LED Chaser

Please explain, how did you did you do that, did you find this circuit somewhere? If not, how does it oscillate, I have tried and tried to make transistor oscillators myself, and just cannot figure it out???

@Issacsutt the trick to these oscillators is that the bias resistor is adjusted to approximately the emitter resistor X the transistor gain, + approximately 10%, this is just at the point of where the transistor will either turn fully on or fully off. The circuit is set so when one transistor starts to turn on, it will trigger the other transistor to turn off.

Multiple LED oscillators using transistors, more stable version; http://everycircuit.com/circuit/6107451243102208

I sent the reply to your link

I see what he did. If you take out all but just two transistors/RC Networks/LEDs, you will notice that he just took a classic astable multivibrator and expanded it.

This is correct...

I still have a hard time wrapping my head around that though. Maybe you can understand where I'm having trouble. Hurz has tried, but has no patience, every oscillator I've ever tried to build on EC will work the way I imagine it would at first, but then, no matter what I do, it always reaches equilibrium somewhere, instead of triggering itself to restart. Here's how I usually think about it... To generate a square wave, you have to have some kind of time delay, and the best way to do that is with an rc network (resistor and capacitor in series, input at resistor, output between R&C, capacitor tied to ground). So, if you could use this rc network to trigger a transistor to turn on at a specific voltage that the rc filter will hit in it's rise time, then this transistor could loop Back around and discharge the capacitor in a quick instant... then theoretically, that would cause the circuit to rise high in t=rc when you turn it on, and then the transistor would turn at at some voltage determined by maybe a voltage divider that would turn it on with about 0.75v when it hits, say a 2.5v threshold, then when it does, it would discharge the capacitor really quickly, and cause it to restart, and charge up again until it triggers the transistor to re-discharge itself once again; thus it oscillates and continues to do so till the end of time, or just whenever the source is drained or somebody interrupts it and turns it off to disrupt the beautiful cycle. But, in practice, it seems it's really not quite that simple, and it just doesn't work that way.🙁

I reflect on the fact that capacitors don't pass DC, only AC. The initial thing that happens is one transistor is slightly more biased to turn on greater than the other, as it turns on its emitter voltage goes a bit more negative than what it was, this change, going negative, is felt through the capacitor driving the base of the other transistor more negative which starts to turn it off more. The emitter voltage on this transistor increases a bit, this is felt through the capacitor of the other transistor with the effect of increasing the base current thus turning that transistor on more, this keeps on happening until one transistor is fully turned on and the Other one is fully turned off. Once the transistors reach this state, the capacitor on the transistor that is turned off can now start to charge positive through it's resistor, when the voltage gets high enough, (somewhere around .75V) that transistor starts to turn on, it's emitter starts to go more negative than what it was, and the whole story starts again. I think part of the success of the oscillator is the fact that the change in Voltage being felt through the capacitor is greater than the amount of voltage charge that is trying to come down thru the resistor, this works best when the biasing resistor is set to the ratio of emitter resistance X gain of the transistor, at this setup, before the oscillations begin the base current is also approximately the max emitter current devided by the transistor gain. Because the initial event starts from a state of rest, the base voltage/current swings are not as great and the first pulse is shorter in duration, but after that pulse because the transistors are fully changing from one state to the other, the negative pulse is gets larger and more driven, thus the pulse duration settles down to the RC time constant.

Why do you say: "it goes a bit more NEGATIVE than it was"?

Watch the blue trace on the attached circuit. When the led is off the emitter is around 7.9 volts and the red trace on base of the other transistor is less than 1 volt. When the blue transistor turns on, its emitter voltage goes to less than 1 volt, this is felt through the capacitor driving the base of the other transistor to -7 volts turning it hard off. But wait there is more, watch the red trace, it is slowly rising by current coming down through resistor, when it gets to around 0.863 volts, it snaps the red transistor on, it's change in Voltage is felt through the other capacitor and turns the blue transistor off. So the story continues. http://everycircuit.com/circuit/4979591824539648

Here we go, check this out. After so many attempts, I finally made a successful transistor oscillator: http://everycircuit.com/circuit/6110092480217088

It is simpler than that; http://everycircuit.com/circuit/6179322051428352

Like this: http://everycircuit.com/circuit/6110092480217088 not sure why it wont start without the switch though.

Make the last resistor 1k.

Oh thanks, I missed that some how

If you want to go that way, try this; http://everycircuit.com/circuit/4842937440796672

Your circuit still works nearly the same if you replace the inductor with a low value resistor such as 50 Ohms, so why use an inductor?

You can also just use a closed switch, the circuit seems to need some type of resistance (maybe for delay) I commonly use A small value inductor in these situations.

Oh ok, cool