Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Newbie LED current/voltage/resistor question

0 views
Skip to first unread message

Dirk Puslich

unread,
Feb 7, 2002, 4:14:11 PM2/7/02
to
Hello,

I am a total newbie but am starting to learn electronics. I have some
dumb questions for you electronics gurus.

I have a blue LED that is rated at 5 Volts and 20 milliamps of
current. Okay.

If my power supply is 12V, then using Ohm's law I would take R = V/I =
(12V-5V)/0.020, which means I'd have to put an 350 ohm (or close to
350) resistor before the LED, correct?

Here is my problem: What if my power supply is 5V, which is the same
as the rating of the LED. In that case, I can't put a resistor on it
or the voltage will drop belw the required 5V. But, if I put no
resistor on it, then Ohm's law says that the current will be 1 amp, or
1000 milliamps. Wouldn't this fry the LED?

Does the LED only "draw" the 20 milliamps even though 1000 is
available (pardon the terminology)? Is voltage the only thing that
matters when it comes to resistors (i.e. is the 20 milliamp rating
just saying that for the required resistance calc we assume the LED is
taking 20 milliamps)?

What I am trying to do is use a PC's power supply cables to power a
blue LED, so there is a 5V source and a 12V source. Does it matter
what kind of resistor I use (e.g. is ceramic okay)? Also, I want to
put heat shrink over the resistor and the wires (for safety), will
this cause too much heat or other problems?

Also, what is meant by "forward voltage" and "forward current" vs.
just "current".

Thanks!

Edmond Cheng

unread,
Feb 7, 2002, 5:03:20 PM2/7/02
to

> I have a blue LED that is rated at 5 Volts and 20 milliamps of
> current. Okay.
>
> If my power supply is 12V, then using Ohm's law I would take R = V/I =
> (12V-5V)/0.020, which means I'd have to put an 350 ohm (or close to
> 350) resistor before the LED, correct?

**** Correct!

>
> Here is my problem: What if my power supply is 5V, which is the same
> as the rating of the LED. In that case, I can't put a resistor on it
> or the voltage will drop belw the required 5V. But, if I put no
> resistor on it, then Ohm's law says that the current will be 1 amp, or
> 1000 milliamps. Wouldn't this fry the LED?

**** 1 amp is incorrect. Resistor value now becomes the internal output
resistance of the power supply (extremely small). Depending on the
tolerance of the 5V forward voltage and power supply output
resistance, you may get a very high current that burn out the LED
(V+>Vf) or the LED would not turn on (V+ < Vf).

> Does the LED only "draw" the 20 milliamps even though 1000 is
> available (pardon the terminology)? Is voltage the only thing that
> matters when it comes to resistors (i.e. is the 20 milliamp rating
> just saying that for the required resistance calc we assume the LED is
> taking 20 milliamps)?

**** Current is limited by the total series resistance (your external
resistor,
output resistance of power supply and internal resistance of LED.
Internal resistance of LED is very large when applied voltage is
below
Vf and becomes very small when applied voltage is above Vf.

> What I am trying to do is use a PC's power supply cables to power a
> blue LED, so there is a 5V source and a 12V source. Does it matter
> what kind of resistor I use (e.g. is ceramic okay)? Also, I want to
> put heat shrink over the resistor and the wires (for safety), will
> this cause too much heat or other problems?

**** Recommend using the 12V source with limiting resistor. No heat
sink is required.

> Also, what is meant by "forward voltage" and "forward current" vs.
> just "current".

**** "forward" means "forward bias" in contrast with "reverse bias".
When the LED (or diode) is reverse biased, only tiny leakage
current will flow. When it is forward biased, current remains small
until voltage is higher than "forward voltage", then it rises sharply
with slight increase in voltage.
"forward current" is the operating current in forward bias mode.


Scott

unread,
Feb 7, 2002, 5:39:44 PM2/7/02
to

I'm also new to electronics, and I'll be hanging out to see what answers you
get to these questions, you raised some good points.

defaultuser

unread,
Feb 7, 2002, 6:47:10 PM2/7/02
to
Good answers.
Another option is find a 3.7v Blue LED
http://google.com/search?q=blue+led+3.7v&hl=en

John Popelish

unread,
Feb 7, 2002, 5:49:13 PM2/7/02
to
Dirk Puslich wrote:
>
> Hello,
>
> I am a total newbie but am starting to learn electronics. I have some
> dumb questions for you electronics gurus.
>
> I have a blue LED that is rated at 5 Volts and 20 milliamps of
> current. Okay.

This LED may already include some resistance. most blue LEDs drop
between 3.5 and 4 volts when passing 20 ma.



> If my power supply is 12V, then using Ohm's law I would take R = V/I =
> (12V-5V)/0.020, which means I'd have to put an 350 ohm (or close to
> 350) resistor before the LED, correct?

If you have the correct voltage drop for your LED, yes. You might try
several resistors starting with 470, 430, 390, and 360, measuring the
voltage drop across each, and calculating by ohm's law how much
current they are passing. Pick the one that does not exceed 20 ma. I
am betting that the actual drop by the diode will be more like 3.5
volts, and the 430 will be the correct value.



> Here is my problem: What if my power supply is 5V, which is the same
> as the rating of the LED. In that case, I can't put a resistor on it
> or the voltage will drop belw the required 5V. But, if I put no
> resistor on it, then Ohm's law says that the current will be 1 amp, or
> 1000 milliamps. Wouldn't this fry the LED?

You might. Without any resistor, you are depending on the internal
resistance of the LED to set its current, and the current will vary
dramatically with very small changes in applied voltage, and the
current varies pretty strongly with temperature, also, at any given
voltage. This is why LEDs are driven with controlled currents, not
with regulated voltages. If the LED really needs very close to 5
volts to make it pass 20 ma, then it is almost unpractical to drive it
from 5 volts. However, if it is a typical blue LED that actually
needs between 3.5 and 4 volts then you have a bit to waste across a
resistor, to stabilize the current.



> Does the LED only "draw" the 20 milliamps even though 1000 is
> available (pardon the terminology)? Is voltage the only thing that
> matters when it comes to resistors (i.e. is the 20 milliamp rating
> just saying that for the required resistance calc we assume the LED is
> taking 20 milliamps)?

The LED has a very steep increase in current with a small increase in
applied voltage, once it begins to conduct. This relationship is
approximately exponential (fixed multiple for a given increment of
voltage).



> What I am trying to do is use a PC's power supply cables to power a
> blue LED, so there is a 5V source and a 12V source. Does it matter
> what kind of resistor I use (e.g. is ceramic okay)? Also, I want to
> put heat shrink over the resistor and the wires (for safety), will
> this cause too much heat or other problems?
>
> Also, what is meant by "forward voltage" and "forward current" vs.
> just "current".
>
> Thanks!

You need to find out the actual voltage your LED drops when you are
passing 20 ma through it. Forward refers to forward biasing the diode
junction in the LED. That is, applying the more negative voltage to
the cathode lead. The LED makes no light when current is driven
backwards through its junction, and may suffer damage at very small
reverse currents. Blue LEDs are especially sensitive reverse current
damage.
--
John Popelish

Adam Aglionby

unread,
Feb 7, 2002, 6:27:55 PM2/7/02
to
Your Blue LED is probably more like 3.2V foward V try:

http://linear1.org/ckts/led.php

LED resistor calculator linked from LED museum:

http://ledmuseum.home.att.net/ all you wanted to know about LEDs

HTH
Adam
"Dirk Puslich" <dirk_p...@hotmail.com> wrote in message
news:d3099e46.02020...@posting.google.com...

Edmond Cheng

unread,
Feb 8, 2002, 9:01:03 AM2/8/02
to

----- Original Message -----
From: "John Popelish" <jpop...@rica.net>
> >
> > I have a blue LED that is rated at 5 Volts and 20 milliamps of
> > current. Okay.

>
> This LED may already include some resistance. most blue LEDs drop
> between 3.5 and 4 volts when passing 20 ma.
>

> John Popelish

****

If it's really a Resistor-LED (LED integrated with current limiting
resistor) and let's
ASSUME Vf is about 3.5V and rated 20mA, you can easily model that by adding
an LED (3.5V) in series with a 75 ohm resistor. You can verify the
configuration by applying a 5V, measure the current, then 4.5V & measure the
current again. The
2 equations I=((V+)-(Vf))/R should solve for R & Vf.

Edmond

John Popelish

unread,
Feb 8, 2002, 10:45:39 AM2/8/02
to

I would start at 3 volts and carefully work up to the voltage that
drives 20 ma. otherwise you might get a blue flash and nothing more
on the first test. :)
--
John Popelish

Larry Dugan

unread,
Feb 9, 2002, 9:11:37 AM2/9/02
to
Edmond Cheng wrote:

It's been a long time since I did anything in electronics but at one point they
made LED's for 12V and 5V. They had an internal current limiting resistor and
all you had to do is to supply the 12V or 5V to them. It sounds like you have
one of those kind. But if you want to play it safe, do like another poster
suggested, start with a low voltage and slowly raise it while observing the LED.

--
"This country was founded by religious nuts with guns."
-- P. J. O'rourke


Code

unread,
Feb 9, 2002, 11:35:57 AM2/9/02
to
I would use a small transistor, something like a NPN 2n2222 or bc547 as a
current-limiter. A small current into the base of the transistor will limit
the amount of current passing thru the emitter-collecter junction to a
certain limit.

mathematically, 133uA(into the base) x 150 gain = 20mA (collector-emitter)

to get 133uA into the base, just use a 37Kohm (or something near) resistor
pulled up to 5V.

Tan Chun Chiek


"Dirk Puslich" <dirk_p...@hotmail.com> wrote in message
news:d3099e46.02020...@posting.google.com...

Rich Grise

unread,
Feb 9, 2002, 7:29:46 PM2/9/02
to
Actually, an NPN with a resistor from emitter to ground makes a pretty
handy current regulator. The transistor all alone would make for a
pretty touchy circuit. But if you want, say, 20mA through the collector,
then put 5V at the base through maybe 1K or 10K, this will give you
4.3V (approx) at the emitter. 4.3V/.02A=215 ohms, so a 220 would be
suitable. In that case, you can put any voltage at the top that
you want to, as long as it's not enough to let the magic smoke out
of the transistor. ;-)

--
Cheers!
Rich

"We have met the enemy, and he is us!"
- Pogo Possum, ca. 1950's

0 new messages