hobo9941 Using an LED on AC should not damage it if the proper resistor is used. If the LED sees higher than rated reverse voltage, if the proper resistor is used, no current will flow through the LED, and there should be no damage. Excess current flow is what burns out an LED.
Using an LED on AC should not damage it if the proper resistor is used. If the LED sees higher than rated reverse voltage, if the proper resistor is used, no current will flow through the LED, and there should be no damage. Excess current flow is what burns out an LED.
That's the key. An LED with the proper resistor for 12V DC, when operated on 12V AC, sees reverse voltage above the rated reverse voltage. Low current, because of the resistor, but still over the rated 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.
iwander And I do not want to take the risks that several of the other repliers have taken. No small fires or exploding LEDs for me, thank you.
And I do not want to take the risks that several of the other repliers have taken. No small fires or exploding LEDs for me, thank you.
Eh? You should be around when a capacitor lets go. ☺
The Route of the Broadway Lion The Largest Subway Layout in North Dakota.
Here there be cats. LIONS with CAMERAS
Thanks for caring about my health...but...the additional wire that I'll be adding is excess wire from old Christmas tree sets. The rest of the wiring under the table is two tone colors. I never work under the table. All wiring so far is done above the table and then dripped into an opening and hooked onto the underside/front edge of the layout. The lighting will be plugged into a surge protector holding several other AC rope lights that light up the background. They are only turned on when quests are present and I'm running the trains.
Well if the LED is connected to a decoder it eliminates those problems - but an LED across the track gets AC and we're right back where we started. There's a reason it's common to use bicolor LEDs as track power indicators for DCC - because they are composed of a pair of LEDs wired back to back (antiparallel) so during the reverse part of the waveform, the 'backwards' LED only sees the forward votlage of the other one. Plus if the DCC system supports zero stretching to run an analog loco, the color of the bicolor LED will shift to more red or more green depending on the direction.
CSX Robert Mark R.: ...running an LED on AC should in theory destroy an LED, when in practical use, I've shown it doesn't ! (?).. All you've shown is that In this one instance, it hasn't, yet. You may not have left them powered long enough, so they may still blow. Also, the particular LED's you are using may have a relatively high tolerance for reverse voltage. The thing about using electronics beyond their ratings is it does not always do immediate damage. By repeatedly exposing LED's to reverse voltages, you may do no damage if you are below what the LED can withstand, or, depending on several variables, you may shorten the of the LED from slightly to drastically. I remember reading on one of the forums where someone was replacing a bulb headlight with an LED in a locomotive that did not have a rear headlight. He didn't know why his LED's kept dying because they would work fine for a while and then quite. He did not have any reverse voltage protection for the LED and when he added that he quit killing LED's. He showed that sometimes in practical use not protecting the LED from reverse voltage does indeed kill it.
Mark R.: ...running an LED on AC should in theory destroy an LED, when in practical use, I've shown it doesn't ! (?)..
...running an LED on AC should in theory destroy an LED, when in practical use, I've shown it doesn't ! (?)..
All you've shown is that In this one instance, it hasn't, yet. You may not have left them powered long enough, so they may still blow. Also, the particular LED's you are using may have a relatively high tolerance for reverse voltage. The thing about using electronics beyond their ratings is it does not always do immediate damage. By repeatedly exposing LED's to reverse voltages, you may do no damage if you are below what the LED can withstand, or, depending on several variables, you may shorten the of the LED from slightly to drastically.
I remember reading on one of the forums where someone was replacing a bulb headlight with an LED in a locomotive that did not have a rear headlight. He didn't know why his LED's kept dying because they would work fine for a while and then quite. He did not have any reverse voltage protection for the LED and when he added that he quit killing LED's. He showed that sometimes in practical use not protecting the LED from reverse voltage does indeed kill it.
Since this thread has started, I've been doing some research on the use of LEDs on AC power and information on reverse voltages. Apparently this can be done provided a diode is placed across the leads in parallel to the LED but oriented opposite to the LED. This diode will protect the LED from reverse voltage .... much the same as your description of having to use a front and rear LED. With just the front LED, the diode is used in place of the rear light.
Having not worked on much DC stuff in many years, I guess I've forgotten more than I learned in that regard - DCC avoids most all those concerns.
Mark.
¡ uʍop ǝpısdn sı ǝɹnʇɐuƃıs ʎɯ 'dlǝɥ
BroadwayLion The LION does use two LEDs in his signals, so one is lit and the other is dark according to polarity. Their is only one resistor since only one lamp will be lit at a time. So the LION'S question is: If the lighted LED carry the current, then the dark diode carries nothing, and they should work in this configuration, should they not. That is to say "The *do* *work* according to my plan, but would they be more at risk or somewhat less so since an alternate path is given to the current. The same principal applies to head lamps and marker lamps. Train moves forward headlamps are lit, train moves backwards and the markers are lit. ROAR
The LION does use two LEDs in his signals, so one is lit and the other is dark according to polarity. Their is only one resistor since only one lamp will be lit at a time.
So the LION'S question is: If the lighted LED carry the current, then the dark diode carries nothing, and they should work in this configuration, should they not.
That is to say "The *do* *work* according to my plan, but would they be more at risk or somewhat less so since an alternate path is given to the current.
The same principal applies to head lamps and marker lamps. Train moves forward headlamps are lit, train moves backwards and the markers are lit.
ROAR
This has already been covered twice in this thread, but when you have two LED's wired in parallel but opposite polarity, the maximum reverse voltage each LED can see is the forward voltage drop of the other LED, usually well within the reverse voltage rating of the LED.
Mark R. ...running an LED on AC should in theory destroy an LED, when in practical use, I've shown it doesn't ! (?)..
I think you are making a big mistake safety wise to do this .There is a lot of wiring under a layout to power track ,switch machines, signals ,lights etc. There are many different types of small gauge colored wires . You may think you will remember whats what but under the bench work with lousy light and a long night of trouble shooting you may make a mistake or have a friend helping that is not so familiar with the your wiring scheme. It would be too easy to misidentify this wire as low voltage accessory wire .The Christmas tree lights start with 110 VOLT AC! I am sure there is no electrical code that would allow this to be done .If the there is any 110 Volt wiring around or under a layout it should be easily identified and marked be it Romex ,BX or in conduit this is not stuff to play a guessing game with. If you are asking if you can do what you are proposing I would guess you are not sure what you are doing. NO TIME FOR GUESS WORK! If you insist on using Christmas lights use the the 10 light battery powered sets. If you want to use the leds from a 110 volt set cut them apart and use then with proper low voltage power and resistors.It would be far better to burn a few Leds on the low voltage that to RISK BURNS OR SHOCK trying to use the lhe 110 AC sets. I am Not an electrician but I know that 110 volt house current is DANGEROUS STUFF! More so when you are not sure what you are doing.
I'm not trying to say it's a great idea - I understand it's not a good idea by common practice. I'm just trying to understand that when according to the "rules" of electronics, running an LED on AC should in theory destroy an LED, when in practical use, I've shown it doesn't ! (?)
Maybe it's like "you can't run a red light" (good practice), where-as you actually CAN run a red light (poor practice - it's a gamble if you get hit or not) .
Mark R. We're vering WAY of course with the original topic of LEDs on AC now. The concern was that the inverse "polarity" of an AC sine wave would be detrimental to LEDs. This inverse polarity would (should) be no different that wiring an LED backwards to a DC source - yet there is obviously no big concern to damage an LED if wired backwards, it just won't work. Which brings me right back to the appropriate resistor comment. The resistor protects the LED in forward bias, doesn't it protect it in reverse bias as well considering the reverse voltage tolerance is a couple volts higher than the forward voltage tolerance ? Yes, in reverse base voltage the LED itself is taken out of the equasion for voltage drop, but isn't that why the reverse base voltage is higher from a manufacturing standpoint ?
We're vering WAY of course with the original topic of LEDs on AC now. The concern was that the inverse "polarity" of an AC sine wave would be detrimental to LEDs. This inverse polarity would (should) be no different that wiring an LED backwards to a DC source - yet there is obviously no big concern to damage an LED if wired backwards, it just won't work.
Which brings me right back to the appropriate resistor comment. The resistor protects the LED in forward bias, doesn't it protect it in reverse bias as well considering the reverse voltage tolerance is a couple volts higher than the forward voltage tolerance ? Yes, in reverse base voltage the LED itself is taken out of the equasion for voltage drop, but isn't that why the reverse base voltage is higher from a manufacturing standpoint ?
I'm not sure why you are determined to show that deliberately exceeding the reverse voltage rating of an LED is a great idea (DC or AC doesn't matter). You make your choice; I'll stay within the guidelines and ratings when designing circuits.
As far as the resistor not protecting a reversed-bias diode, that's fairly easy to understand. In a circuit with no current flowing, the full voltage is present across any break in the circuit. The reversed-bias diode is both the break in the circuit, and the reason why no current is flowing. Without current flowing through the resistor, there is no voltage drop across the resistor. The full reverse voltage is passed through to the diode.
The variation of voltage drop across a resistor in proportion to the current flowing through it is why potentiometers (rheostats) became such a poor choice for controlling motors with widely varying current draws.
Many things are possible; many fewer things are wise choices.
Fred W
Hooking an LED to a battery backwards isn't really good for it either. hort term, the amount that the LED's life is shortened is miniscule, and you and your grnadkids STILL won't live long enough to see it finally die. But you generally don't leav eit liek that, you hook it up, no light - oh yeah, backwards - so you fix it and now it works. Running on AC, it will get reverse voltage constantly. There may be no real noticeable difference IF the voltage is down near the LED's forward voltage anyway - when it's low enough ot only need a small resistor. LEDs ARE very different from a silicon rectifier diode. COmmon rectifier dioides have PIV (peak inverse voltage) ratings of 100, 200, and 400 volts. Those are the common small ones like 1N4001, 1N4002, and 1N4004 (the last digit is the PIC in hundreds - NOT amps as commonly inferred). LED PIVs are typically around 5V.
If you want your LEDs to live long and healthy lives, use DC, and limit the current to well under the LED's rating, which is a maximum.
Just for giggles - for the last number of hours, I've had three Leds (white with 1000 ohm resistor) attached to a 12vac source, a 12vdc forward bias source and a 12vdc reversed bias source. The two on the AC and forward bias DC are still working fine. When I flip the polarity on the reversed biased LED, it still works fine as well .... so where is this big concern ?
Is it a matter of reducing the life of the LED over the long term ? So instead of getting 100,000 hours, I'm only going to get 50,000 ? The signals on my layout are still showing their AC yellow indication after 15 years, and for how long they actually need to display yellow, I'm sure they will still outlive me.
Mark R. Do all steam engines that have a directional LED headlight ALWAYS have a corresponding rear LED light as well ? (not a steam fan, so don't know) I'm trying to wrap my head around why reverse voltage is such a concern when the LED has a proper resistor attached for the given voltage. You always read .... " If the LED doesn't light, it may be wired backwards - it won't hurt it, it just won't light" - for doing DCC installations. If this reverse voltage is such a concern, why aren't there warnings saying DO NOT WIRE THE LED BACKWARDS ! I know we're flip-flopping between bi-polar DC and AC sources here, but the theory is basically the same isn't it ? Mark.
Do all steam engines that have a directional LED headlight ALWAYS have a corresponding rear LED light as well ? (not a steam fan, so don't know) I'm trying to wrap my head around why reverse voltage is such a concern when the LED has a proper resistor attached for the given voltage.
You always read .... " If the LED doesn't light, it may be wired backwards - it won't hurt it, it just won't light" - for doing DCC installations. If this reverse voltage is such a concern, why aren't there warnings saying DO NOT WIRE THE LED BACKWARDS !
I know we're flip-flopping between bi-polar DC and AC sources here, but the theory is basically the same isn't it ?
The bi polar or pulsed DC is not seen by the LED, it is already conditioned by the decoder. LEDs list their max reverse voltage since it can be a problem.
I've run into similar situations when using diodes to generate a voltage drop for lighting cabeese or psgr cars. Normally you can use diodes for constant lighting because the motor is drawing enough amps. Two diodes will generate enough voltage drop to light a 1.5V bulb, if you have enough draw. So I used to use 1156 or similar bulbs to replace the motor, but cover that bulb of course. Ultimately I didn't like the circuit but the idea is the same. With 4 diodes you make a reverse parallal circuit. I tried resistors by themselves to just experiment and that's when I learned they don't work without current.
At least that's how I've understood it for a while now, hope it helps.
Richard
Mark R. Ok - the logic sounds reasonable .... but under what circumstances will this happen ? I know numerous people who run AC into the bi-color LEDs (red / green) to achieve a simulated yellow indicator - and have been doing so for years with no adverse effects. Is this a case where practice trumps theory ? .... "it shouldn't work - but it does !" Mark.
Ok - the logic sounds reasonable .... but under what circumstances will this happen ?
I know numerous people who run AC into the bi-color LEDs (red / green) to achieve a simulated yellow indicator - and have been doing so for years with no adverse effects. Is this a case where practice trumps theory ? .... "it shouldn't work - but it does !"
Actually, Randy already touched on this several posts back. When you have two LED's wired in parallel but opposite polarity(which is what the two lead bi-color LED's are), each LED will only see a reverse voltage as much as the other LED's forward voltage. You have the same situation with LED headlights on non DCC locomotives, they are wired in "reverse parallel" so that one lights in one direction and the other lights in the other direction. When running opposite the direction of one LED, that LED will only see the forward voltage drop of the other LED. Remove one of the LED's and run the locomotive in the direction of the removed LED and you will likely blow the other LED.
Thank you everyone who responded so far. It was a little overwhelming and sometimes TMI. I have 43 buildings for my downtown and the 60 lights will be distributed to most of them. I will cut one wire at a time and reatach them before going on to the next cut. The total will remain 60 and I am not afraid of 110 or120 AC under and on top of my train board. No more than the extension cords and power pack cords under the table. This won't take place for awhile but I'll let you know what happens. Thanks again to everyone.
Mark R. .... But if the resistor is safely keeping the forward operating voltage at 3.4 volts, does not that same resistor also keep the reverse voltage at the same level ? In other words, the resistor keeps the upper ac sine wave (positive bias) at 3.4 volts, wouldn't it also keep the lower sine (negative bias) at the same 3.4 volts ?
.... But if the resistor is safely keeping the forward operating voltage at 3.4 volts, does not that same resistor also keep the reverse voltage at the same level ? In other words, the resistor keeps the upper ac sine wave (positive bias) at 3.4 volts, wouldn't it also keep the lower sine (negative bias) at the same 3.4 volts ?
No, the voltage drop of the resistor depends on the current through the resistor. No current, no voltage drop. The full reverse voltage then falls across the diode.
That being said, the reverse voltage rating of any given LED is usually around a couple volts higher than the forward voltage. Based on that, wouldn't an LED running on AC be destroyed on the forward bias before the reverse bias is even attained ? Mark.
A diode is a non-linear device. In the forward direction, there is a small almost linear operating region, followed by a near constant voltage drop across the diode, regardless of current. Which is why LEDs need resistors to limit current whenever supplied voltage is equal to or higher than the forward voltage drop.
When reverse biased, the junction inside the diode fails when the reverse bias voltage exceeds the rated reverse voltage. And the diode is no longer a diode.
hope this helps
That being said, the reverse voltage rating of any given LED is usually around a couple volts higher than the forward voltage. Based on that, wouldn't an LED running on AC be destroyed on the forward bias before the reverse bias is even attained ?
hobo9941 Actually, running LEDs on AC should not harm them, so long as the voltage is not excessive...
Actually, running LEDs on AC should not harm them, so long as the voltage is not excessive...
This is true, emphasizing the "so long as the voltage is not excessive" part(more on that later).
hobo9941 After all, they are.................ta da, DIODES, used to rectify AC into DC. They are light emitting diodes, but still diodes, just like those in a rectifier. Just make sure you use the right resistor.
After all, they are.................ta da, DIODES, used to rectify AC into DC. They are light emitting diodes, but still diodes, just like those in a rectifier. Just make sure you use the right resistor.
Unfortunately, I often see this advice given on forums. As Joe mentioned, LED's are different than rectifier diodes. Here is the data sheet for some common rectifier diodes: http://www.fairchildsemi.com/ds/1N/1N4001.pdf. Here is the data sheet for a random white LED: http://media.digikey.com/pdf/Data%20Sheets/Lite-On%20PDFs/LTW-2S3D7.pdf. The rectifier diodes in that data sheet have a reverse voltage rating from 50 to 1000 volts, where the LED has a reverse voltage rating of 5 volts. If you try to power that LED with 12 volts AC, then you will likely kill it.
Many LED's will actually withstand a higher voltage than rated, especially the red, green, amber, and yellow ones; however, white LED's tend to be much less tolerant of reverse voltages. Also note that excessive reverse voltages do not always kill the LED quickly, but will often drastically shorten the LED's life.
hobo9941 Heh heh heh. If you run LEDs on AC, you don't have to worry about polarity. Actually, running LEDs on AC should not harm them, so long as the voltage is not excessive. After all, they are.................ta da, DIODES, used to rectify AC into DC. They are light emitting diodes, but still diodes, just like those in a rectifier. Just make sure you use the right resistor.
Heh heh heh. If you run LEDs on AC, you don't have to worry about polarity.
Actually, running LEDs on AC should not harm them, so long as the voltage is not excessive. After all, they are.................ta da, DIODES, used to rectify AC into DC. They are light emitting diodes, but still diodes, just like those in a rectifier. Just make sure you use the right resistor.
Actually they are a little different in that they don't tolerate reverse voltage very well and need provision made to handle that.
Joe
There's nothing wrong with cutting upt eh strings and using them as indivudual or groups of LEDs - with low voltage and the appropriate resistors. White LEDs with 12Volts are goot with a 1K resistor - if you plan and wire them all the same, instedof having numerous resistor values you can buy a whole box of oen value - in such quantities they are 5 cents or less each, when purchased from an electronic supplier like Mouser. Thus adding the resistors does not really negate any savings from picking up the cheap string of lights.
Mark R. I agree with the monitor refresh rate - I can't see it, but my wife can ! With an LED, I think the brightness tends to overcome the pulsations. If you put about a 20,000 ohm resistor on an LED hooked up to a 12vac source, the LED is dim enough that you can detect the pulsations. Mark.
I agree with the monitor refresh rate - I can't see it, but my wife can !
With an LED, I think the brightness tends to overcome the pulsations. If you put about a 20,000 ohm resistor on an LED hooked up to a 12vac source, the LED is dim enough that you can detect the pulsations.
This is correct. AND modern high intensity LEDs do get quite hot. They are off about 1/3rd of the time, but we are talking severaly hundred cycles per second. Here are two photos. Your eye cannot see the difference, but if your camera is too fast, it sure can!
BTW: You want to know if your TV remote is working... Just look at it through your camera, you will see the light!
Well, then.....your responses prove once again that there is no such thing as a stupid question. I won't be modifying Christmas light strings any time soon.
Thanks.
The St. Francis Consolidated Railroad of the Colorado Rockies
Denver, Colorado
Many people *can* see that 60 Hz flicker. I have one computer user who could never look at a regular computer monitor without getting a headache from the flicker. I never could see the flicker. I stopped short of telling him it was in his head, and simply bumped up the monitor to 85 Hz.
As for "why not" extend the wires... The danger comes in your reassembly of them, one solder joint not perfect will cause 120 volt arcing at that point, and will lead to heat and fire. I have had some small 'fires' on my layout at 12 volts since my track circuit can put out 15 amps. I will have to install a fuse to limit it to 3 amps, since I know now that 3 will be sufficient for my requirements. But I can have as many as 12 "locomotives" potentially running at once on this layout.