Simulated Candle Flame

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

Simulated Candle Flame

Post by jollyrgr » Wed Jan 01, 2003 4:52 pm

Does anyone know a location (URL) for a simulated candle flame schematic? I want something to drive yellow and/or orange LEDs to simulate a candle flame similar to the devices sold by C Crane and Sharper Image. I can handle the charging circuit myself. But how do you simulate a flickering candle? I don't want just a simple flasher (I can do that easily) but a psuedo random flicker.
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!

russlk
Posts: 563
Joined: Thu Dec 06, 2001 1:01 am
Location: New Hampshire
Contact:

Re: Simulated Candle Flame

Post by russlk » Wed Jan 01, 2003 7:40 pm

I recommend that you use a noise generator. You may find information on the web, or if you just cascade 4 opamps with a gain of 100 each (use capacitive coupling), the output will be noise. You will need to lowpass filter the noise so the flickering will be visable. Another possibility is the Radio shack opto receiver, 276-137, its output is noise in the absence of optical input.

Dimbulb
Posts: 324
Joined: Sun Oct 13, 2002 1:01 am

Re: Simulated Candle Flame

Post by Dimbulb » Wed Jan 01, 2003 10:20 pm

Searching for "white noise random" should bring up a few schematics. White noise source have been made with zeners such as an LM336, 15k resistor on the cathode and a 9 volt battery. Others use a pnp transistor. A dual op-amp presents a few interesting charecteristics. <p>The duration and the amplitude would need to have variation. A 555 timer and a sample and hold may be helpful. Since the frequency is too fast then taking one value every so often may be adjusted to give your application a sample rate that looks more realistic than a flickering that is too fast, but a really nice flame might use some variation in its frequency as well<p>If you analyze a flame with a photo-transistor you could build a screen that does not let the light through except a tiny pin hole. The infra red should be filtered with a red lens because you want what people can see. The output could be fed to a sound card for different portions of the flame.<p>Some random patterns of audio from FM station such as *classical or a talk show definetly has an interesting random pattern. Some music has a more predictable pattern to it. <p>There exists an interesting simularity in very ultra precision DC reference circuits, noise increases if the leads are not protectected in an inclosure this a very small noise which is seen against a DC background. Some op-amps have noise measurement circuits that amplify this while blocking the DC. I have seen plenty of these on data sheets but you only need a basic audio amplifier circuit.<p>A candle flame may turn out to be identified by an almost peuedo-sequential pattern with smaller variations at the base of the flame and more pronounced random variations at the tip of the flame. Concidering different noise you may find close approximation to the noisey air convection currents that can effect op-amps. There is alot written on the 555 and op-amps are inexpensive you may need a quad.<p>[ January 01, 2003: Message edited by: dim bulb ]<p>[ January 01, 2003: Message edited by: dim bulb ]</p>

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Simulated Candle Flame

Post by MrAl » Thu Jan 02, 2003 9:16 am

Hello there,<p>I once built a controller with a Z80
processor and 2k of RAM and about 1k of
ROM which i programmed to accept 'uploads'
from a host computer. I also included
a simple i/o port where you could connect
anything you wanted to really, so i connected
a dual color led with green and red colors.<p>Pulsing the green and red using the bits of
the output port, i got the led to look like
a little 'flame' that might appear on a candle
that you were talking about. It looked so
real that you were almost afraid to touch it
for fear of getting burned.
Part of the reason for this is because the
red and green led chips internal to the dual
color led are slightly offset from one another.
This means as the two colors are pulsed at
different rates the thing acutally looks like
it 'flickers' because the origin of the light
keeps moving back and forth.<p>If you dont want to go thought the trouble of
building up a controller board (or perhaps
Basic Stamp) then you might try two 555's driving
the red and green dual color led.
You can then easily experiment with the different
frequencies for each led, and their relationship
(causing a 'beat' flicker).<p>Has to be seen to be believed how real it looks !<p>Good luck with your circuits,
Al
LEDs vs Bulbs, LEDs are winning.

k7elp60
Posts: 226
Joined: Wed Dec 05, 2001 1:01 am
Location: Ivins, UT
Contact:

Re: Simulated Candle Flame

Post by k7elp60 » Fri Jan 03, 2003 6:53 pm

This is a simple solution that works fairly well.
Connect 2 reqular flashing LED's in parallel, and the parallel pair in series with an orange LED.
When power is first applied the orange LED will flash on and off. When the two flashing LED's get out of sync with each other the orange will flicker like a flame.
Ned :)

gadgeteer
Posts: 102
Joined: Fri Mar 07, 2003 1:01 am
Location: Ok!
Contact:

Re: Simulated Candle Flame

Post by gadgeteer » Fri Mar 28, 2003 1:40 am

I bought a Hallmark Christmas ornament a few years ago, something about the "Bearenstein Bears". It had a nice fireplace, with living-room-floor and mantle etc. (you had to buy the bears extra). But their FIRE! Ugh! They thought a FIRE could be simulated with a fast-flashing red LED! Their fire was molded clear yellow plastic, painted logs; backlit by that obnoxious red strobe light...<p>SOOOO---on a day when I had much more time than sense, I used a 4060 (I think that's the one with the built-in oscilator---just add resistor & capacitor); then a BCD to 7-segment LED driver (is that a 4017? Whaddever...) I chose the 4000 family 'cause they'll run fine on 3 volts, which was the design of the fireplace---two AA's. AND--- 7 yellow LEDs, tied to the segment outputs of the decoder/driver. I'm thinkin' I used basic diffused T1 yeller LEDs. The slower-flashing ones I placed close to the logs, the faster ones towards the periphery...<p>And THEN (drum-roll please) ...I scratched a thin line into the paint, into the crevice between the two logs; and placed a red-orange steady LED (so the crack will glow like embers).<p>The fireplace is spectacular...

Post Reply

Who is online

Users browsing this forum: No registered users and 14 guests