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!

Are resistors needed?

7649 views
31 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    October 2005
  • 1,047 posts
Posted by betamax on Tuesday, May 31, 2011 4:53 AM

All is good, except... when the junction heats up, it's characteristics change.  Inside a closed shell it will get hotter than it does in free air.

Another point is that doing it the conventional way with a series resistor is less likely to have problems, which will not frustrate people. Powering a LED off a 1.5V output is pushing it, unless there is someway to insure that there is current limiting available.  White LEDs have a much higher Vf, while being a lot more sensitive to current.  

While connecting an ammeter across a cell will not show a million amps, as the internal resistance will limit it, doing it can result in burns or explosion of the cell. The internal resistance is what results in a dead cell in time.

While your demonstration works, it isn't something most of us would recommend as a normal practice.  

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Monday, May 30, 2011 3:46 PM

richg1998

It appears some here need to do some research. I think I see opinions at times. The below link shows quite a lot about resistors, light bulbs and LED's. Get your test equipment out and do some experimenting.

http://members.optusnet.com.au/nswmn1/Lights_in_DCC.htm

Online LED resistor calculator. Many here do not realize there are many electronic/electrical calculatros online.

http://www.hebeiltd.com.cn/?p=zz.led.resistor.calculator

Various online calculators.

http://www.ifigure.com/engineer/electric/electric.htm

Rich

 

Good Timing.

Click on the first calculator.

In the calculator select source voltage of 3 volts

Then select voltage drop of 3 volts

Then enter 30miliamps for desired LED current.

Read the value of resistor needed.   It's zero

That means no resistor is needed.

While the output will sometimes creap up, the calculations are done based on the rated output of the circuit.

 

 

Springfield PA

  • Member since
    November 2002
  • From: Colorado
  • 4,075 posts
Posted by fwright on Monday, May 30, 2011 3:31 PM

Hamltnblue

Once again

It is a FACT that if you supply a 3 volt LED with a 3 volt power source it will Not blow.

Correct for a battery, not correct for a power source.

As to batteries. I challenge anyone here to place an Amp Meter on a single AAA, AA., or D battery. You will  get more than an amp out.  Surely it will be much higher than what an LED will draw.

Back to the original idea though. If a power supply or LED output is putting out a set voltage and that voltage is at or below the LED's rating you do not need a resistor.  Actually using a resistor in that scenario could lower the voltage below the operating range of the LED or make it dim.

Correct again, with my emphasis.  The problem is that a 1.5 volt (or any other nominally rated) power supply will often put out more than 1.5 volts under very light loads.  Measure your wall wart output voltage at various loads, and see how much it varies from the nominal rating.  Or measure a DC power pack full voltage output with no load, and with load.  Your decoder output voltage is going to vary slighty with the track voltage and with the load - there is not enough space on the decoder for better than 5% output regulation.

Your computer has a very highly regulated power supply with many components that a decoder has no hope of fitting.  Even so, there are allowable variations in the voltage output of the power supply.

A battery by itself will not exceed its full voltage rating because it is chemically limited as to how much voltage it can put out.

You are counting on the decoder output to never exceed its rated 1.5 volt nominal rating under any conditions.  I'll bet against you in a heart beat.  I like my odds of winning.

Fred W

  • Member since
    October 2006
  • From: Western, MA
  • 8,571 posts
Posted by richg1998 on Monday, May 30, 2011 2:56 PM

Yes, I do realize some just want to carry the conversation along.

Rich

If you ever fall over in public, pick yourself up and say “sorry it’s been a while since I inhabited a body.” And just walk away.

  • Member since
    October 2006
  • From: Western, MA
  • 8,571 posts
Posted by richg1998 on Monday, May 30, 2011 2:55 PM

It appears some here need to do some research. I think I see opinions at times. The below link shows quite a lot about resistors, light bulbs and LED's. Get your test equipment out and do some experimenting.

http://members.optusnet.com.au/nswmn1/Lights_in_DCC.htm

Online LED resistor calculator. Many here do not realize there are many electronic/electrical calculatros online.

http://www.hebeiltd.com.cn/?p=zz.led.resistor.calculator

Various online calculators.

http://www.ifigure.com/engineer/electric/electric.htm

Rich

 

If you ever fall over in public, pick yourself up and say “sorry it’s been a while since I inhabited a body.” And just walk away.

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Monday, May 30, 2011 2:33 PM

Once again

It is a FACT that if you supply a 3 volt LED with a 3 volt power source it will Not blow.

As to batteries. I challenge anyone here to place an Amp Meter on a single AAA, AA., or D battery. You will  get more than an amp out.  Surely it will be much higher than what an LED will draw.

Back to the original idea though. If a power supply or LED output is putting out a set voltage and that voltage is at or below the LED's rating you do not need a resistor.  Actually using a resistor in that scenario could lower the voltage below the operating range of the LED or make it dim.

Showing the battery was only to answer those who somehow thought that an LED without a resistor in series with the 3 volt supply would somehow run away and blow.

The million amp availabilty noted earlier was simply to use an extreme to say it doesn't matter what the Amp Capacity is. It's the voltage that matters in this case.

Springfield PA

  • Member since
    November 2002
  • From: Colorado
  • 4,075 posts
Posted by fwright on Monday, May 30, 2011 2:29 PM

I forgot to mention that a light bulb is also a decidedly non-linear device.  But since a bulb is usually operated in one of 2 states - cold and hot (emitting light) - and both states are stable, we use separate resistances for the 2 states when applying Ohm's Law.

Tying this back to the OP's question regarding lights and decoder outputs - a battery will never exceed its fully charged output voltage, no matter how light the load.  Any other type of power supply does not have this characteristic unless very heavily regulated.  The output voltage tends to creep up above the nominal rating as the load approaches zero, or there are short-term voltages above rating during start-up or turn-on.

As Randy has pointed out, these over-voltages are deadly for an LED being operated near its forward voltage rating.  Which is why a battery is such an ideal power source for LEDs - there is no worry about exceeding the voltage rating of the LED.  Rapido (and others) use batteries -without charging circuits - for LED lighting kits for passenger cars.  The minute a battery charging circuit is introduced, the potential for over-volting the LED has been introduced.

Which is why the 12 volt light output with current limiting resistors is recommended for LEDs.  Running 1.5V LEDs off the 1.5 volt decoder output without current limiting exposes the LED to over-volting.  1.5 volt light bulbs, on the other hand, generally impose a higher load from start on the decoder output which means the voltage regulator works better, and the bulbs can better absorb any momentary over-voltages.  Unless the decoder manufacturer specifically recommends using his 1.5V output for LEDs, I would use the 12V output with current limiting resistors.  Reserve the 1.5V output for light bulbs unless the decoder manufacturer advises differently.

my thoughts, your choices

Fred W

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Monday, May 30, 2011 2:23 PM

 Well there's 'nominal voltage' the consumer expects, and 'nominal voltage' the real nominal voltage defined by the battery chemistry. 1.2 volts is the nominal voltage of a NicCad cell, it's just the non-scientific consumer has been 'trained' that "a battery" is 1.5 volts.

                  --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
    November 2002
  • From: Colorado
  • 4,075 posts
Posted by fwright on Monday, May 30, 2011 1:48 PM

Hamltnblue

 

 

This is basic Ohms law stuff and I can't believe soo many people don't understand it.

If you have a single LED rated at 3 volts and 30 miliamps and you supply it with 3 volts at 1 million amps you get the same light output and no burnout.

The issue is not Ohm's Law, but the abuse of Ohm's Law with non-linear devices.  By non-linear device, I mean a component that significantly changes its internal resistance with current or voltage applied.  In a model railroad context, LEDs, motors, and transistors are all non-linear devices.

As far as the LED/battery scenario goes, both of you are right.  As long as the voltage supplied by the battery is below the reference forward voltage of the diode, the diode will draw only the needed current to emit the light.  The LED is operating in its linear range (a pretty small voltage range).  This principle is used in all of the LED flashlights, headlamps, lip lamps, and other LED lights I have used in aviation.  The batteries are set up for nominal 1.5 or 3 volt operation depending on the LEDs used, never more.  So yes, in most cases you can run an LED directly off a 1.5V nominal battery.

Another fact to note:  very few modern battery formulations serve up the full nominal voltage.  Nicads are among the worst offenders - a fully charged nicad cell only puts out 1.2 volts.  And of course, the output voltage of a battery does reduce over time and with load.  All these factors give a further safe margin for use as an LED power source.

just my experiences

Fred W

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Monday, May 30, 2011 12:13 PM

 Just keep walking that thin line - you're ok if you do not even SLIGHLTLY exceed Vf on the LED - since there's no excess voltage to bleed off. The minute you exceed the Vf of the LED with the power supply, you'd better have some current limiter in place or the LED will avalanche right past breakdown and go pop. At least 9V is best - then you can actually hear the pop, usually.

 Don't confuse people, you can't 'supply' 1 million amps, you can have a power source capable of sourcing  a million amps but that doesn't mean any attached circuit will actually take a million amps. That's something even more people are confused over, based on how many people seem t think that with a 5 amp DCC booster, every loco has 5 amps running through it.

 I'm not disagreeing with you - I paid a lot of money to learn this stuff. Discharge characteristics vary greatly between battery chemistries, alkaline vs NiCad vs NiMH vs Lithium Polymer, etc. Alkalines are some of the most limited by internatl resistence - the tongue test is good to see if there's any juice left. A niCad though can deliver a LOT of current VERY quickly - so they are better for those things that need short bursts at high amperage. Not always a good idea to touch your tongue to one of those. A good example is a digital camera. Ones that can use standard batteries, see how many shots you can get with a set of ordinary alkaline batteries vs a fully charged set of NiCad or NiMH batteries. Not only to the others outlast the alkalines, they can then be recharged and used over. Double bonus.

                         -_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
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Sunday, May 29, 2011 11:54 PM

rrinker

 WHat type of batteries were you using? It DOES matter - different battery chemistries have different internal resistences and that limits just how fast you cna actually draw the power from them. 1200mAH does not mean the batteries can supply 1200ma.

 It works because the voltage is below the LED's actually Vf and the battery internal resistence keeps it from sucking over the maximum current. Pick a 9V battery and try it - instant LED pop even though the 9V will have LESS mAH than the AA's.

              --Randy

 

I just showed 3 different batteries. All alkalines including D cells, and you don't think a D cell can feed over the rated current of a single LED?

If an LED is rated at 3VDC and you supply it with 3VDC it will draw it's rated current if the source is capable.

Oh and trust that even a single AA battery can easily supply a full amp of power if the resistance is low enough.  Put your amp meter across one and see what it tells you.

We're talking about a single LED for crying out loud with a pair of AA, AAA and D batteries.

My 1979 LED watch did it with a single battery.

Oh and look at the pics again   There are engergizer and Duracell batteries shown. 

This is basic Ohms law stuff and I can't believe soo many people don't understand it.

If you have a single LED rated at 3 volts and 30 miliamps and you supply it with 3 volts at 1 million amps you get the same light output and no burnout.

 

Springfield PA

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Sunday, May 29, 2011 10:05 PM

 WHat type of batteries were you using? It DOES matter - different battery chemistries have different internal resistences and that limits just how fast you cna actually draw the power from them. 1200mAH does not mean the batteries can supply 1200ma.

 It works because the voltage is below the LED's actually Vf and the battery internal resistence keeps it from sucking over the maximum current. Pick a 9V battery and try it - instant LED pop even though the 9V will have LESS mAH than the AA's.

              --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
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Sunday, May 29, 2011 9:16 PM

I'm bored tonight so I'll go ahead and do a show and tell.  We can do an ohms law explaination later if anyone is interested. By the way I'm just having fun with it and not at anyone's expense.

First up is a single LED with a pair of AAA batteries.  These babies have a 1200mah (1.2 amp hour) capacity.  Surely they'll blow the 30 miliamp led without current limiting resistors.

Neat Huh. Note the brightness and it's not smoking or burning up.

Next up the AA battery. Twice the size of the AAA and carries a 2700mah (2.7 amp hour) punch. Close your eyes if you can't bear to watch.

Wow It survived, and guess what?  Same brightness. Applause in the background.

Now we'll go even further. Whoooooooooooo.  A pair of D's with a 12000 (12 amp hour) capacity.  Those dirty minded people thinking Double D's give it up. 

Whooo Ho Ho.  The LED is the same brightness and still burning.  Amazing.

But Wait. He goes even further.  Adds a AA battery in series with the Double D's for a total of 4.5 volts.

And the crowd goes wild.  The led is a little brighter but still survives.  and with NOOOOOO resistors.

Note: the led will eventually fail at this voltage.

Randy usually chimes in about now but the fact is NO you don't need resistors if the supplied voltage is the same or a little lower than the rated voltage of the LED.

 

 

Springfield PA

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Sunday, May 29, 2011 8:43 PM

Current is part of the PIE in OHMS law.

If yo have 1000 amps available at 3 volts and you supply a 3 volt 30 miliamp LED you get the same light as if you have 3 volts and 50 miliamps available. 

Dropping resistors are used to bring the voltage of a circuit down to divide the supplied voltage with the LED so that the LED has only it's rated voltage or lower available.

If you have a 3 volt LED and a 12 volt supply, you have to add the resistor required to remove or drop 9 volts.  If the voltage is already at the LED's rated level there is no resistor needed.

Also An LED Will light at a voltage lower than what it is rated at.  The Result will simply be a dimmer LED.

This is why we often recommend a 1K resistor instead of the 680 ohm value that works mathmatically.

The LED is a little dimmer but it will last longer since it won't be running at maximum brightness.

Obviously there are several here who have a basic understanding of the subject  but dont understand Ohms law.  I can follow up with a couple of pictures if needed of an LED hooked directly to different sized batteries with no resistor and having the same brightness and not burning out.

I can use a pair of AA batteries directly to a 3 volt, 30ma LED with no resistor.  The AA batteries have a 2 amp hour rating and can produce over an amp when shorted. (over 1000 miliamp)

I can also go further and show the same LED with 3 batteries in series. Neither of them blowing the LED.

It would be magic for some but no LED's would be harmed in the test I promise Smile, Wink & GrinLaugh

Springfield PA

  • Member since
    October 2005
  • 1,047 posts
Posted by betamax on Sunday, May 29, 2011 3:24 PM

LEDs don't work if the supply voltage is less than Vfwd.  They are current devices. A dry cell can also supply more current than needed to fry the junction.  

 

A 4.5V battery is a lot more than the typical Vfwd of an LED.  There is more to your flashlight than meets the eye.  

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Saturday, May 28, 2011 10:34 PM

Resistors are only used to drop the voltage to the rating of the device.  By doing so you limit the current.  If the supply voltage is already at or lower than the LED's rating you do not need a resisitor. The led will only draw it's rated current at it's rated voltage.  I have an LED flashlight.  Open it up and guess what. No resistors. It has 3 double A batteries that feed directly into the LED.

Springfield PA

  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Saturday, May 28, 2011 6:14 PM

to expand on what betamax had already said.

a lamp is just a piece of wire.  the more voltage, the hotter it get producing more light and increasing its resistance to limit the current, within limits.  Hewlett-Packard got started using a lamp to regulate current to make a very stable oscillator.

an LED is a diode, a semiconductor, a device normally used to restrict current flow in one direction (think A.C. to D.C.).  Once the forward voltage is exceeded, the current increase exponentially (1.5 is not 1.4).  Without a resistor, the current limit can easily be exceeded and it will burn out.

The forward voltage drop across many LEDs was typically 1.4 volts.   The resistor value is the source voltage - 1.4 divided by the desired current (2.6k for 12v and 4ma).  

I'm not familar with the types of LEDs available today, but i'm guessing that there are probably LED packs that operate at higher voltages.

greg - Philadelphia & Reading / Reading

  • Member since
    October 2005
  • 1,047 posts
Posted by betamax on Saturday, May 28, 2011 5:15 PM

Hamltnblue

 

 

 

Huh? ConfusedConfused

You don't use a resistor on an LED when there is a 1.5 volt output.

How the device produces light has nothing to do with the voltage that is feeding it. 

 

It has a lot to do with it.  LED means Light Emitting Diode.  When forward biased, it is almost a short, with a small amount of voltage drop across the junction.  They are current operated, and without the series resistor to limit the current (and provide a voltage drop) they burn out.  Really quickly.

An incandescent lamp, like used in many household fixtures, has 120V across it at all times while operating.  A fluorescent lamp requires a lot of voltage to strike the arc, and once the arc is going, the ballast limits the voltage and current across the lamp.  Ohms Law says they are all related. The difference is one increases in resistance when lit, the other decreases. An LED is a much different animal. It will not tolerate excessive voltage or current.

  • Member since
    November 2005
  • From: Phoenix, AZ
  • 693 posts
Posted by woodone on Saturday, May 28, 2011 2:47 PM

 I made an error in my last post ( Some LED's will light up with 1.5 volts). should have said - Most of the LED's will not light up with 1.5 volts. So with a 1.5 volt output you will never get  most LED's to work.

 

 

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Saturday, May 28, 2011 2:44 PM

Here's a copy of the manual.  Note it say that when using LED's to use the 12 volt wiring. This would require resistors.  The reason they say this is because many LED's require more thant 1.5 volts for full brightness.  If you use the 12 volt then you would use a 1k resistor.  You could go down to 680 ohms or a little lower but the LED will last much longer with 1K.

http://www.mrdccu.com/graphics/jpg/TSU-AT1000Instructions.jpg

Springfield PA

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Saturday, May 28, 2011 2:28 PM

You don't need anything to limit current.  If a device is listed for say 1.5 volts and 20 miliamps, and you apply 1.5 volts, it will draw 20ma.

By the way, I installed one of he GN1000's and it put out 1.5 volts as advertised. No resistors needed.

If you're nervous about doing it, you won't hurt anything by trying it with a resistor.  You will probably just get little or no light.

 

Springfield PA

  • Member since
    September 2007
  • From: Charlotte, NC
  • 6,099 posts
Posted by Phoebe Vet on Saturday, May 28, 2011 1:03 PM

Yes they will, but there is nothing to limit the current.  The 1.5V output of a circuit board may have the current limited on the board, but a battery does not..

There are some LEDs that have a resistor built in.

Dave

Lackawanna Route of the Phoebe Snow

  • Member since
    November 2005
  • From: Phoenix, AZ
  • 693 posts
Posted by woodone on Saturday, May 28, 2011 12:58 PM

Well that would depend on the LED would it not? Some LED's will light up with 1.5 volts.

  • Member since
    September 2007
  • From: Charlotte, NC
  • 6,099 posts
Posted by Phoebe Vet on Saturday, May 28, 2011 12:33 PM

Go ahead, try it.

Dave

Lackawanna Route of the Phoebe Snow

  • Member since
    May 2008
  • 4,612 posts
Posted by Hamltnblue on Saturday, May 28, 2011 12:07 PM

Phoebe Vet

 tomikawaTT:

The 1.5V bulb IS a resistor.  OTOH, an LED needs a resistor to limit the current flow.  It's easy to confuse the requirements, since both the bulb and the LED produce light - but in entirely different ways.

Chuck (Modeling Central Japan in September, 1964)

 

That is true.  Put a 1.5 V LED across a 1.5 volt battery and watch how fast it burns out.

Huh? ConfusedConfused

You don't use a resistor on an LED when there is a 1.5 volt output.

How the device produces light has nothing to do with the voltage that is feeding it. 

Springfield PA

  • Member since
    September 2007
  • From: Charlotte, NC
  • 6,099 posts
Posted by Phoebe Vet on Saturday, May 28, 2011 11:04 AM

tomikawaTT

The 1.5V bulb IS a resistor.  OTOH, an LED needs a resistor to limit the current flow.  It's easy to confuse the requirements, since both the bulb and the LED produce light - but in entirely different ways.

Chuck (Modeling Central Japan in September, 1964)

That is true.  Put a 1.5 V LED across a 1.5 volt battery and watch how fast it burns out.

Dave

Lackawanna Route of the Phoebe Snow

  • Member since
    February 2005
  • From: Southwest US
  • 12,914 posts
Posted by tomikawaTT on Saturday, May 28, 2011 10:48 AM

The 1.5V bulb IS a resistor.  OTOH, an LED needs a resistor to limit the current flow.  It's easy to confuse the requirements, since both the bulb and the LED produce light - but in entirely different ways.

Chuck (Modeling Central Japan in September, 1964)

  • Member since
    November 2005
  • From: Phoenix, AZ
  • 693 posts
Posted by woodone on Saturday, May 28, 2011 9:47 AM

The AT ( TSU-AT 1000 ) version of the Tsunami decoders do have a pad that provides a 1.5 volt output.

Not sure it is a resistor on board.  I think the AT has a voltage regulator for the 1.5 volt output. 

Ar any rate,  if you use the 1.5 pad for your 1.5 bulbs you will not need a resistor.

  • Member since
    October 2006
  • From: Western, MA
  • 8,571 posts
Posted by richg1998 on Saturday, May 28, 2011 9:04 AM

richhotrain

I don't know anything about that particular decoder, but when it comes to 1.5 volt incandescent bulbs, if a resistor is not required it is because there already is a resistor built into the decoder.

Rich

Do not "assume" when it comes to DCC installs. There are "Gotchas". Those who smoke a LED or decoder usually do assume. The different decoder manufacture sites have decoder manuals and the decoders come with instructions. With the Internet, you can find the manuals easily .Ignorance is not bliss.

A digital multimeter to check the install can be had for about $5.00. I have three of them and cheap to replace if I misuse it.

Rich

If you ever fall over in public, pick yourself up and say “sorry it’s been a while since I inhabited a body.” And just walk away.

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!