Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Wiring LEDs in Series

8733 views
26 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    September 2004
  • From: Dearborn Station
  • 24,281 posts
Wiring LEDs in Series
Posted by richhotrain on Friday, May 31, 2013 6:46 AM

I am operating an NCE PH-Pro 5 amp system.

I want to wire three LEDs in series and use one resistor.

The LEDs each have an absolute maximum current of 20 ma.

If my DCC system puts out 16 volts, to determine the proper resistor, do I subtract 3.5 volts (LED) from 16 volts (DCC system) and divide the ohms into the result to determine the proper resistance.   If so, that would indicate a resistor with a value of approximately 520 ohms or more.  Is that correct?

Rich

Alton Junction

Moderator
  • Member since
    June 2003
  • From: Northeast OH
  • 17,249 posts
Posted by tstage on Friday, May 31, 2013 7:52 AM

Rich,

If you have three (3) 1.5V LEDs, did you mean reducing by 4.5V rather than 3.5V?

Tom

https://tstage9.wixsite.com/nyc-modeling

Time...It marches on...without ever turning around to see if anyone is even keeping in step.

  • Member since
    December 2004
  • From: Bedford, MA, USA
  • 21,483 posts
Posted by MisterBeasley on Friday, May 31, 2013 7:59 AM

Start with a 1K resistor.  If the LEDs aren't bright enough, put in a smaller one.

It's nice to do the math and know you're in the right ballpark, but in the end what's important is how it looks on your layout.  In my experience, you never want to run lights at their full rated voltage.

It takes an iron man to play with a toy iron horse. 

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Friday, May 31, 2013 8:10 AM

 Or if you mean 3 LEDs in series, each with a voltage of 3.5V, then you subtract 3 x 3.5V. If this is going on the function output of a decoder, measure the output and see what the actual voltage is, it's usually a bit less than track voltage.

         --Randy


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    April 2004
  • From: Ontario Canada
  • 3,574 posts
Posted by Mark R. on Friday, May 31, 2013 9:19 AM

This online calculator might help you ....

http://www.hobby-hour.com/electronics/ledcalc.php

When you select more than one LED, you get an optional drop box to select either series or parallel. For input voltage selection, the voltage output on a decoder function is 1.4 volts less than the track voltage.

Keep in mind, these online calculators provide absolute ratings. Always better to use a resistor value a bit higher than recommended so you aren't running your LEDs at their maximum rating. For example - if the calculator says 680 ohms, you could just as easily use 1000 ohms and not notice much difference in brightness, but you wouldn't be stressing the LED as much.

Mark.

¡ uʍop ǝpısdn sı ǝɹnʇɐuƃıs ʎɯ 'dlǝɥ

  • Member since
    July 2006
  • From: North Dakota
  • 9,592 posts
Posted by BroadwayLion on Friday, May 31, 2013 9:21 AM

LEDs really do not care about voltage. a 3.5 volt LED means that it needs AT LEAST 3.5 volts to light up. 5, 12, 16, 24 volts are all OK as long as you do not pass more than 20mA of current through any one LED.

The resistor is there to limit the current, not the voltage, although it will do both. (if)...

It is not considered good practice to put LEDs in series, although I have done it and it does work. Here is how the LION wired the platforms at Prospect Park...]

I have problems with those that I wired in series, and will have to go back and change them, but yes, they will work that way.

Remember, and LED is more like a restifier than a light bulb, it passes current rather than drawing current.

...(if) LION uses resistors embedded in the track to control train speed. A series of three 5.1 ohm resistors lead into each station. A VOLT METER will not show any reduction in voltage if there is no train motor in the circuit. (Because the circuit is drawing no current). As a train passes each resistor it does slow down a little bit, which was after all the intention of the circuit. So to make a short story a little longer, a resistor will not do anything in your circuit until it has something to do.

ROAR

The Route of the Broadway Lion The Largest Subway Layout in North Dakota.

Here there be cats.                                LIONS with CAMERAS

  • Member since
    September 2004
  • From: Dearborn Station
  • 24,281 posts
Posted by richhotrain on Friday, May 31, 2013 9:35 AM

tstage

Rich,

If you have three (3) 1.5V LEDs, did you mean reducing by 4.5V rather than 3.5V?

Tom

Tom,

I must confess that my limited knowledge of electronics gets in the way when I start threads like this one.  In this case, I don't know the voltage of the LEDs that I am purchasing.  I guess that I thought that LEDs generally were 3.5V.  So, I need to find that out?

Rich

Alton Junction

  • Member since
    September 2004
  • From: Dearborn Station
  • 24,281 posts
Posted by richhotrain on Friday, May 31, 2013 9:41 AM

rrinker

 Or if you mean 3 LEDs in series, each with a voltage of 3.5V, then you subtract 3 x 3.5V. If this is going on the function output of a decoder, measure the output and see what the actual voltage is, it's usually a bit less than track voltage.

         --Randy

Randy, thanks, after you mentioned reducing voltage supply by 3.5V for an LED on another thread, I just drew the conclusion that all hobby LEDs were 3.5V and I did not realize that LEDs in a series were multiplied before subtracing, as you are indicating here.

Rich

Alton Junction

  • Member since
    September 2004
  • From: Dearborn Station
  • 24,281 posts
Posted by richhotrain on Friday, May 31, 2013 9:42 AM

Mark R.

This online calculator might help you ....

http://www.hobby-hour.com/electronics/ledcalc.php

When you select more than one LED, you get an optional drop box to select either series or parallel. For input voltage selection, the voltage output on a decoder function is 1.4 volts less than the track voltage.

Keep in mind, these online calculators provide absolute ratings. Always better to use a resistor value a bit higher than recommended so you aren't running your LEDs at their maximum rating. For example - if the calculator says 680 ohms, you could just as easily use 1000 ohms and not notice much difference in brightness, but you wouldn't be stressing the LED as much.

Mark.

Mark, thanks for that discussion and the link to the calculator, all very helpful.

Rich

Alton Junction

  • Member since
    April 2004
  • From: Ontario Canada
  • 3,574 posts
Posted by Mark R. on Friday, May 31, 2013 10:11 AM

Series and parallel loads have inverse power requirements.

Let's use three 3.5 volt / 20ma LEDs as an example. In series, you add the total voltages of all the LEDs, but the current requirement stays the same - so you would need 10.5 volts at 20ma.

In parallel, you add the currents and the voltage stays the same. So these same three LEDs in parallel would require 3.5 volts at 60ma.

I much prefer to run multiple LEDs in parallel. If one goes bad, the other still works. In series, if one goes bad, they all go out and you then have to determine which one went bad. I always use two LEDs in parallel on a single 1000 ohm resistor for headlights. If one LED goes bad, the 1000 ohm resistor is still adequate for the one left still working without damaging it. Another good reason to use higher than maximum rated resistors.

 

Mark.

¡ uʍop ǝpısdn sı ǝɹnʇɐuƃıs ʎɯ 'dlǝɥ

  • Member since
    July 2006
  • From: North Dakota
  • 9,592 posts
Posted by BroadwayLion on Friday, May 31, 2013 10:11 AM

richhotrain
I must confess that my limited knowledge of electronics gets in the way when I start threads like this one.  In this case, I don't know the voltage of the LEDs that I am purchasing.  I guess that I thought that LEDs generally were 3.5V.  So, I need to find that out?

1) LEDs are not light bulbs. They do not have a "rated voltage"

2) The voltage that you see is the smallest voltage that it will pass forwards (and light up).

3) Once you reach that voltage, assuming that you are not trying to run too much current through the device, it will emit light.. (A light emitting DIODE -- It *is* a diode, not a light.)

4) It will pass any voltage until the current goes over 20mA (usually -- specs do vary), and then it dies.

Think of it as a rectifier... a component in your circuit. It has a job to do and it happens to light up. It does not know that the LIGHT is the product that you want from it, so you cannot power it like you do a light bulb. It needs something else in the circuit, and that something else for want of something else is a resistor.

ROAR

The Route of the Broadway Lion The Largest Subway Layout in North Dakota.

Here there be cats.                                LIONS with CAMERAS

  • Member since
    September 2004
  • From: Dearborn Station
  • 24,281 posts
Posted by richhotrain on Friday, May 31, 2013 10:14 AM

LION, thanks for that explanation.  Very helpful.  I am learning - - - slowly, but surely.

Rich

Alton Junction

  • Member since
    April 2004
  • From: Ontario Canada
  • 3,574 posts
Posted by Mark R. on Friday, May 31, 2013 10:32 AM

BroadwayLion

richhotrain
I must confess that my limited knowledge of electronics gets in the way when I start threads like this one.  In this case, I don't know the voltage of the LEDs that I am purchasing.  I guess that I thought that LEDs generally were 3.5V.  So, I need to find that out?

1) LEDs are not light bulbs. They do not have a "rated voltage"

2) The voltage that you see is the smallest voltage that it will pass forwards (and light up).

3) Once you reach that voltage, assuming that you are not trying to run too much current through the device, it will emit light.. (A light emitting DIODE -- It *is* a diode, not a light.)

4) It will pass any voltage until the current goes over 20mA (usually -- specs do vary), and then it dies.

Think of it as a rectifier... a component in your circuit. It has a job to do and it happens to light up. It does not know that the LIGHT is the product that you want from it, so you cannot power it like you do a light bulb. It needs something else in the circuit, and that something else for want of something else is a resistor.

ROAR

Wow - so many incorrect assumptions !

1) - LEDs DO have a rated voltage. If that voltage is exceeded, they WILL break down and ultimately destroy themselves.

2) - LEDs don't pass voltage, they pass current. Voltage is supplied. A 3.5 volt LED will light even at 2 volts at the required current, only it will be considerably dimmer.

3) - A diode is an inverse voltage limiting device, and also a voltage dropping device. A 20ma rated LED will limit a maximum current flow of no more than 20ma, thats why a resistor is required - to ensure that no more than the maximum rating of 20 ma of the LED is exceeded.

4) - Again, voltage isn't passed - it's supplied. Your SUPPLY voltage cannot exceed the LEDs rated voltage. Again, this is why a resistor is used. The resistor limits the excess current and absorbs the the excess voltage giving it off as heat .... remember Ohm's Law ?

An LED is in essence a load within itself. If the supply voltage and current rating are a match to the specifications of the LED, no resistor is really required. A 3.5 volt 20ma LED lights up just fine on two 1.5 volt batteries in series with no resistor required.

 

Mark. 

¡ uʍop ǝpısdn sı ǝɹnʇɐuƃıs ʎɯ 'dlǝɥ

  • Member since
    December 2006
  • From: Lancaster, PA
  • 512 posts
Posted by claymore1977 on Friday, May 31, 2013 11:07 AM

Just read through the thread and would like to re-iterate/clarify a few points.

1) An LED (or any diode for that matter) has a 'minimum' and 'maximum' voltage.  If your supply voltage is below minimum, the LED won't turn on.  If supply voltage is above maximum voltage, then you run the serious risk of causing the LED to fail.

2) LEDs in series is not such a great idea for few reasons.  

  a) Troubleshooting becomes harder, quite akin to the 'which light in the string of christmas lights is causing the whole string to not work' scenario.  

  b) the 'minimum voltages' are cumulative.  3 LEDs with a 'minimum' voltage of 1.8V each, when connected in series, will require a minimum of 5.4V for them to turn on.  And it will be all on or all off.  5.4V may not be an issue unless you are thinking of running them off of batteries or other lower voltage supplies.

3) 'Minimum' voltage is also often referred to as the 'Forward Bias Voltage'.  Aka, what voltage does it take to re-arrange the electons in the semi-conductor such that current can start flowing.

4) As much of a PITA it may be, the best design you can have will be 1 Resistor per LED.

5) There are circuits called "current limiters" that can solve a lot of LED problems.  What they do is prevent any more than X miliamps to be supplied to the LED.  They are, however, a bit more complex than a single resistor. :)

Just my $0.02.  Cheers!

Dave Loman

My site: The Rusty Spike

"It's a penny for your thoughts, but you have to put your 2 cents in.... hey, someone's making a penny!"

  • Member since
    October 2005
  • From: Detroit, Michigan
  • 2,284 posts
Posted by Soo Line fan on Friday, May 31, 2013 4:47 PM

Mark R.

This online calculator might help you ....

http://www.hobby-hour.com/electronics/ledcalc.php

Nice link, thanks Mark

Jim

  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Friday, May 31, 2013 5:40 PM

below is a curve indicating the voltage vs current relationship of a diode (LED).    The corresponding curve for a resistor, where I = V / R is simply a straight line from bottom left to top right and passing through the origin where both voltage and current equal zero.   Think for a moment about what both the resistor and diode curves mean.

the straight line of the resistor curve, I = V/R, means that current (I) increases proportionally as the voltage across the resistor increases.   If the voltage is negative, the current increases in the opposite direction.

the lower left portion of the diode curve indicates that as voltage across the diode is made more negative, very little current is passed until the breakdown voltage is exceeded, the diode fails, becoming an open circuit which means no curve flows.

the right had side of the curve means that as you increase the voltage across the diode, current increases exponentially, a small increase in voltage results in a large increase in current.   A typical diode (not and LED) is passing max current at ~0.7V.   If the current exceeds some threshold, the diode overheats, fails and becomes an open circuit.

If a resistor is put in series with the diode and the voltage is increased across both the diode and resistor, there is a balance between the current passed thru the diode, the voltage across the resistor due to this current and the remaining voltage across the diode.   When the voltage is small and the current is small, most of the voltage is across the diode.  As the voltage and hence current increases, more voltage is "dropped' across the resistor and the voltage across the diode increases very slowly.

this thread, from back in january describes a simple 20ma current regulator that will limit the current through a circuit, which as it describes, is very useful for LED circuits, especially with varying voltages or multiple LEDs.   This means that the Cl2N3 drops as much voltage as necessary (within limits) to maintain as much voltage as necessary across the diode so that it never passes more than 20 ma.

circuits like these maintain an equilibrium between voltage and current.  science is wonderful

greg - Philadelphia & Reading / Reading

  • Member since
    December 2006
  • From: Lancaster, PA
  • 512 posts
Posted by claymore1977 on Friday, May 31, 2013 6:53 PM
BOOM! Greg just dropped a semiconductor theory bomb on us! Damage report! Casualty assessment!

Dave Loman

My site: The Rusty Spike

"It's a penny for your thoughts, but you have to put your 2 cents in.... hey, someone's making a penny!"

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Saturday, June 1, 2013 11:38 AM

 You effectively cannot 'overvolt' an LED, not on forward voltage. It's the CURRENT. Those christmas light LEDs - they are runnign the LEDs on 120V. You just need a high enough value resistor such that the current does not exceed the LED rating. ANd I'm willing to bet those cheap christmas lighs run the LEDs at near or even slightly over maximum current for the highest brightness, since they don;t care if they last 10 years or only 3.

How many volts a given LED drops depends on the chemistry of the junction. Most white types of LEDs are around 3.5 volts, most color LEDs these dats are around 2.1 volts. Don't assume, get the specs on the LEDs you are using to make the resistor calculations, because makign the wrong assumption can lead to too much current, and the end of the LED.

LEDs do have a limit on reverse volage too, that it MUCH lower than a typical rectifier diode. Typical rectifiers can take (and block, since current only flows in one direction through a diode) 50, 100, 200, or even 400 volts (and some much more) without breaking down. LEDs are much mor emodest, generally about the same as their forward voltage rating - which is why the suggestion that if running LEDs on AC, or some othe rbipolar source like being wired across the track, there is a regular diode or another LED wired anti-parallel so that the reverse current flows through the other diode and does not try to be 'forced' through the LED, 2 lead bi-color LEDs are like this, that's why you see them as the track power indicator on DCC, With very low current limits (say a 1K resistor on a whote LED, limiting it to 9ma or less at 12V supply), it may be ok, but I still say over time it will reduce the life of the LED. In this case maybe 10 years instead of it lasting 20 with no reverse voltage.

          --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    July 2006
  • From: North Dakota
  • 9,592 posts
Posted by BroadwayLion on Saturday, June 1, 2013 3:20 PM

rrinker
LEDs do have a limit on reverse volage too, that it MUCH lower than a typical rectifier diode. Typical rectifiers can take (and block, since current only flows in one direction through a diode) 50, 100, 200, or even 400 volts (and some much more) without breaking down. LEDs are much mor emodest, generally about the same as their forward voltage rating -

In building signals LION puts them back to back with one resistor, reversing the voltage does no harm to the reversed LED because the current is going through the one that is lit.

ROAR

The Route of the Broadway Lion The Largest Subway Layout in North Dakota.

Here there be cats.                                LIONS with CAMERAS

  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Saturday, June 1, 2013 5:16 PM

rrinker
With very low current limits (say a 1K resistor on a whote LED, limiting it to 9ma or less at 12V supply), it may be ok, but I still say over time it will reduce the life of the LED. In this case maybe 10 years instead of it lasting 20 with no reverse voltage.

The Truth About LED Lifetimes.    BTW, 1 year is 8760 hours

greg - Philadelphia & Reading / Reading

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Saturday, June 1, 2013 6:08 PM

 Pretty much in line, 100K hours is over 11 years. ANd they are talking about high power LEDs, like the incandescent replacement LED 'bulbs', not a single white LED operating at say 9ma.

You will notice most of the repalcement LED bulbs have a large heat sink as part of the design, as mentioned in the article the heat accumulates in the junction and without adequate heat sinking would quickly destroy themselves. Put a good enough heat sink on an ordinary LED and you can pulse them at well over their rated current - just not continuously operate them. An ordinary T1 or T1 1/2 LED is too well insualted to conduct heat outside the die.

            --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    August 2010
  • 1 posts
Posted by mountainZ on Friday, June 7, 2013 2:24 PM

Rich,

     The forward voltage on LEDs generally varies by color from red to blue/white from 1.2V to 3.6V.  The advise to start with  a 1-2K ohm resistor and adjust the value to get the lighting intensity you want. When an LED is rated at 20mA, that is just a recommended value, it will easily work  down to 5mA.

    You mention powering from DCC, which is actually an AC voltage. LEDs are generally limited to about 4 V when the polarity is reversed. You should use a rectifier dioded such as a 1N4001 to prevent an over voltage in the reverse mode.

     Another option is to use a current limiting diode (CLD), They're pricey and can be hard to find, but they solve the issie of calculating voltage drops.

 

Mark

  • Member since
    September 2004
  • From: Dearborn Station
  • 24,281 posts
Posted by richhotrain on Friday, June 7, 2013 4:34 PM

mountainZ

Rich,

You mention powering from DCC, which is actually an AC voltage. LEDs are generally limited to about 4 V when the polarity is reversed. You should use a rectifier dioded such as a 1N4001 to prevent an over voltage in the reverse mode.

My original plan was to wire three LEDs in series with one resistor on a decoder to provide lighting for a locomotive.  But, I have abandoned this plan in favor of wiring in parallel with a resistor on each LED.

Rich

Alton Junction

  • Member since
    July 2006
  • From: North Dakota
  • 9,592 posts
Posted by BroadwayLion on Saturday, June 8, 2013 7:05 AM

richhotrain
My original plan was to wire three LEDs in series with one resistor on a decoder to provide lighting for a locomotive.  But, I have abandoned this plan in favor of wiring in parallel with a resistor on each LED.

GOOD PLAN! LION wired entire station with LEDs in series. It stopped working. LION pulled roof off of station and replaced all of the LEDS with new ones, and wired them with resistors. Him then glued roof back on station. Looks good, but there are three other stations with series lighting, and all of them are in trouble: the circuits will have to be rebuilt. This means taking the stations apart and re doing them.

ROAR

The Route of the Broadway Lion The Largest Subway Layout in North Dakota.

Here there be cats.                                LIONS with CAMERAS

  • Member since
    September 2004
  • From: Dearborn Station
  • 24,281 posts
Posted by richhotrain on Saturday, June 8, 2013 7:43 AM

Yeah, the original plan was to save space under the hood, but I wound up using 1/4 watt resistors and I shortened the legs of each resistor and saved considerable space.

Rich

Alton Junction

  • Member since
    July 2006
  • From: North Dakota
  • 9,592 posts
Posted by BroadwayLion on Saturday, June 8, 2013 9:49 AM

Yes, use 1/4 watt resistors. You can get 1/8 watt resistors, but they will cost you megabucks more.

LION uses 1/2 watt resistors for track automation, should be much higher except they are in use for only seconds at a time, our radio expert said that bigger ones were not necessary. If they get to hot they will catch fire. With 12 amp power supply, the breaker will never trip for one of these.

ROAR

The Route of the Broadway Lion The Largest Subway Layout in North Dakota.

Here there be cats.                                LIONS with CAMERAS

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Saturday, June 8, 2013 10:22 AM

 If space is at a premium and you have lots of lights, particularly if there are instances where you have dual lights that would be on at the same time, check out this handy board from Ulrich:

http://www.ulrichmodels.biz/servlet/the-597/LED-Lighting-Resistor-Board/Detail

Has inputs for 6 functions, each with its own resistor (already soldered to the board, so you don;t have to worry about soldering surface mount resistors) plus two of them are doubled up - two outputs in parallel with each other. You solder the decoder wires to one side of the board and solder the LEDs to the other side.

        --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

There are no community member online

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!