I am trying to build a circuit to flash, alternately, 2 LEDs, with something close to a 50/50 duty cycle (which I have read is a bit complex for the 555, but anything close is fine, no need for perfection in this case) at somewhere around 5-10 flashes per second. I have a circuit that seems to work, and I can get the LEDs to flash, but the problems I am having are:

(1) How do I get the appropriate flash rate that I am looking for? I know changing the resistor and/or capacitor values will change the rate/cycle but how exactly do I figure out what resistors to use?

(2) After about 10 seconds or so, the LEDs stop flashing and remain steady, both lit, one very dim, and the other at a normal level. If I disconnect the power for a minute or so, and then reconnect, I get the same problem. If I disconnect power for only a second or two, then reapply the LEDs will flash a few more times, but then stop.

I am using high power 3-watt LEDs, could this be a major problem? (The resistors for the LEDs themselves are 5W 27Ω resistors which get a bit warm, but not hot while running). The supply voltage is 12v from a battery, and I am currently using a 2.2kΩ R1 and a 1MΩ+470kΩ series for R2, with a .1μF capacitor (these will end up being changed, I am sure in response to part one of my question).

This is the diagram I am using:

Thanks in advance to any and all help! If you can only answer part of my question, that is perfectly okay!

 

The typical NE555 has an output current rating of 200mA and a max power dissipation of 600mW, but those vary quite a bit over the various flavors and manufacturers. What is the exact brand and model number of the one you are using?

You don't say what the forward voltage drop is for your LEDs, but let's assume that it is 3V. With a 12V supply, that means you are dropping 9V across your 27Ω resistor, which translates to 333mA.

So you are overstressing your 555 and it is probably heating up and either going into some kind of thermal protection mode or else the performance characteristics are getting so skewed because it is so hot. You say that the resistors get warm, but what about the 555 itself?

Also, a 1MΩ timing resistor is pretty large. If the leakage current on the timing pins (threshold, trigger, discharge) gets high enough because of the high temperature, the thing will stall.

Try using the output of the 555 to drive the base of two transistors, one NPN and one PNP, to act as switches. You will need to put a current limiting resistor between the output and each base (two separate resistors!) and size it to limit the base current to approximately 25mA to 50mA in order to ensure that the transistors saturate.

A slight variant, in which the transistors are kept from saturating, and hence dissipate a bit more power, but which minimizes the base current to something around 1mA to 3mA, is the arrangement shown in the attached diagram.

 

Thanks for the prompt response! I knew when I found this forum it would be the right place to ask.

The issue with too much current for the 555 to handle is what I was guessing the problem was. After further investigation, it does seem that the 555 gets quite warm, so to avoid damage to it, I have put the project aside for the moment

When the circuit was left to set overnight and tested again today, the LEDs flashed for much longer, and in fact lasted a good minute before I noticed the 555 getting warmer and warmer, so I am thinking that leaving it to cool overnight is what made it work.

In any case, the idea with 555 driving the transistors as switches sounds like a good solution to me. I am fairly new to this kind of work with electronics, are there specific transistors I need to be using (ie, what do they need to be in order to handle the draw of the high power LEDs?)

Also here is more information on the specs of the LEDs and the exact 555 I am using:

555 IC: Looks like a Texas Instruments Part Number 18AG83M Model NE555P (from Radio Shack)

The LEDs are Vollong 3W RGB (TriColor) High Power LED chips, so they are actually 3 sets of specs for them. I am currently using the Red and Blue for desinging the circuitry, but I eventually would like to be able to use the other colors as well, with different circuits or something so they can be switched (ie, for a Red/Blue alternate mode or a Blue/Green alternate mode, etc etc)

Forward Current (Red) 400mA
Forward Voltage (Red) 2.5v
Forward Current (Blue or Green) 350mA
Forward Voltage (Blue or Green) 3.4v

More info on the LEDs here: http://www.superbrightleds.com/moreinfo/component-leds/vollong-3w-rgb-high-power-led/899/

Also, I want to eventually add some white LEDs to the mix as well (for RGBW operation of each module I am making). They are very similar, specs for those are here: http://www.superbrightleds.com/moreinfo/component-leds/vollong-3w-white-high-power-leds/896/

Thanks again for the help! If we get this figured out, I will also still need help on getting the flash rate/duty cycle set properly (I have a fairly decent array of various resistors/capacitors on hand, so if you have recommendations to try, I should be able to experiment.)

As the end result of this project, I am building light "modules" into project boxes that have the RGB and W Leds inside with a lens on the box, and I want to build a sort of controller that will hold the flasher module(s) I am constructing with switches for various pre-defined color flashes (Red/Blue, Red/White, Green/Red, etc etc.) I had figured on building a seperate flasher-circuit for each color-combination. The modules themselves will only flash one color at a time, with another module flashign the other color (as in, with 4 modules total, a switch could be flipped to make 2 flash red and two flash blue, or another switch could be flipped to make 2 flash green and two red)

Sorry for the long post, but with the Tri Color LEDs, can I safely use two colors at once (Red and Green for example) to make amber if they are on the SAME CIRCUIT but with their own current-limiting resistors)?

Please feel free to address just the first (MAIN) issue at this time: getting the flasher circuit working without burning anything up Thanks! The rest of the info is just in-case it means going in another direction to solve the first problem.

 

These transistors should work and they are cheap. Less than 50 cents each from a major seller. A 3 watt LED that uses 3 volts will also use 1 amp of current. That leaves 9 volts, or 9 watts of heat to get rid of.

12V supply-3V for the LED-.7V for the transistor = 8.3v
series resistor should be more than 8.3 ohms. Try 10 ohms, 10 watts.
The transistors will be coasting along at .7 watts which is less than half their free air rating.

Meanwhile the base driving resistors will need to supply .01 amps to each transistor. 12V/.01 amps is 1200 ohms.

See also:http://forum.allaboutcircuits.com/showthread.php?t=19075

 

Thanks for the detailed description of what you are trying to do. As you indicate, it can be very helpful to keep us moving along a path that is consistent with getting to the desired destination.

Looking over the data sheet for the TI NE555P, it has the typical +/-200mA drive capability. I would allow for the output voltage only getting down to about 1V and only getting up to about 2V below the supply voltage (but you also have to allow for it getting almost all the way to the rails, too). So you have to be sure that you can not only turn them on, but also that you can turn the transistor switched off.

There are a number of ways to design the circuit. Some real quick and dirty ways that are probably good enough, and some ways that will give you a bit more uniform performance from module to module.

As for the resistor values to get the frequency you want, look over the datasheet I linked and see if you can figure that out (you are working with an astable multivibrator, in case you didn't know that term). If you still aren't sure, we can walk through it with you.

 

Just occurred what you were asking for.

High Power LED Flasher

This from...

Completed Projects Index

This particular design is more than a little efficient. It will not quit until the batteries are sucked dry.

 

Thanks for the additional replies!

I saw that there is a solution that was already created, however, I don't want to completely ignore something that does indeed work, but I would like to continue with what I am working on still if only for the sake of learning.

I was able to get the PNP and NPN transistors that were mentioned and I attempted to hook them up but the result is that I simply end up with both LEDs lighting and staying lit.

In addition to exactly what I had before, I connected the BASE of each transistor to the output (pin 3) of the 555 through 300Ω resistors (actually its two 150s in series, but it should be close enough), the EMITTER for the PNP to V+ and the emitter for the NPN to V-. The load, in this case my LEDs are connected between the COLLECTOR of the transistor and the opposite V for the transistor (ie, red LED is between NPN collector and V+ and blue LED is between PNP collector and V-).

I can honestly say I don't totally understand what I am doing here so this may be totally wrong. Can someone help to straighten me out? As I said, the main goal of this project is the learning experience and not necessarily the end result, so I want to try and make it work this way, with the transistors, and not just wire it up based on another solution and call it a day.

Thanks to everyone who has helped!


EDIT: I also wanted to add, if I didn't make it totally clear before, I am very new to electronics and am probably jumping into something here a bit more difficult than I am ready for. I have played with the little electronics for kids type kits in years gone by, so I understand the basics of electricity, but a lot of the terminology and abbreviations I don't totally understand what they mean, so everything should be fully explained and nothing assumed. That probably sounds needy or pushy, but what I really mean is, you are all a lot smarter than me so don't hesitate to dumb it down to my level