[Paul Norton]

I am installing a 10mm (large) white LED in an old diesel but can’t get it to light. I am using a freshly charged lithium-ion battery and a 27 MHz TE receiver to drive the circuit. A 5mm LED will work, but the 10mm will not. The branch of the parallel circuit containing the LED has a 1K ohm resistor in it to limit current.
 
Does anyone know for sure what is the recommended voltage and maximum current are for this device?

 

[Torby]

Seems to be 3.6 to 4v at 25mA.

You don't say what voltage you're running at, or I'd figure you a resistor. 1k seems really big unless you have like 28v.

 

[Del Tapparo]

I assume youv'e tried the obvious and turned the LED around (changed polarity) to see if that works? What voltage is the LED connected to? I am assumming the TE is providing a relay or open collector output (I don't know), and the other side of the LED / resistor is connected to your battery or track power?

 

[Paul Norton]

I could not find anything conclusive, but I am guessing a 10mm white LED is a 3.6 volt device because I could light it with 3 volts. I had to make sure it was not burnt out.
 
The A-C lithium-ion battery was freshly charged at 24 volts, but a couple of volts are lost within the TE. From my calculations a 1K resistor would be perfect to limit the current to a maximum of 20 milliamps at full voltage. This is the same maximum as a 5mm white LED.
 
I did try both directions and no light.
 
I do have a couple of bi-polar, red/green, 3mm LEDs in another branch of the parallel circuit and these LEDs change color when the polarity changes. I almost get the feeling the current is taking the easier path through the smaller LEDs and their two 1K resistors, and ignoring the other rung on the circuit, as they are bright for 2 volt devices.
 
Tomorrow I’ll try removing the smaller LEDs and see what happens.

 

[Dan Pierce]

Or it could be the 10mm is defective. Do try it alone with the 1k resistor to find out.

 

[Torby]

I get about 720 ohms, so 1k should be a reasonable approximation.

 

[Paul Norton]

Some daze, like yesterday, you can’t see the forest for the trees.

 
This morning after a good night’s rest I set up a simple test circuit with a 1K resistor and the 10mm white LED. The power supply was my power car with a TE and A-C Lithium-Ion battery. The input voltage to the circuit was 22.5 volts.
 
Voltage Supplied â€" Voltage of the Device (LED) / Maximum Current for the Device (LED) = Resistor Required
 
22.5 Volts â€" 3.6 Volts / 0.020 amps = ? Ohms
 
18.9 Volts / .020 Amps = 945 Ohms
 
Hooked up the circuit with a 1K just like before and the LED didn’t light until the TE was at full throttle. Well ya dummy, full throttle voltage was what you used to calculate the resistor. DUH! Give you head a shake, you brain is stuck.

 
So maybe a 5 volt voltage regulator and a 100 ohm resistor (5.0 â€" 3.6 / .020 = 75) would provide constant lighting. A diode on the positive input leg of the voltage regulator would prevent magic smoke when the current was reversed and provide directional lighting. Correct?
 
But what do I do with my Christmas (red/green LED classification) lights? If I add them to a parallel branch, they will only light one way?
 
Sound of grinding gears heard, temperature rising, smoke begins to waft ……… I need a constant voltage supply for all the devices? The battery! But how can I get directional lighting if I use the battery?

 
My head hurts! Time for lunch!

 

[Torby]

Hmm. I drew this some years ago for somebody who wanted to turn his battery lights off when he wasn't running the train. 

If you put only 1 of the diodes in, the transistor would turn on going one direction, not the other.

At 20mA, you're not looking at a very big transistor. A small signal line 2N222 would do fine.

 

[jebouck]

Here's a current calculator for LEDS.
http://ledcalc.com/

 

[Greg Elmassian]

Paul, when you bought the LED, you did not get the specs?

The main reason you had a problem is that the voltage drop of the LED larger than most LEDs, thus it has become a significant part of the voltage drop in the circuit of the resistor and the LED.

For most "ordinary" LEDS, simply calculating the required resistance as though the LED was a dead short works fine, and gives you a bit of safety margin. But on this white LED, it has significant voltage drop. Thus your resistance was too high, and at lower voltages, not enough current flowed.

Your other LED worked because the lower voltage drop allowed a wider supply voltage to get minimum current through the LED.

Regards, Greg

 

[Paul Norton]

Hi Greg!
 
I bought the large 10mm LEDs a number of years ago to use as FA-1 headlights. I never used them however, because a 5mm LED fits perfectly in the headlight lens. Now they are in a small parts drawer with no information./DesktopModules/NTForums/themes/mls/emoticons/crying.gif
 
As I can light them with 2 AA cells, I assume they are 3.6 volt devices like the 5mm white LEDs or they wouldn’t light. They also appear bright at 20 ma, so again I assume the have the same recommended maximum current of 20 ma as 5mm white LEDs do.
 
Yes I realized after a good night’s sleep that I had based my calculations on the maximum voltage of the power supply (TE) and used a slightly larger resistor, so unfortunately no current could flow even at full throttle. I am too used to the 75 MHz lighting circuits that provide a constant voltage. /DesktopModules/NTForums/themes/mls/emoticons/satisfied.gif
 
My problem now is how to wire:
-         constant intensity and directional front and rear headlights (white LEDs),
-         constant intensity and directional front and rear classification lights (bi-polar red/green LEDs), and
-         a constant intensity and non-directional cab beacon (flashing yellow LED).
 
I guess my best bet is to test and rewire the power distribution board that came with this old locomotive. It would probably take less time and cost less than bridge rectifiers and voltage regulators. Unfortunately as Dave Winter has found out, there is no guarantee the board will work. /DesktopModules/NTForums/themes/mls/emoticons/crazy.gif
 
I have to learn to stop taking things apart or at least save them in one piece. /DesktopModules/NTForums/themes/mls/emoticons/blush.gif

 

[Paul Norton]

This above reply hung when I pushed the submit button, and it’s only copied text with GIFs added. Something is definitely wrong!
 
After waiting more than a minute I pushed the stop icon on the browser and then the cancel button for the post. A while later it posted anyway?

 

[Greg Elmassian]

5 volt regulators, protected with a diode? that would let them turn on only when the track polarity was right, and then use a dropping resistor on the 5 volts for your 3+ volt led to get right current?

Oh, sounds like you may be battery (although the 75 mhz setup was an onboard system), in that case, hook your regulators to the motor leads.

Regards, Greg

will see if a smiley makes a problem posting... no subsequent post means no problem...

 

[Paul Norton]

Greg I thought about two, 5 volt regulators with a diode for the front and rear head lights (white LEDs). I could also have used a bridge rectifier and a 12 volt regulator for the beacon (flashing yellow LED). Running the bi-polar red/green LEDs (classification lights) with voltage regulators and diodes might also have worked. Time is of the essence however, as I have spent way too much time on this locomotive already. The buyer also has a power car he wants me to assemble.
 
I am making progress with the factory electronics in this old locomotive however. The track power wiring was replaced with an MU plug for a 27 MHz power car. The motors are hooked up and run fine. I tested a bunch of spare light bulbs and found enough that will handle 24 volts for the class lights, headlights, number boards, and beacon. Tonight I found the circuit for the beacon and have it flashing even at slow speed. I also found the circuit for the directional front and rear headlights. The front headlights are installed and working. I will install the rear headlights and classification lights tomorrow.