# Electronics LED Resistor Question

Discussion in 'Modding' started by KID52, 28 Dec 2007.

1. ### KID52Member

Joined:
19 Aug 2007
Posts:
335
6
I have no idea how to work this out... basically I have some blue LEDs, I bought them with resistors to run them off of 12v, the resistors have a label on them saying 470ohm, how would I work out what resistor I need to run them off of 5v instead? (I don't know what voltage or anything the LEDs require, all I know is that they work on 12v using the resistors I already bought)

2. ### sammo1999New Member

Joined:
13 Jan 2005
Posts:
378
1
http://led.linear1.org/1led.wiz

So I think...

12 V source with an led requiring 3.3 V and 20 mA would require a 470 ohm resistor.

5 V source with an led requiring 3.3 V and 20 mA would require a 100 ohm resistor.

The variance in the voltage supplied to the LED (that is the 3.3 v part) will determine the brightness of the LED; usually at around 6-8 volts the LED will turn a funny color much more could fry it. So a resistor around 100 ohms should do fine.

I'm not sure if there are any negative effects when running an LED at a voltage higher than it is rated for, say 4.5 v instead of 3.3 to make it brighter. Perhaps it will burn out faster but I doubt that would be a problem unless you go to the point it turns funny colors. Anyone else know?

Hope this helps.

3. ### CinnanderNew Member

Joined:
19 Apr 2007
Posts:
393
2
Up to a point the LED will be fine, but as the voltage goes up the current through it goes up too. This is because, assuming you're far from the 0.7V area, R of the LED is almost constant. As I=V/R, I will increase as V increases, and this is what kills the LED. Power = I{param}R, of which some is light but the rest is resistive heating. So if you run the LED over voltage it will run hotter and, as with transistors, diodes, ICs, etc, have a shortened lifespan unless you heatsink it.

(haxed from an image here)
Suppose the diode has a maximum dissipation of 50mW before it runs the risk of being damaged, then using anything over 0.8V across the diode will shorten the life (such as if your resistor was too small). This isn't a particularly good example, though, as the curves would ideally go up to 5V or so.

(better example, haxed from here)

Last edited: 28 Dec 2007
4. ### KID52Member

Joined:
19 Aug 2007
Posts:
335
6
Thanks a lot for the help guys, one last question, does anyone know where to get cheap 100ohm resistors from in the UK? Maplin sells them for 13p each which is pretty expensive... cheapest I have found them for so far is £2.50 for 100

5. ### capnPedroHacker. Maker. Engineer.

Joined:
11 Apr 2007
Posts:
4,381
241
eBay. Although that £2.50 offer seems OK 2.5p/each is not too bad. Although I guess you want a lot less than 100, right?

6. ### KID52Member

Joined:
19 Aug 2007
Posts:
335
6
Had a look on ebay, cheapest I could find was 100 for £0.99 + £1.50 delivery... so £2.49

7. ### r4tch3thmmmm....

Joined:
17 Aug 2005
Posts:
3,166
48
Resistors are usually bought in larger quantities so buying them individually costs alot. Maybe its time to start getting into electronics and build up a collection of parts. I needed 3 200K resistors so bought 2000, worked out cheaper than buying 5 individual ones and now I will have 200K resistors whenever I need one.

8. ### cpemmaEcky thump

Joined:
27 Nov 2001
Posts:
12,328
55
If you can wait up to 3 weeks for delivery, about 5p for 10 with around £2 delivery is about as cheap as it gets.

But try your LEDs on 5V with the 470R resistors, they'll be bright enough for indicator use on 3mA.

Cinnander, you've posted the curve for a silicon diode. LEDs have different chemistry and higher forward voltage, and no way is 'R' constant; current goes up rapidly with small increases in voltage, the resistor is there to limit that current.

9. ### CinnanderNew Member

Joined:
19 Apr 2007
Posts:
393
2
I know, I did wonder if that would get picked up on It was the overall shape of the curve I was after though, but you caught me red handed! I was trying to demonstrate that as the voltage across the diode goes up the current will increase and it'll quickly expire.

This is what I meant by R is constant - it effectively is if you are sufficiently away from the forward voltage area (and avoid the 'smoke and flames' area). That is, the characteristic curve has a [nearly] constant gradient. Here's a better example. Hopefully that clears up my meaning (though I appreciate that it will be not a perfectly straight line, but near enough)
(I thought LEDs started to conduct at about 0.7V too? ah well! Learn something new).

Last edited: 29 Dec 2007
10. ### Jaguar_InfinityJust some guy

Joined:
8 Oct 2003
Posts:
363
1
Just a question, if you braught 12v LEDs instead of 3.3v LEDs could you just wire them straight into the 12V loop without the Resistors? same question with 5V leds on a 5v loop

11. ### KID52Member

Joined:
19 Aug 2007
Posts:
335
6
Thanks for all the comments and help, that link for the resistors is great thanks, I also tried using the resistors I have like you said, the LEDs were still very bright so I suppose I can use them

12. ### r4tch3thmmmm....

Joined:
17 Aug 2005
Posts:
3,166
48
12V LEDs are usually standard LEDs with a resistor built in, so if it says 12V then no, you don't need a resistor.

Joined:
27 Nov 2001
Posts:
12,328