idea needed for a simple, automatic voltage regulator for 3V LED

Thread Starter

meehow

Joined Jun 3, 2021
8
An idea needed for a simple, automatic voltage regulator for 3V LED.

The application is for an electric model train (HO) which operates on 0V - 16V DC power supply.
Traditionally, these locomotives utilise a 12V micro lightbulb which changes brightness in line with voltage applied to track.
This isn't realistic from a modeller perspective. Brightness is expected to be constant regardless of train speed. The reasonable speed at which the train should travel calls for ~7V supply and that results with a dim headlights.

What is the best and easy way to achieve a constant 3V to the LED installed in the locomotive while the voltage at the track is variable?
The solution must not interfere with a DC motor fed from the same track.

Would this be suitable:
LF33CV
Fixed LDO Voltage Regulator, 3V to 16V, 400mV Dropout, 3.3Vout, 0.5Aout, TO-220-3
https://uk.farnell.com/stmicroelectronics/lf33cv/ic-v-reg-ldo-3-3v-to-220-3/dp/9755322
 

ronsimpson

Joined Oct 7, 2019
2,985
Something simple:
1622760288519.png
A 3V LED does not like 3V. But with current it will have a voltage near 3V. The voltage varies with temperature and from one batch to another.
Above circuit R1=1 ohm = 1.25A, 10 ohm 125mA, 100 ohm 12mA (set R1 to match the current you want)
There is also the LM317L which is a small low current version in a small package. Best for small trains.
This should work for 7 to 16V like you want.
 

Papabravo

Joined Feb 24, 2006
21,158
An idea needed for a simple, automatic voltage regulator for 3V LED.

The application is for an electric model train (HO) which operates on 0V - 16V DC power supply.
Traditionally, these locomotives utilise a 12V micro lightbulb which changes brightness in line with voltage applied to track.
This isn't realistic from a modeller perspective. Brightness is expected to be constant regardless of train speed. The reasonable speed at which the train should travel calls for ~7V supply and that results with a dim headlights.

What is the best and easy way to achieve a constant 3V to the LED installed in the locomotive while the voltage at the track is variable?
The solution must not interfere with a DC motor fed from the same track.

Would this be suitable:
LF33CV
Fixed LDO Voltage Regulator, 3V to 16V, 400mV Dropout, 3.3Vout, 0.5Aout, TO-220-3
https://uk.farnell.com/stmicroelectronics/lf33cv/ic-v-reg-ldo-3-3v-to-220-3/dp/9755322
Only with a current limiting resistor, or as a constant current source if it can be configured that way. Check the datasheet for details.
PS you do not want to use a fixed voltage regulator WITHOUT a current limiting resistor. LEDs are current mode devices with a forward voltage drop. Any voltage source you use in conjunction with a current limiting resistor has to be greater than the forward voltage drop. There are LEDs and LED strings that have integrated resistors. You really really need to know what you are doing before you let the magic smoke out.
 

AnalogKid

Joined Aug 1, 2013
10,986
The wide range of variability in a constant current diode's output. For example, the 10 mA part is +/-20%.

The 3-terminal regulator you linked will give a constant voltage. That plus one resistor will feed the LED a constant current.

Note that the power dissipation is the same no matter what type of linear regulator circuit you use.

LED part number - vendor page - data sheet - ?

ak
 

ErnieM

Joined Apr 24, 2011
8,377
With the constant current diode there is still a dead zone where the train moves and the LED is off. Heck, last time I saw a real train stopped in the station the headlight was on.

Before I was going to type out what I did once I Googled "model railroad constant lighting circuit" and found hits to the same idea. My idea came in #4.

(Actually I lied... I used Duck Duck Go and not Google.)
 

crutschow

Joined Mar 14, 2008
34,280
Below is the LTspice simulation of a simple, 2-transistor, constant-current circuit, that works down to about 0.7V above the LED operating voltage:
An R3 value of 65 ohms gives about a 10mA LED current.

R3 can be changed to give the LED current you want.
Q1 and Q2 can be just about any small PNP BJT transistor.

1622814539987.png
 
Last edited:

Tonyr1084

Joined Sep 24, 2015
7,852
Nothing to add except this: With trains with headlamps lit while sitting still, perhaps a small supercapacitor can be wired in parallel to maintain voltage for a short period of time. Not quite the same thing as real life - but it's closer than the light going off just as the train comes to a stop. Depending on how much room you have inside the HO engine, you can get quite fancy with electronic gear.

Like I said, beyond that I have nothing much to add.
 

Thread Starter

meehow

Joined Jun 3, 2021
8
@meehow Have you considered what will happen when you reverse the voltage to the track?
yes! these are directional lights. only one end of the locomotive is supposed to be lit at any given time.
the entire 'setup' has to be doubled for opposite polarity taking care of another end of the loco.
can also wire small, red 'position' LEDs to opposite end.
 

MrChips

Joined Oct 2, 2009
30,706
A large enough resistance makes a decent constant current source.
Just select the largest value that gives a reasonable LED brightness.
 

ElectricSpidey

Joined Dec 2, 2017
2,757
@meehow

Next question:

Does your system automatically switch the power from the forward lighting to the back?

My concern is reverse polarity protection for some if not all of the options provided.

I know for a fact the LM317 will not like it, and I'm not sure about the rest and even if the circuits can survive the LED might not.
 

crutschow

Joined Mar 14, 2008
34,280
For my circuit, you could add a Schottky diode or logic-level P-MOSFET (source to supply, gate to ground, drain to circuit) in series with the power source to protect it from reverse voltage.
The diode will drop about 0.4V, the MOSFET would have essentially zero drop.
 
Last edited:

AnalogKid

Joined Aug 1, 2013
10,986
@meehow Have you considered what will happen when you reverse the voltage to the track?
If the train is expected to run backwards with the light on, then add a small diode bridge to Wally's #13. This adds 1.2 V to the minimum voltage necessary for light, which is still below the ~7 V mentioned in post #1.

ak
 
Top