I'd like to build a circuit that will pulse a series of LEDs to obtain a twinkling christmas light effect. I found this circuit that I'm trying to build... specifically, v6. Unfortunately, it seems that it doesn't want to work for me. I've built it a couple of times, but only LEDs 4 and 5 actually light up, but do not pulse or fade. I also tried using the v7 of the circuit, but using individual LEDs rather than a multi-color. The difference is that v6 uses an LM339 while v7 uses an LM324. With this version, the LEDs all light up, but do not pulse. I'm not really sure what I'm doing wrong. I've built both circuits a couple of times now and achieved the same results over and over... so I'm either doing it right and it just doesn't work, or I'm just incompetant and I'm making the same mistake time and time again. Any help will be greatly appreciated. Oh yeah, I'm running this off the 5V rail of my PSU, so I'm within the theoretical voltage requirements. Also, the LEDs I'm testing this with are 2.2V@20mA cheapo greenies. Thanks
This should work I modified v7 a little and here is the result: http://pages.videotron.com/zwox/pictures/pulser.gif I ran a simulation, because I was bored, and it seems to work. I have added resistors to the leds. The led labeled LED1 is only a power ON led. Also, you might consider replacing C5 with 10nF to slow slow the pulsing effect. You say the leds are ON all the time, could it be that you have used smaller resistor or capacitor values than the ones in the original diagram? This might make the leds pulse so fast that they seem to be always lite. Let me know what happens, Zwox
Yeah, the resistors I used were a bit smaller as I didn't have the exact values handy, but according to the page, they're not really critical values, they just have to be between 1M and 10M. I'll try it with proper values tonight (i'll string a couple in series if necessary. As to your schematic, I'll give that a try tonight when I get home. Thanks for the help!!
Both circuits need led resistors, and to get any visible change in brightness use ones that will give a very low led current, say 5mA max. AFAICS the upper gates are slow saw-tooth oscillators, the bottom one is a fast saw-tooth. When compared by the 324 or 339, you get a PWM stream that (in theory) varies from 0-100% duty cycle and back down to 0% over the period of the slow wave. In practice, if the wave amplitudes aren't exactly the same, it will either dim to fully off for a bit or never quite get there. I can't see what the bottom-right inverter does here, or why that's the one with the variable resistor. In simulation, moving the pot just stops it oscillating. You may need low-leakage tantalum caps with the 2M+ resistors in the slow oscillators.
Well, I tried it with exact values last night and no dice. I also tried Zwox's circuit with similar results. It seems that only some of the LEDs light up, and I'm beginning to think that they're just flashing way too fast to be seen dimming. Grrr... ah well, I may just build the circuit I saw on your page, cpemma. Of course, I'll have to build a couple of them so that I can have multiple flashing LEDs, but with some variance in the throbs throught the pot. One other thing, anyone remember those circuits that used to be sold by 5fame?? It was a PWM LED pulser that could have 4 LEDs connected. Unfortunately, they're no longer in business it seems, but if I recall, they used a MIC502 or similar to achieve the pulsing. Any thoughts on this being a possibility??
I made a PWM throbber with an LM324, on a 12V supply you can run about 10-11V-worth of leds (5 reds or 3 blues) in series. 1 op-amp for the fast oscillator, 1 for the slow oscillator giving the sawtooth reference voltage, 3rd for the comparator. I used the 4th for a second comparator so as one led brightened another dimmed. It was going to be the Ultimate Led Throbber, but needs a few tweaks before it gets that title.
