# Basic DC electricity question

Page 2 of 4
• posted on February 16, 2007, 9:17 pm

On Fri, 16 Feb 2007 12:58:43 -0600, Mark Lloyd

If you're going to be pedantic about it, "a resistor" isn't equivilent to "resistance", or to "limited current", either.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 16, 2007, 4:53 am
Chop Suey wrote:

For the sake of my education let me ask. Doesn't an LED have a predictable voltage drop across the junction? I seem to recall 1.5 volts is that correct? If it is I only need as many LEDs in series as will equal the voltage and they will need no external resistance. Is my memory wrong? -- Tom Horne
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 16, 2007, 7:18 am
snipped-for-privacy@gmail.com writes:

The voltage-current characteristic of an LED has a pretty sharp "knee" characteristic of most diodes. Very little current flows as voltage increases up to the turnon voltage, and then current rises very rapidly with very little change in voltage. The actual voltage depends on the type of LED (different colours use different chemistry and have different voltage) and temperature.
LEDs are easy to drive from a constant-current power source. You set the current you want, and the power supply adjusts its voltage to whatever is needed to make the LED draw that amount of current. Put a bunch in series, and all get the same current.
But most power sources are closer to constant-voltage instead of constant-current. It's nearly impossible to set the voltage needed for a particular current, since a small voltage change results in a large current change, and the voltage needed depends on temperature (which depends on current among other things!).
So you generally use a resistor in series with the LED, or string of LEDs. The power supply voltage is equal to the sum of the voltages of the LEDs in the string, plus a few extra volts. The resistor is sized to give the desired LED current with the "extra" voltage applied across it. This gives a system where small changes in LED turnon voltage produce small changes in the voltage across the resistor, which produce small changes in current through the LEDs. Without the resistor, you'd get large and unwanted changes in current.
As others pointed out, it is possible to build a system using only a LED and a battery *if* the battery's own internal resistance is enough to stabilize the system.
Dave
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 16, 2007, 7:08 pm
On Fri, 16 Feb 2007 04:53:13 GMT, Thomas Horne

It varies depending on the LED. In an electronics catalog I got last year, I see forward voltages ranging from 1.5V to 4.8V (that high one was blue). I'd expect it to be constant for a particular LED.

Maybe if that voltage is EXACTLY right. Theoretically, you might see this a few times during the life of the universe. In reality, don't expect it.
--
Mark Lloyd
http://notstupid.laughingsquid.com
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 15, 2007, 12:21 am

A LED acts as a voltage regulator (regulating the voltage across it to about 2V). It will draw as much current as needed to do this. If the LED is across a 12V battery it will "happily" destroy itself trying to maintain 2V across that 12V battery. I made that mistake once. Immediately, there was a POP and half the LED disappeared. If the supply voltage is below that, you don't get any light at all.
The flashlights that use LEDs without resistors use batteries that cannot supply more current than about 20mA. The nearly-infinite load from the LED lowers the battery output voltage until it's below that which the LED will conduct. Then the battery recovers and the cycle begins again.
That is, you need a battery with limited current capacity. You're using the battery's internal resistance (Rth, if you care) as the LED series resistor.
--
Mark Lloyd
http://notstupid.laughingsquid.com
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 15, 2007, 2:19 pm

I hooked them up in parallel and they seem to work just fine. Incredibly bright too.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 15, 2007, 5:38 pm

LEDs in parallel with each other? It's unlikely that all would light, since the threshold voltages would be slightly different, and the one that's lowest would prevent the others from lighting.
As someone said earlier, your LEDs probably are actually LED modules, and come with built-in resistors. You have a separate resistor in series with each LED.
Modern LEDs can appear very bright. I noticed that with the holiday lights I had this year (yes, I know that's "last year", but it is still less than 2 months ago). Those LEDs look brighter than the miniature incandescent's.
--
Mark Lloyd
http://notstupid.laughingsquid.com
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 15, 2007, 8:02 pm

If they're all from the same batch, they're probably pretty well matched in forward voltage, and it only takes a little bit of series resistance within each LED to approximately balance the current between LEDs. This isn't a *good* way of connecting multiple LEDs, but it's not automatically doomed to failure.

The little 1 W and 3 W LEDs appearing in flashlights are now brighter than anything I ever saw from any incandescent bulb in the same size of flashlight (2 AA battery).
Dave
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 16, 2007, 7:11 pm
On Thu, 15 Feb 2007 20:02:12 +0000 (UTC), snipped-for-privacy@cs.ubc.ca (Dave Martindale) wrote:

Yes. I have one of those 3W LED flashlights. It's brighter and whiter than an incandescent flashlight.

--
Mark Lloyd
http://notstupid.laughingsquid.com
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 15, 2007, 9:18 pm

He said 3V LEDs in one post - that implies the internal resister.
Bob
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 16, 2007, 7:12 pm
wrote:

Not necessarily. As I said in another message, some LEDs have forward voltages that high. That's the LEDs themselves.

--
Mark Lloyd
http://notstupid.laughingsquid.com
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 16, 2007, 9:54 pm
wrote in message

Considering the circumstances, I doubt that that is the case here.
Bob
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 17, 2007, 1:47 am
wrote:

What I said definitely IS the case. What I said is just to disagree that "3V" IIMPLIES "internal resistor". Nothing else.
--
Mark Lloyd
http://notstupid.laughingsquid.com
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 17, 2007, 2:21 am
wrote in message

These are "3V LEDs" provided for a school project. These would seem very likely to be LEDs designed to operate easily off of 3 volts, since LEDs are rarely described as "X volt" in any other circumstance. Certainly not at the "consumer" level.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 14, 2007, 5:32 pm

Nothing wrong with that as long as they are in series.

No. a good mix of math and science.

Go back to Ohms law. 2X resistance = 1/2 current

Aahhh... Remember that we've been talking incandescent light bulbs. A LED is a diode and not a bulb even though it gives off light. Although the same principles apply since LED's use a dropping resistor to limit current, similar to items in parallel.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 14, 2007, 5:38 pm

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 15, 2007, 12:24 am

2x resistance * 2x voltage = 1/1 current (as in 2 6V bulbs in series on 12V).

LEDs can be connected in series as long as the voltage is high enough. Then, only one resistor is needed.

--
Mark Lloyd
http://notstupid.laughingsquid.com
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 14, 2007, 6:15 pm

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 14, 2007, 6:29 pm

So what's the problem? Connect them in series.

Too much cumulative resistance at some point, anyway -- not necessarily all the way at the end.

Browse around at the web site for Edmund Scientific //http://scientificsonline.com / Lots of good stuff there.
--
Regards,
Doug Miller (alphageek at milmac dot com)
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
• posted on February 15, 2007, 12:25 am
On Wed, 14 Feb 2007 18:29:08 GMT, snipped-for-privacy@milmac.com (Doug Miller) wrote:

Or in parallel if the power supply is big enough. C cells may not be.

--
Mark Lloyd
http://notstupid.laughingsquid.com