Go Down

Topic: Why my LED dies when connected directly to a 9v battery? (Read 19178 times) previous topic - next topic

KenF

That was reply #55 back a few pages ago.  I guess this is destined to be a perpetual thread.
I don't think we've covered the possibility of using one LED as a current limiting device for a second LED.  Since the resistance of the average LED is about 120 ohms, two LEDs in series with 5v driving them would be about right wouldn't it?

raschemmel

And if the forward voltage is slightly different for one of the leds , then what ?
Arduino UNOs, Pro-Minis, ATMega328, ATtiny85, LCDs, MCP4162, keypads,<br />DS18B20s,74c922,nRF24L01, RS232, SD card, RC fixed wing, quadcopter

Paulcet

I don't think we've covered the possibility of using one LED as a current limiting device for a second LED.  Since the resistance of the average LED is about 120 ohms, two LEDs in series with 5v driving them would be about right wouldn't it?
That would be a "wind up", right?!


rdfeil

Ken,

What color combinations will work to limit current properly at 5 volts?  Does the red limit for the blue or visa-a-versa?  How about a white and a green....  I won't even ask about a RGB in the mix  :smiley-money: ...

R

And there is the pitch ............................

jarrod0987

Simple answer: Too much power.  Try 2 AA or AAA batteries and 60 ohm resistor in series.

racemaniac

#96
Jan 22, 2015, 07:52 am Last Edit: Jan 22, 2015, 08:28 am by racemaniac
And if the forward voltage is slightly different for one of the leds , then what ?
it would not be the stablest way to do it (because as mentioned often in this thread already, the forward current is dependent on temperature etc...), so it's a bad idea. But if you take enough margin, it would certainly work :).
Btw, what did you mean by that reply? when putting leds in parallel slighty different forward voltages are a pain in the a**. one led will go too bright, one will hardly give light. In this case we're putting them in series. They can have completely different forward voltages, as long as their total forward voltage is higher than what the supply is giving, you're kind of safe (but it's still a bad idea). They're in series, so the same current will go trough both of them :).

My issue with the discussion is there are people here taking the mantra that you shouldn't put a voltage on leds (because they're current driven) literally. The seem to take it to mean that putting a voltage on a led and getting something reasonable without putting a resistor in the chain is impossible... it's obviously not the best idea, but it's certainly possible...
They also seem to think that a constant current source somehow delivers current without just being something that gives feedback to the voltage it's giving to keep the current constant. Current is always the result of a voltage. So even if you're feeding your led a current, it's just a system that regulates its voltage to keep the current constant. So whatever you're doing, it's a voltage difference driving the led (but one with the proper feedback mechanism).

Maybe i'm looking at it from a too technical point of view, but i'm pretty sure i understand the situation very good by now ^^. I certainly learned to be more current minded when working with leds (and as soon as i have half an hour of spare time, i'm going to take a led, fire up my variable supply, and do some tests plotting its current vs voltage behaviour, it's gonna be interesting ^^. I also HAVE to try to give it some temperature differences to see the effect of that :D). For me without having tried it yet, my main feeling is that some here might be slightly exaggerating how sensitive a led is to voltage changes in the region of its forward voltage. will a 1% difference in voltage suddenly double the current? I would expect the answer of that to be no, but it'll be closer to the truth than i previously realised ^^.

raschemmel

#97
Jan 22, 2015, 08:35 am Last Edit: Jan 22, 2015, 08:37 am by raschemmel
The voltage drop across EACH led is a function of the voltage drop across BOTH leds. If one led has a lower forward voltage than the other it will draw more current. Whether the other led limits that current depends on the forward voltage of the other led. Both will initially be off, then as the voltage rises , one of them will conduct , drawing more current. As the voltage increases further , the other will conduct.  The first led cannot draw the current it needs to turn on completely because it is limited by the second led, which for that brief moment acts as a current limiter. It is not until the total voltage drop across BOTH leds is sufficient to turn on both leds that they both light up. One cannot light without the other.  When VTotal = VForward-1 + VFoward-2 , then BOTH leds can conduct. Think of it like a handgun with TWO safetys. You can take Safety-1 OFF, but it still won't fire until you take the second Safety Off. When the total voltage = Vf1+Vf2, they both light, but when they do , the current will rise exponentially in both leds if the voltage further increases. It is impossible to know which of the two leds will self-destruct first. It's like a game of chance. You increase the voltage until they both turn on. If you increased it gradually enough you can hold the voltage at a safe point once they are both lit. If you continue to increase the voltage , one of them will suddenly get very bright and burn out. If you measure the forward voltage on both of them as soon as they both turn on, and note which one has the lower forward voltage , then you can reasonably assume it will be the first to self destruct with an increase in voltage.  What is the minimum voltage for turn on and what is the maximum without damage, can only be determined empirically.
Arduino UNOs, Pro-Minis, ATMega328, ATtiny85, LCDs, MCP4162, keypads,<br />DS18B20s,74c922,nRF24L01, RS232, SD card, RC fixed wing, quadcopter

AWOL


racemaniac

#99
Jan 22, 2015, 09:48 am Last Edit: Jan 22, 2015, 09:59 am by racemaniac
<long explanation>
I'm sorry, but the forward voltage is NOT the threshold voltage at which the led starts emitting light. The first thing i did when i got my leds was put a white & red led in series (the sum of their forward voltages would be about 5.0-5.2v)
Then putting the below 5V (about 4.6V i think) from an arduino on usb from it, they both light up nicely (just not as bright as they are capable). And the current is nicely limited.
You seem to be thinking of some magical ideal led that has 1 exact voltage drop. below that voltage it doesn't turn on, above that  voltage it short circuits, and only that exact precise voltage you have to limit the current to make sure it doesn't burn out...
I'm sorry, but that isn't how a real led works either! Constant current sources that control the current by varying the voltage would be unable to drive a led like this!

I think it's time to wrap this one up, please.
May i ask why? i think a lot of interesting thoughts are coming from this issue, like above raschemmel confusing the forward voltage at ideal current with the threshold voltage at which a led starts conducting and lighting up.
This is not some huge wind up or infinite circle, i've learned a lot in this discussion, and have the impression that a lot of other people are also learning (or at least i hope they will learn something, rather than being stubborn).

JimboZA

I think it's time to wrap this one up, please.
May i ask why? i think a lot of interesting thoughts are coming from this issue
Personally, I think it's now a giant pissing contest, na-nana-nana, I know more that yoo-oooo.

But that's just me.....
Johannesburg hams call me: ZS6JMB on Highveld rep 145.7875 (-600 & 88.5 tone)
Dr Perry Cox: "Help me to help you, help me to help you...."
Your answer may already be here: https://forum.arduino.cc/index.php?topic=384198.0

Coding Badly

May i ask why?
Because it is painfully obvious from your response that you either failed to read @raschemmel's post or you failed to understand it.  Or both.

In other words... https://www.google.com/search?q=talking+past+each+other

Quote
or at least i hope they will learn something, rather than being stubborn
Uh huh.  Maybe this will help...
https://www.google.com/search?q=pot+calling+the+kettle+black


racemaniac

#102
Jan 22, 2015, 10:13 am Last Edit: Jan 22, 2015, 10:20 am by racemaniac
Personally, I think it's now a giant pissing contest, na-nana-nana, I know more that yoo-oooo.

But that's just me.....
I think it's probably what your mindset is about it. If it's just a pissing contest, then it should be shut down. For me i got inspiration for some led experiments i'm gonna do this weekend, mapping out how leds react to current and voltage, getting to know their behaviour even better.
I also learned a lot better about how i should be careful to not damage leds (or other components) by accidentally putting too much current on them because i was thinking too much voltage wise, not current wise.

I do have the impression that some people believe putting a voltage on a led and getting a sane result is impossible, and i hope they will learn that that is not right either.

But if it's just a pissing contest, then just close the thread and be done with it :).

Because it is painfully obvious from your response that you either failed to read @raschemmel's post or you failed to understand it.  Or both.

In other words... https://www.google.com/search?q=talking+past+each+other

Uh huh.  Maybe this will help...
https://www.google.com/search?q=pot+calling+the+kettle+black


I think i understood his explanation perfectly. But he seems to think of the forward voltage as something very constant, while it isn't. Putting 2 leds in series and giving them a voltage a bit below their ideal summed forward voltage works very well, and is sane to do in my book...

And if i'm really not getting his point, then tell me where i missed his point, i hope it's clear i'm not putting all this effort in this thread to feel self important, but because i learned a lot already, and if there are still things i'm missing, i would love to understand them. I'm for sure gonna do some experiments on leds very soon to confirm everything i learned so far. I'm not in here for the pissing contest.

I'm hoping this thread to explain clearly how it all works, not perpetuate some mystery about how leds magically short circuit if you put a voltage that is not 100% exactly their ideal forward voltage on them...

alnath

I think it's probably what your mindset is about it. If it's just a pissing contest, then it should be shut down. For me i got inspiration for some led experiments i'm gonna do this weekend, mapping out how leds react to current and voltage, getting to know their behaviour even better.
I also learned a lot better about how i should be careful to not damage leds (or other components) by accidentally putting too much current on them because i was thinking too much voltage wise, not current wise.

I do have the impression that some people believe putting a voltage on a led and getting a sane result is impossible, and i hope they will learn that that is not right either.

.........

I'm hoping this thread to explain clearly how it all works, not perpetuate some mystery about how leds magically short circuit if you put a voltage that is not 100% exactly their ideal forward voltage on them...
I think that a datasheet like this one datasheet  (there are plenty of them) gives you all the answers you need. You might want to have a close look at the curves, especially the "Fwd Current vs Fwd Voltage" one .

racemaniac

I think that a datasheet like this one datasheet  (there are plenty of them) gives you all the answers you need. You might want to have a close look at the curves, especially the "Fwd Current vs Fwd Voltage" one .
That's indeed what i'd expect reading that diagram the ideal voltage is 3.5 V, but starting at 2.7V the led starts giving a current, and emitting light. So If you put 2 of these in series, and apply 6V to it, they'll both light up, but the current will be about 4mA, and so they will be pretty dim, and each will have a voltage drop (and at that moment forward voltage) of about 3V (small differences between the leds are of course possible).  Even though their ideal forward voltage to reach their 20mA is 3.5V.
I think i'm understanding it very well?

Go Up