Now, first off, to be clear, I am a complete amateur at this, so I am going off of vague hypothesis and someone else's original schematic. I was also, up to this point, trying to use parts salvaged off of other stuff, so not sure if the leads on them are even connecting on the test board, so I can test it (I just ordered replacements from digikey for the ones that may be dead/not connecting right.

Anyway, there are a mess of these LED fader with 555 timers out there, using what seems to be called and a-stable state. So, what I am aiming at is two things, for a Halloween rig, neither of which is part of *any* of the designed that are already out there:

1. I need two "eyes" that don't blink at all.
2. A super-bright led that fades, but not all the way to off. This is for a hollow mouth that will glow and pulse.
3. For testing purposes, use a potentiometer to adjust the fade timing. I will replace this, of course, with the right ohm resister, once I have it fading the way I want.

So, my hypothesis is - connect the "eyes" so they are not part of the main circuit, but on the same battery, and then parallel resistors in the actual fader, so that when "off" it will still glow, at a low level, and when "on" it will be fully bright (more or less). Here is the circuit layout I came up with for this:


So, am I right about this? According to an online resistance calculator, the result of using 3.3k ohm and 5.1k ohm should be "roughly" the original 2k ohm that the original had. But, I have no clue, at all, if this might somehow effect the timer/fader part of the circuit in some unpredictable way. And, until I get my new parts, I am not even sure why the thing isn't doing anything at all, but, I suspect the leads on both the transistor and the capacitor are so short they just are not contacting anything in the test board.

 

Pin 3 slaps high and low so you have that slowed down with the 100K pot into the 220uf capacitor. That gives you a maximum time of 24.2 seconds. Pins 2 and 6 are waiting for the voltage on the capacitor to change to either 1/3 of the power voltage or 2/3 of the power voltage, then they switch pin3 to whatever it isn't at the moment.

Problem: The 100K pot must never be adjusted to zero or things get messy. Maybe even smoke. Better put a 10K in series with the pot so it can never get faster than 2.4 seconds for a full cycle.

2K seems kind of dim at about 3 ma to the LEDs. Better put a 2K (or 2.2K) in series with each eye LED...or even 1K for each. You can't put two LEDs strictly in parallel because one of them will hog all the current. Feed them separately.

That's about all for now. I didn't check the 555 connections to see if it's really an oscillator.

 

Welcome to AAC!

First off, you shouldn't operate LEDS in parallel; invariably, one will draw more current than the other (s). That can cause a cascading failure.

UV LEDS don't emit much light in the visible spectrum, are you using something that flouresces?

Your timer circuit is unreadable. I don't care to memorize pin function on many devices... The pot connection to the timing cap seems odd.

 

Hmm. Ok. Pins should be right on the 555, since I went off of the setup from several existing designs, including ones that show them on youtube actually working, all of which seem to have the same basic layout (though some seem to use more pins, but they seem to actually flash, not fade on/off). Will double check that, of course.

Thanks for the advice on the pot and splitting the eye LEDs so they each have their own resister. Though, they seem to work fine without that now. Its the only part actually working at the moment, since its the rest of the components, other than the IC and resistors, which I have new ones of, that are iffy, having been salvages and having short leads. But, glad to know that I got the parallel resistor part right. Was afraid that doing that might mess with the timer, somehow. Not like I have like... any comprehension of this, beyond some limited knowledge from like.. 20 years ago, following instructions on one of those Radioshack kids kits. lol

 

Pins should be right on the 555, since I went off of the setup from several existing designs, including ones that show them on youtube actually working, all of which seem to have the same basic layout (though some seem to use more pins, but they seem to actually flash, not fade on/off). Will double check that, of course.

This is your "schematic" drawn in a more conventional manner:

EDIT: Need to redo my analysis. At first I thought the timer did nothing, but then I noticed trigger starts at ground and will trigger the one shot.

 

Look at the datasheet for the UV LEDs for the Forward Voltage (Vf). If it is less than 3 V (each), then you can put the two UV LEDs in series and cut the current draw for that section by 50%. Adjust the 2K resistor to get current to them:
(Assume 2.5 V LED Vf)
(Assume you want 10 mA through the LEDs)
9 V - 2.5 V - 2.5 V = 4 V -> this is the voltage across the resistor when the LEDs are on.
R = E / I -> R = 4 V / 0.01 A = 400 ohms -> use 390 or 430 ohms

ak

 

Need to redo my analysis.

When power is first applied, the one shot will be triggered. The voltage on C1 will rise exponentially until it starts turning on Q1 and will brighten the LED from it's dimmest level. When/if the voltage C1 rises to 2/3Vcc, the timer will toggle and C1 will discharge through Q1. When the voltage on C1 decreases to 1/3Vcc, the one shot will trigger again.

But the LED will never return to it's dimmest level as Q1 will always be conducting.

 

So, not wrong about my adjustment then.

Suppose will have to wait for my parts deliveries, to see how much it does "dim". I may still need to adjust the resistances so it stays brighter when in the dim state than it would be just with the timer in there. That was the main reason for paralleling the resistors, so I could adjust how "dim" it actually managed to get. Too dim and the effect would be less interesting than I want.

For purposes of anyone looking to reproduce this, might be useful to post a "corrected" schematic, which shows the extra resistor in series, as suggested by #12, to prevent accidentally dropping the resistance of the pot to 0, and, as he suggested, maybe doing bad things to some of the components.

But, thanks all. Seems like it should work like I need, once I actually have reliable parts to test it.

 

One thing I'm noticing about how you want the green LED to dim and brighten: Like a rising tide the green LED will grow brighter and brighter until it's as bright as it's going to get, then like the tide's ebb the green LED will grow dimmer and dimmer until it is as dim as it's going to get. Don't know why, but I got the impression you wanted the green LED to be (for the moment lets call it) full bright for a set period, let's call it 10 seconds, then you want it to dim out, but not all the way, then return to full brightness, spending only a second or two at its lowest level.

If not, then what you want is a light that brightens and dims in a linear fashion, kind of like a triangle wave. If this is what you want then you don't need a 555 timer. All you need is an RC circuit that rises to a trigger point then falls to a reset point, repeat. Let me see (if this is what you want) if I can find one.