cheap fade effect?

guartho · Post by **guartho** » Sat Jun 10, 2006 9:38 am

Is there a nice cheap n' easy way to have a fade effect on a circuit? I would think it could be done with capacitors, but my grasp on basic electronics is pretty tenuous and my experiments haven't worked. I'd like for some of my lights to fade up to on when activated and fade out to off when deactivated.

Sparky · Post by **Sparky** » Mon Jun 12, 2006 7:21 am

Yes there is, Zog sent me schematics of this, I think we have to repost them (the sticky threads expired once). Let me see if I can find the image to post a link.

Oops I see that these don't exaclty match your fade on when powered on fade off when powered off request. . .Zog might be able to toss in here before I get my cup of coffee this morning. . .

Sparky · Post by **Sparky** » Mon Jun 12, 2006 7:27 am

Here they are:
<A HREF="http://www.kc6sye.com/images/circuits/s ... _blink.jpg" target="_blank">Slow Pulsating Blink</A>

<A HREF="http://www.kc6sye.com/images/circuits/s ... _wtran.jpg" target="_blank">slow Pulsating Blink usin' a transistor</A>

guartho · Post by **guartho** » Mon Jun 12, 2006 10:01 pm

Thanks for the links. I'm not having much luck. I've been trying the top circuit. Am I correct in thinking that .047 Farads = 4.7 uF?

When I put power to my circuit the LED simply comes on. It behaves as it would without a capacitor. Am I using too small or too large a capacitor? I also had a 10 uF capacitor that I tried. I got the same results.

jwrjr · Post by **jwrjr** » Mon Jun 12, 2006 10:16 pm

No. .047 farads = 47,000 uf. My fader is predictable and reliable, but costs a bit more.

Sparky · Post by **Sparky** » Tue Jun 13, 2006 11:34 am

It has to be a very large cap, you are using it as a temporary short, as the cap takes power from the battery (time=0+) it appears as a short and no current flows through the LED. As the cap starts to take current (time>0) it starts to develop a resistance and some current will now decide the LED is a better path to ground and flow through the LED. Eventually the cap has all the electrons it can handle and it appears as an open circuit, and all current flows through the LED.

If you have a small cap the time that it takes to reach full charge is so quick that the LED appears to just come on.

guartho · Post by **guartho** » Tue Jun 13, 2006 1:06 pm

Hey, that's great to hear! It means that my understanding of a capacitor is not flawed, just my understanding of uFs!

Thanks guys. I can't wait to try some bigger caps.

guartho · Post by **guartho** » Wed Jun 14, 2006 9:17 pm

Well, of course the only possible local store, Radio Shack, has limited selection and even more limited knowledge.

But I did find what I'm looking for at The Electronic Goldmine. Since I won't be able to experiment with different sizes and then order what I need in quantity I was wondering if you guys could answer some more questions.

About how long would a .047 farad cap take to charge on a 12 volt system? What about other voltages like 9 or 3? Is there a simple formula for calculating this so I don't have to bother you guys? Is there a way to calc the size cap I'd need for a given time? For one effect I have in mind, it doesn't really matter, but my appetite is whetted now so I thought I'd try a torpedo tube or something. Thanks once more for the advice.

jwrjr · Post by **jwrjr** » Wed Jun 14, 2006 10:18 pm

The time to charge depends on the resistor you have controlling the current. Note: there is always a resistor, even if it is one internal to the power supply. A very large capacitor could draw enough current to damage your power supply, if the current is not limited. The equation is a moderately nasty one.

guartho · Post by **guartho** » Thu Jun 15, 2006 10:57 am

Thanks for the info as always JWRJr. I went hunting an online capacitor calculator and I think it will guide me to my desired results. I'd like to check my numbers though.

If I've got a 12 volt power supply-> 30 ohm resistor -> .047 Farad capacitor -> LEDs/LED&Resistor-> ground would that give me about a 2 1/2 second fade up? (within a half a second accuracy)

jwrjr · Post by **jwrjr** » Thu Jun 15, 2006 10:17 pm

That depends on how you are using the capacitor voltage to control the lights. If you are controlling them directly from the capacitor, then the current drain will cause the capacitor to charge more slowly. Here's another 'gotcha' - during that first 0.1 second that resistor will be dissipating over 4.5 watts (decreasing as the capacitor charges). If the resistor is not rated at high enough power, it will burn out.

Sparky · Post by **Sparky** » Thu Jun 15, 2006 10:46 pm

Here's a link to the equation not to nasty, you will have to pick your resistor (probably whatever is needed to protect the LED at full power). Also your capacitor's voltage rating should be twice the powersupply voltage this way it can with santd bieng fully charged and having the full voltage of the power suppyl present.

http://hyperphysics.phy-astr.gsu.edu/hb ... apchg.html

BTW this calculation will not be exactly right, since the diode will start stealling current right away. Hopefully it get's you in the ballpark.

Otherwise you're equation gets ugly, the LED's resistance forms a curve so you'll get some exponetial exquation for it. It can be done, if you know the LED's mathmatical representation, you could add it to the equation and solve with it in place. Maybe theres a note on that somewhere on google. . .

MillenniumFalsehood · Post by **MillenniumFalsehood** » Fri Aug 18, 2006 4:38 pm

Hey, that's a cool little circuit there, Sparky! Thanks for the link!

I have a question: Could I hook up a flash circuit(like in camera flashes; I have a disposeable camera that has one I could use) to the LED that triggers when the LED is at full brightness, to simulate TMP era torpedo launchers? If so, could you provide me an illustration?

I know that there is one in the works by one hobby electronics supplier, but I thought I'd try my hand at making my own.

Sparky · Post by **Sparky** » Sat Aug 19, 2006 7:23 am

Yes something with diodes or zener diodes, I'll have to hunt around for the diagrams

raist3001 · Post by **raist3001** » Wed Aug 23, 2006 12:36 pm

jwrjr wrote:No. .047 farads = 47,000 uf. My fader is predictable and reliable, but costs a bit more.

I can certainly vouch for this. JWRJR makes fantastic effects with circuits. Absolutely fantastic!! Not in the least bit expensive, and his circuits are plug and play. Simply wire your LEDS to them and your good to go

MillenniumFalsehood · Post by **MillenniumFalsehood** » Mon Aug 28, 2006 9:36 pm

Sparky wrote:Yes something with diodes or zener diodes, I'll have to hunt around for the diagrams

You found them yet? Just curious. . .I'd like to use something like this in my Bird of Prey, but I didn't want to pay big money for a kit when I could build one myself.

Sparky · Post by **Sparky** » Mon Aug 28, 2006 11:13 pm

sorry I havent' looked I'll get the circuits book out tomorrow and start skiming.

Sparky · Post by **Sparky** » Fri Sep 01, 2006 12:11 am

Ok update I looked through my old course book and thought what you wanted would be a voltage detector. But this might not be it, what you really want is 2 circuits, a voltage detector and a single shot trigger. One detects the voltage at the LED has reached 'full power' and the other fires a strobe once, when it is kicked by the voltage detector.

I have found a simple Schmitt trigger circuit that uses a 555 timer, 2 caps and a 9 volt source. The function it provides is this, after an upper threshold is reached the circuit output turns on, after a lower threshold is reached it turns off. ues a transistor to take this signal and switch on the strobe, the problem is that it wont be a 'flash' it will be on until the lower thresh hold is reached, what you need then is a single shot 555 timer circuit that pulses an on signal to your strobe light (easily uses an LED isntead fo the shroter lived strobe light).

If the speed at which the strobe light's cap. can charge up is as fast as the fastest flash rate of your LED, you should be fine, the strobe will always fire when the LED voltage reaches full power.

I will do some searching on 555 timers, used to have a nice card with the basic modes and operations it could do for you, the equations needed to get all the timing and voltage levels set.

I can transfer the diagram to the computer tomorrow, err later today.

MillenniumFalsehood · Fri Sep 01, 2006 11:49 am

Thank you!

Sparky · Post by **Sparky** » Wed Sep 06, 2006 1:03 am

Ok here's the schmitt/voltage comparitor:

http://home.cogeco.ca/~rpaisley4/LM555.html

search for schmitt trigger and follow the anchor link to the section on the page.

This output feeding the trigger on the Monostable Oscillator (also on this page) should allow you to turn on a white LED or strobe.

Don't forget a simple 2N2222 as a driver transistor for the LED (whites need a lot of current compared to logic circuits).

MillenniumFalsehood · Post by **MillenniumFalsehood** » Mon Sep 11, 2006 1:02 pm

So if I understand this right, I hook up the Schmitt Trigger to the fader, and the flash circuit to the Trigger, making sure that the capacitor for the fader is larger than the one in the flash circuit so it has time to recharge before the plasma-weapon light gets to full brightness?

Sparky · Post by **Sparky** » Mon Sep 11, 2006 1:28 pm

Yes that sounds about right; I would suggest bread boarding this so you can easily pull the resistors and capacitors as you adjust timing.

The issue with the strobe is that, if the power supply cannot provide enough juice to get the strobe's cap charged before the pulsing signal wants the strobe flashed, you won't get a flash for every peak. If you're using batteries for the power supply you might see this happening. You will notice it as the batteries fade for sure.

A wall-wart can easily be resized to provide more current getting the strobe cap charged up in time.

I'd liked to reiterate that camera flash strobes have a limited life compared to LEDs; a disposable camera's probably has even less life. If you consider it might be flashing X times an hour for a 4 hour model show that will start to impact how long it will last. I think they're rated for number of flashes, not hours of operation.

jwrjr · Post by **jwrjr** » Mon Sep 11, 2006 9:29 pm

The fader that I build has 3 parts and costs approx. $2.00.

Sparky · Post by **Sparky** » Tue Sep 12, 2006 11:18 am

I'm thinking at this point, a micro controller is your best bet. The fade effect can be simulated with a PWM output, most of the chips will have one pin configurable as a dedicated PWM function. And the code can be written so that at the peak it activates another pin for some amount of time, however long is needed to fire a storbe or white LED (it'll need a FET to handle the current).

With the microcontroller you can set it up so it doesn't fire every time, saving on the life of the strobe and adding fancier effects, all in code no change in hardware past the initial wiring.

jwrjr · Post by **jwrjr** » Tue Sep 12, 2006 2:03 pm

Exactly.

Eight pin microcontrollers can be purchased for less than $1, even in small quantities. What you can do with them is limited only by your imagination and programming ability (or the ability of whoever programs it for you). However, the more exotic programs may be a bit more hardware-intensive and thus cost more.

MillenniumFalsehood · Post by **MillenniumFalsehood** » Thu Sep 14, 2006 5:17 pm

Sparky wrote:I'm thinking at this point, a micro controller is your best bet. The fade effect can be simulated with a PWM output, most of the chips will have one pin configurable as a dedicated PWM function. And the code can be written so that at the peak it activates another pin for some amount of time, however long is needed to fire a storbe or white LED (it'll need a FET to handle the current).

With the microcontroller you can set it up so it doesn't fire every time, saving on the life of the strobe and adding fancier effects, all in code no change in hardware past the initial wiring.

I really don't have any programming skills whatsoever.

But it's not going to go continuously, just a simple push-button control and a timing circuit(if I can find one) to keep the current flowing until the LED shuts off. Geez, this is getting complicated.

jwrjr · Post by **jwrjr** » Thu Sep 14, 2006 10:42 pm

You could always get the fader from me. That sort of thing is what I do.

MillenniumFalsehood · Post by **MillenniumFalsehood** » Sat Oct 07, 2006 8:59 pm

Would you be able to draw up the circuit I want for free? I'm a bit strapped for cash right now.

jwrjr · Post by **jwrjr** » Sat Oct 07, 2006 10:47 pm

Send a request to jwrjr@bellsouth.net (so I can attach a drawing to the reply). It uses a pre-programmed chip. I don't charge much for those ($5.00, includes shipping within the USA). For a single fader, you will only need 2 other parts. But the drawing I'll send for free.