I am in the process of adding working classification lights to an Atlas S-2. I am using 2 - 3.4 V, 20mA SMD LEDs in series, with Digitrax DN135D decoder. My DCC output is 14V. If my math is right I will need a 360 ohm 1/4 watt resistor at minimum (I'll use 470 ohm on hand). Ok so far?
I have never run LED's from the same Fx before, so I want to make sure I have this straight before I solder everthing up (30 gauge magnet wire is a PITA to work with - not to mention the SMD's!). Below is my crude circuit diagram. Please let me know if I'm on the right track.
Blue common wire====470 Ohm R=== (+)LED1(-)===(+)LED2(-)===Green wire (Fx1)
Thanks for your input.
Richard
Your theory is correct on how to wire the LEDs, but are you sure they are rated at 3.4 Volts ? Most LEDs work on 1.2 Volts, so two in series would be 2.4, not 3.4, and the current rating will not be additive.
As an initial test, I would use a 1K Ohm resistor and see if they light up. If not, drop the resistor value in increments until they do. Too small a resistor value due to miscalculation can shorten their life.
My calculations indicate that you need at least 530 Ohms for the resistor at 3.4 Volts, or 580 Ohms for 2.4 Volts at 14 Volts input from the decoder, and it's always good practice to use a higher value resistor. IMHO, a 470 Ohm resistor can shorten their life.
Below is the info that came with the LEDs. They are 3.4V EACH! I was surprised myself - they are no bigger than a crumb! I initially hooked one up to a AAA battery to test and couldn't get it to light, that's when I "read the instructions" and discovered they were 3+V LEDs. I should have been more specific in my initial post. They do light up (VERY BRIGHTLY) when hooked up to 3V source.
My math was based on two 3.4V LEDs in series = 6.8V. With this new info does everything add up?
Electrical-Optical Characteristic (Ta=25oC)
Item Conditions
Yes the math adds up. The voltage you should use isn't the track voltage but what is at the FX output, which is probably 12v. The next thing is it might be better to not run the LED's at max. Start with a larger resistor, say 680 ohm and see if it's bright enough. If not then go to the next value. Running LED's at max brightness is usually too high. It also reduces the life of the LED. A relatively lower voltage will significantly increase the life.
Springfield PA
My opinion, would be to run the blue (+) to each led seperately as well as the function (-) wire. A 1k ohm resistor would be placed with booth leds, does not matter which side. While you can group leds together, it's still good practice to wire them seperatly.
As for soldering smd leds, I've got them down to about 5 seconds each. Cut your magnet wire longer then you need. Stick the led onto a piece of double sided tape or similar surface. Use some Tix brand liquid flux and flux the ends of the magenet wire. Use a toothpic to put a tiny drop of flux on the led solder pads. Touch the soldering iron to your solder to cath a bit on the tip. Touch the iron to the pad of the led for a brief second to transfer solder. Repeat for the other side. Do the same on the wire but if will take a little longer since you are also burning throgh the coating. Take your tinned wire end and hold it to the led and touch the iron to them for another brief second. Test your led!
White LEDs are almost ALWAYS 3.2-3.4 volts. Even most color LEDs these days are 2.1 volts, not 1.4 anymore.
Simple calculation is (voltage of decoder - LED voltage)/led current
20ma is .020 amps for the equation. But do NOT base the calculation on 20ma, that is the MAXIMUM the LED can handle. With standard resistor valus plus 10% tolerance you cna easily exceed the maximum. Instead base the calculation on half the maximum, the LEDs will still be plenty bright.
Or just forget all the calculating as it's been done many times before, and when usign an LED with a 3.2-3.4 volt rating with DCC, use a 1K resistor. It works out to the 10ma current level anyway.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
The reason I suggested 680 over 1k is because he's wiring them in series.
With a 14 volt DCC output, the decoder output will probably be around 12.5, due to the voltage drop of the bridge rectifier on the decoder. (12.5 - 3.4 - 3.4) / 470 = ~12.1ma, well under the maximum rating of 20ma for the LED's, so that should work fine; however, depending on the LED's they may be too bright.
graphitehemi My opinion, would be to run the blue (+) to each led seperately as well as the function (-) wire. A 1k ohm resistor would be placed with booth leds, does not matter which side. While you can group leds together, it's still good practice to wire them seperatly...
My opinion, would be to run the blue (+) to each led seperately as well as the function (-) wire. A 1k ohm resistor would be placed with booth leds, does not matter which side. While you can group leds together, it's still good practice to wire them seperatly...
You gain two things by wiring them in parallel: If one LED goes out, the other one will still keep working, and you can adjust the brightness of each LED seperately by using different resistor values.
By wiring them in series, you get less current draw from the decoder, and less heat build up in the single resistor than you would have in each of the two resistors if wired in parallel.
I'm always glad that i ask this type of question before I fire up my soldering iron. I have often wondered why my incandescents seemed a bit dim. With the explanation of the 12v -vs- 14 v at the Fx output it makes sense.
My reasoning for wiring the light in series is the lack of room under the shell. I'd like to keep the weight, so I'm trying to minimize the extra components.
If I get time I plan to take some pictures and illustrate my method of lighting the brass castings. It's delicate work for sure, but someone else may find it of use. If it comes out as I plan it will look cool too!
Thanks again for all of the input.