Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Incandescent LED

Status
Not open for further replies.

mtrails

New Member
What, you ask? Incandescent LED?

I want to know if there is a way to regulate the current, switching on and off of an LED, that will make it light, and extinguish like a conventional filament light bulb. You know, when a light bulb is turned on, it takes a millisecond (or more) to reach maximum output, and likewise when turned off, vs. an LED which is at maximum output instantly.

I'm thinking, capacitor charge/discharge circuitry. Any ideas or suggestions are appreciated.

Thanks.
 
A PWM circuit which starts at a low duty cycle and increases to a high duty cycle will achieve this effect.
 
So, this type of circuit would delay the current going to the LED when switched on, and would light it up at a partially slower rate until continuous current (@1.6V) is reached, then allow current to dissipate rapidly when the base current is removed? Is there a general schematic for this type of circuit available?
 
No, PWM is a common way of converting digital signals to analogue signals. It is commonly used with a microcontroller.


No extra circuitry is usually required. It relies on the fact that you can turn an LED (and many other devices) on and off faster than human perception can keep up.



If you want your LED at 100% brightness you apply the right voltage to it (and its protection resistor) for 100% of the time.

If you want 50% brightness you rapidly switch the LED off and on so that it is on for 50% of the time and off for 50% of the time.

To get 1% brightness you switch the LED on for 1% of the time and off for 99% of the time.



I hope that helps. Im a newbie myself...
 
There is no delay of current. There is only current or no current for different amounts of time. If the ratio of current "on time" to current "off time" is small the LED will be lit so dimly that you cannot perceive it. If the ratio is large then the momentary absence of current is not detectable by the eye.
 
Um, brightness isn't what I am after. Controlling it's output level is also not what I seek from the LED. When I turn it on, I want the voltage to increase at a fixed rate until it reaches peak voltage, then stay there, making the LED light up like a filament bulb. Take a halogen lamp for example. Turn it on, and it takes at least two milliseconds to reach maximum output. When you turn it off, as the heat dissipates from the filament, the light output takes up to four milliseconds to completely diminish. This is the effect I want to achieve when turning on and off an LED.
 
hmm, I don't think you fully understood what people said.

first off, why do you care HOW it happens? Isn't the perceived effect of the bulb more important. That is what they are saying.

secondly, an LED is a current device. ramping voltage is not the best way to control it. However, you should think about PWM as a way to control voltage over a fixed interval. 50% PWM or 1 mS is similar to applying 1/2 the voltage over 1mS. So going from 100% duty cycle to 0% duty cycle in a couple of hundred mS (I'm pretty sure it's not 4 mS to go off) will do EXACTLY what you said.

I think if you stopped resisting the suggestions and try them out you will discover that the solution is pretty much what you described.
 
mtrails said:
Um, brightness isn't what I am after. Controlling it's output level is also not what I seek from the LED. When I turn it on, I want the voltage to increase at a fixed rate until it reaches peak voltage, then stay there, making the LED light up like a filament bulb. Take a halogen lamp for example. Turn it on, and it takes at least two milliseconds to reach maximum output. When you turn it off, as the heat dissipates from the filament, the light output takes up to four milliseconds to completely diminish. This is the effect I want to achieve when turning on and off an LED.

Did you try it?
And why this circus, may I ask?
 
I apologize, I'm not up to par with electronics, terminology, and common circuits. No I didn't understand the information given to me, but I was hoping by joining this forum, that I could get a schematic for the circuit I would like to build. It's for a model railroad project, to make the crossing signals.

As far as a few comments made toward my plight, I'm sorry not everyone can offer their advice without being demeaning.

Thanks everyone.
 
LED dimmer

Just an RC combo across the LED should do it nicely...
 
Try to switch the LED with a NPN transistor like a BC 547.
Turn the transistor on via a 10 k Ohm resistor and put a small capacitor from the base to the emitter say 2.2 µF.
The LED will be in between the positive of the supply and the series resistor ( e.g. 1 kilo Ohms for 12 Volts DC ), then the collector of the BC 547, then emitter to negative or ground.
The turn on will be delayed via the 10 kilo Ohm resistor charging the Cap and turning the transistor on slowly. When turned off the opposite happens and the cap will slowly discharge and turn the transistor off slowly.
You may have to experiment with a few cap values and perhaps put a bleed resistor in parallel with the cap.
 
Tell us your supply voltage, whether it is DC or AC, the amount of LED current you want, and what you are using to control the switching (switch, voltage level, etc.). I realize that descriptions of circuits is probably not going to work for you. We should be able to come up with a schematic.
EDIT: Are you sure you want millisecond switching? I read that incandescent traffic lights take on the order of 100-200 milliseconds to come on. I would think crossing signals would have similar response times.
 
Last edited:
Traffic signals are a special case. They are driven under the design spec of "soft-start, and keep-warm." What normally is required is the lamp to come to full brightness over a period of time, and go off the same way. If the lamp is to be flashed, it is driven at a non-visiable, but filiament heating level when off. The filiament glows slighty, but is not visable unless you look really closely. This means the filiament never actually completely cools off when flashed, dramatically easing mechanical stress on the lamp, extending its life by several fold. The soft start is used for the same reason.

PWM "sweep" is the most common way this is done. One simple way is to have a triangle wave generator drive a VCO, so that you get a rising/falling frequency output. Gate this so you can control the start and stop, and you have a non-micro, or non-pic variable PWM.
 
One of the problems I have with 2ms turn-on time is that the bulb is undoubtedly powered by AC, either 50 or 60 Hz. Need I say more?
I suppose it's possible for halogens bulbs running on DC to switch on that fast.
 
What they are trying to say is that LED are diods and pretty much like a fixed
voltage. (on or off)

But if you switch this on/off so fast the human eye will think you dimmed it.

So if you create a circuite that is:

on for 1ms and off for 9ms 10 times
on for 2ms and off for 8ms 10 times
on for 3ms and off for 7ms 10 times
and soo on.........

You get a nice slow dimmer effect.
The above numbers would take it from 10% to 100% in 900ms
That is just a tad under 1 second

If you put a 10ms square wave on step-pin on below curcuit and
you are pretty much set.

LED_Dimmer.htm
 
Last edited:
Another explanation, you play with the duty cycle of a square wave.

Lower your duty cycle will be, lower the brightness will be, and so on.

As stated, you trick the human eye. PWM is usefull in many different way. you can use it as a variable voltage source, just send the signal through an RC filter and OPamp buffer and TADA. Kinda DAC converter.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top