hon30critter RR_Mel Dave I’m not very good taking pictures but I tried on the SMTL4-RGY Okay Mel, you convinced me to break down and spend the money for the LEDs that you suggested. They were considerably cheaper by the hundred so that's what I bought. Now I just have to wire the things! I'm going to blame you for my stiff neck!!! Seriously, thanks for all your help with this! Cheers!! Dave
RR_Mel Dave I’m not very good taking pictures but I tried on the SMTL4-RGY
Okay Mel, you convinced me to break down and spend the money for the LEDs that you suggested. They were considerably cheaper by the hundred so that's what I bought.
Now I just have to wire the things! I'm going to blame you for my stiff neck!!!
Seriously, thanks for all your help with this!
Cheers!!
Dave
RR_MelDave I’m not very good taking pictures but I tried on the SMTL4-RGY
I'm just a dude with a bad back having a lot of fun with model trains, and finally building a layout!
Hi Mel,
Thanks for taking the time to do that! I like the colours (Randy won't like my spelling of that word!).
I'm going to approach the club to see how much the colour of the green indication matters. One distinct advantage to going with the tri-coloured LEDs is that it gives us the option of setting up a proper signal system down the road. Right now all we want is turnout position indication but that could very well change if we want to install a central control panel and use a proper dispatcher some time down the road.
Thanks again Mel! You (and others) have given the club a lot of support! We are very gratefull for that.
richhotrainIs all of this correct?
Yes.
P.S. That is the shortest answer I have ever given on the forums!!!!
rrinker The other point everyone is trying to get across is that at 20ma, most LEDs will be far too bright, and you will most definitely want to use a larger resistor, resulting in a lower current.
The other point everyone is trying to get across is that at 20ma, most LEDs will be far too bright, and you will most definitely want to use a larger resistor, resulting in a lower current.
Rich
Alton Junction
The unresistored LEDs would draw whatever power your DCC system can supply - for a nanosecond, as they all instantly flashed into nothing. Then they would draw zero current.
All LEDs MUST have a resistor to limit the current to no more than 20ma. Without a resistor, the LED will draw whatever current the power supply can supply, even if that exceeds the limit and destroys the LED.
The other point everyone is trying to get across is that at 20ma, most LEDs will be far too bright, and you will most definitely want to use a larger resistor, resulting in a lower current. For the LEDs Mel is using, it's truly a miniscule current that results in a brightness he likes.
Bottom line of it all is - it's not going to matter at all to run LED turnout indicators right from the DCC bus. Unless you are running the whole layuout with one of the really low power starter systems. But a 5 amp system or more - the LEDs will be negligible. It's like having resistor wheels on cars for detection. It's a resistor right across the rails, so it draws current from the system. My cars have 5K per car. It would take 1000 cars to draw 3 amps.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
richhotrain All of this talk about about milliamps prompts me to ask, how do you determine the amount of milliamps on an LED. I have seen various reports stating LED current draws to be 20 to 30 ma without a resistor attached. Is there a way to measure exact milliamps for a specific LED? Rich
All of this talk about about milliamps prompts me to ask, how do you determine the amount of milliamps on an LED. I have seen various reports stating LED current draws to be 20 to 30 ma without a resistor attached. Is there a way to measure exact milliamps for a specific LED?
So, if I have 50 signals, the unresistored LEDs would consume 1,000 ma, or 1 amp. However, my resistored LEDs (carried over from my prior layout) each only draw 0.01 amp, or 10 ma. So, I assume that the resistors that I am using reduce the current flowing through each LED by 50%, thereby consuming only 1/2 amp in total.
That seems acceptable, but if I want to reduce the total current draw to 1/4 amp, then I need to increase the resistance value of the resistors used by another 50%.
Is all of this correct?
Yeah, these 3 in one LEDs wouldn't work for me anyway, nearly all of my signals need to be 3 light Type G signals. There are single green LEDs with the right color frequency to equal the more aquamarine of railroad signals. If you ever get to see the inner workings of a signal, take a look at the 'green' filter - very definitely bluish green, not a bright grass green.
On LED current - 20ma is a MAXIMUM. LEDs do not work on voltage like a filament bulb does, they work on current. Otherwise the same things apply. A filament bulb will start to glow well below the rated voltage. Exceed the rated vooltage, and the life drops off sharply, excced the rated voltage by enough and the bumb burns out nearly instantly. With an LED, they will light up when sourced with well below their maximum current. Just nooot at maximum brightness. Exceed the rated maximum current, and the lifespan is shortened. Exceed it by a lot, and you create DEDs, dark emitting diodes.
A filament bulb will take all the voltage you give it, to the point of destructions. But it will consume only a certain amount of current, say it's a 540mA bulb, then it draws 50mA. (clightly complicated by the fact that the filament material has a different resistence depending on if it's hot or if it's cold, but basically it's one current level). An LED will take all the current supplied to it, even to destruction, but it will only ever drop a set voltage, say maybe 2.1 volts for a red LED. Doesn't matter if you allow only 2mA, or 20mA, the voltage across the LED will be 2.1 volts if that's what the specifications say. (subject to the fact that the perfect semiconductor does not exist and the voltage/current graph is not perfectly linear, but the idea is close enough for our purposes).
The voltage controls the brightness of a filament bulb. The current controls the brightness of an LED> The only thing you can say for sure about what current to use with an LED is to never exceed the maximum. The data sheet might give a chart of current vs light intensity, but without a light meter, those things are almost meaningless. Experimentation is your best bet. The only way someone else's resistor values would also work would be if they are using EXACTLY the same LEDs that you are using, their values will not likely match yours.
Glad it appears we have found Rich a solution. Some interesting info here.
I use LED signals but decided against targets in favor of D type three lamp and two lamp color light heads.
And my wiring is done to allow LED's or 12v lamps.
Sheldon
gregc with a light meter. 20ma is typically the max. But over the years, LED optical intensity has increased. The value for different color LED and manufactures may be different.
with a light meter.
20ma is typically the max. But over the years, LED optical intensity has increased. The value for different color LED and manufactures may be different.
greg - Philadelphia & Reading / Reading
RR_MelThe LEDs I went with are SMTL4-RGY from Bivar.
Thanks for the resistor values.
Do you have a picture of the green indication? The LEDs that I am using for the club's signals are about $4.00/100. The Canadian price for 100 of the LEDs that you are suggesting is $155.00 incl. tax and shipping. I'm not happy with the green tone on the cheap LEDs but I'm reluctant to spend that much money unless the green is hugely superior.
rrinker You have better vision than I do, at better than 30 years older than me. No way can I do those super fine wires on SMD devices. And unfortunately nothing can fix that, I'm for the most part only seeing out of one eye, so it's not just the sharpness, my depth perception is off. Are those the RGY LEDs from RR-CirKits? The ones he sells have a perfect aqua green for railroad signals - the typical green LED is WAY off. You might not notice it until you see what one of the nice ones looks likes. --Randy
You have better vision than I do, at better than 30 years older than me. No way can I do those super fine wires on SMD devices. And unfortunately nothing can fix that, I'm for the most part only seeing out of one eye, so it's not just the sharpness, my depth perception is off.
Are those the RGY LEDs from RR-CirKits? The ones he sells have a perfect aqua green for railroad signals - the typical green LED is WAY off. You might not notice it until you see what one of the nice ones looks likes.
RR_MelI run my three color LEDs at; Red 2ma Yellow 11ma Green 4ma
I'm working on tri-colour signals for a friend. I've only worked with tri-colour LEDs once before and those were for classification lights so having them bright wasn't too much of an issue. I'm curious to know what value resistors you are using.
Thanks,
RR_MelRich 10ma on LEDs is a bit too bright for me, I would add some resistance to drop it to 5ma. That would drop your total current from 1 amp to a ½ amp. Mel
Hi Rich,
I agree with Mel. Increasing the value of the resitors will not only save you some power, but you may also find that the brightness of the LEDs is toned down to the point where they don't burn your eyes if you look straight at them. I have seen some signal systems where the LEDs were so bright they were almost blinding.
richhotrain I just tested with my RRampMeter. Each signal draws 0.01 amp. That would mean that it would take 100 signals, more than double what I need, to consume 1 amp of DCC power. Wouldn't that be acceptable? Rich
I just tested with my RRampMeter. Each signal draws 0.01 amp. That would mean that it would take 100 signals, more than double what I need, to consume 1 amp of DCC power. Wouldn't that be acceptable?
i think so
OK, I found the solution to the bi-polar LED issue.
By placing a diode in line with the Green LED wire, I can prevent a current backflow that was turning the red/green LED to an orange blended color.
Now, my only concern is bleeding off too much DCC power by wiring the signals to the DCC-powered rails. I do have a RRampMeter hooked up to the layout, so later today I will check the effect of voltage drop. One or two signal hookups should not matter, but I have a considerable number of signals to wire up, so voltage drop may be a serious issue. Any thoughts?
richhotrain ATLANTIC CENTRAL Rich, do you have any diodes in these LED circuits? It seems to me that this might simply be the bipolar LED not liking the DCC signal on the one color? LED's are after all, DC devices. The DCC signal is not really DC. The thing that concerns me about this is voltage drop? Sheldon No, I do not have any diodes in the signal wiring, but I wonder if diodes are the solution to the bi-polar LED wiring. I wish that I understood more about such wiring. When you say that you are concerned about voltage drop, are you saying that powering the LEDs off the rails may pull too much power from locomotive performance? Rich
ATLANTIC CENTRAL Rich, do you have any diodes in these LED circuits? It seems to me that this might simply be the bipolar LED not liking the DCC signal on the one color? LED's are after all, DC devices. The DCC signal is not really DC. The thing that concerns me about this is voltage drop? Sheldon
Rich, do you have any diodes in these LED circuits?
It seems to me that this might simply be the bipolar LED not liking the DCC signal on the one color?
LED's are after all, DC devices. The DCC signal is not really DC.
The thing that concerns me about this is voltage drop?
No, I do not have any diodes in the signal wiring, but I wonder if diodes are the solution to the bi-polar LED wiring. I wish that I understood more about such wiring.
When you say that you are concerned about voltage drop, are you saying that powering the LEDs off the rails may pull too much power from locomotive performance?
Yes
gregc richhotrain But when I do that with the search light signal, the bi-polar LED turns amber on one of the routes. isn't this a good thing? should the signal be clear if the turnout is routed straight, and indicate a lessed indication if switched to the diverging route? and shouldn't the signal toward the turnout on the diverging route be either stop or approach (yellow)?
richhotrain But when I do that with the search light signal, the bi-polar LED turns amber on one of the routes.
isn't this a good thing?
should the signal be clear if the turnout is routed straight, and indicate a lessed indication if switched to the diverging route?
and shouldn't the signal toward the turnout on the diverging route be either stop or approach (yellow)?
richhotrainBut when I do that with the search light signal, the bi-polar LED turns amber on one of the routes.