and one around 660nm (deep red). You know how plants look green? That's because the leaves reflect green light, instead of absorbing and using it.

I know, just taught a 6-hour class on the subject of hydroponics earlier today :slight_smile:

I have seen several research papers recently that indicate that plants actually do benefit from the rest of the spectrum - IR not so much (that just heats up and burns the leaves), but they do absorb some green and into the near UV. I have seen growlights that include a few UV LEDs even. How much more beneficial it is, is not clear. It is of course a point that plants have evolved in the full spectrum of sunlight, not the limited red/blue LED spectrum.

Application and purpose matters of course big time. In those red/blue LEDs the plants themselves look completely black. Very interesting to see, and it looks totally wrong even though I know perfectly well what is going on. I'm mostly looking at home growing myself, so aesthetics are very important.

Not so long ago fluorescent light was also still popular, largely for the low investment, also excellent light for growing. Can't imagine that to last long though considering how fast LEDs develop.