looking for 1 transitor led flasher circuit..that works!!!!!

[Thomeeque]

edit: so like this then!...gave me an excuse to try a schematic to pdf!......

http://dl.dropbox.com/u/7464107/simplest%20led%20flasher.pdf

Simple, but requires two grounds!!

T.

[deadastronaut]

Rob I've got some of those 2222A's from bitsbox.
If you want to hold fire I'll give this circuit a quick breadboard tonight when I get home form work. 

nice one manny, i ended up ordering 10 off ebay instead late last night ....they're cheaper, but just have to wait a bit longer..i needed solder anyway...

but yeah that'll be cool if you can try it too....cheers .rob.

[deadastronaut]

edit: so like this then!...gave me an excuse to try a schematic to pdf!......

http://dl.dropbox.com/u/7464107/simplest%20led%20flasher.pdf

Simple, but requires two grounds!!

T.

2 grounds?...can you elaborate!...

[Thomeeque]

2 grounds?...can you elaborate!...

Sure!



Sorry, that was just a(stupid) joke really

Have a nice day, T.

[R.G.]

OK, this took me a while. I think that's happening is that the transistor is set up to break over the transistor's base-emitter junction and base-collector junction in series. The break over happens at the zener voltage of the B-E junction plus the forward junction of the B-C. Apparently for some transistors this so fills the broken B-E junction with charge carriers that it latches into avalanche mode and conducts until it sweeps out all the charge carriers. Perhaps the cap tossing in a lot of current helps keep it conducting until it is discharged. Normally the junction would just zener, and stay there, not discharging the cap. But the capacitor forcing through a lot of current makes for hot spots which keep conducting until the cap is discharged.

I think.  

I suspect - and what you've been posting - that the transistor's characteristics have a big effect on when, where and how it will go into avalanche oscillation. I also have some suspicions of the longevity of the transistor used like this. Could be fine though.

[deadastronaut]

@thomeque....i was gonna put 147 ground symbols on it..... i was on the whisky last night!...

@RG. hmmmm...so the tranny might take a kicking eventually then,..i'll try it out and see how it performs..and leave it on all day and night to test for longevity and any kind of randomness in light rate etc..

i tried simulating this in my circuit wizard programme...it didnt flash in  simulation....i guess theres nothing better than hands on real world breading for this!.

hopefully manny tries it tonight too...and we get a run down of behaviour...and led popping too lol...

though armando's worked ok too..for his bike!.. we shall see.....cheers rob.

[R.G.]

@RG. hmmmm...so the tranny might take a kicking eventually then,..i'll try it out and see how it performs..and leave it on all day and night to test for longevity and any kind of randomness in light rate etc..

I think the flash rate, etc. will be fine. I was just thinking about the fact that the current through the transistor is high and not well determined when it avalanches. The cap pours a current through that is dependent only on the internal resistances of the cap, the avalanched transistor, and the LED, none of which are known well. Could internally heat enough to "wear out". But maybe it's OK. I think a high resistor across the LED would change and possibly stabilize the flash rate. Try about 100K - 1M.

i tried simulating this in my circuit wizard programme...it didnt flash in  simulation....i guess theres nothing better than hands on real world breading for this!.

This is one of those places wherre circuit simulation is not all that great. Breakdown behaviour and especially things like avalanche behaviour are too complex for most models to include.

I thought the transistor would go into zener mode and stay there. It took me a while to tumble to it latching on by hot spot formation as a theory.

But once you find suitable transistors, they are probably cheap, so any wear out behaviour will be easy to fix if it ever happens. Buy two, keep a spare, like fuses. 

[arma61]

But once you find suitable transistors, they are probably cheap, so any wear out behaviour will be easy to fix if it ever happens. Buy two, keep a spare, like fuses. 

Hi

you talk about transistor "longevity", what about the cap, any "danger", I still have mine on the breadbord, I'll turn it on now and report back in 3-4 hours, it's located quite safe, it's on my balcony outside, so not big danger!      

[R.G.]

you talk about transistor "longevity", what about the cap, any "danger", I still have mine on the breadbord, I'll turn it on now and report back in 3-4 hours, it's located quite safe, it's on my balcony outside, so not big danger!      

Actually, I think that it will either work fine forever, or might over many months of use get hard to start or erratic. There's no "danger" as you know. In any case, the cure would be to pop in another transistor, possibly another cap. Not a big thing. The circuit's too simple and easy to replace to worry much about this. I was just noting what one might run into.

The avalanche mechanism was always described as "poorly understood" every time I read about it. That's what this circuit relies on - I think.

[Manny]

Hi All,

I just tried this with a couple of 2N2222A's and no luck I'm afraid.

I tried several of the transistors with hFEs of 205, 215 and 225. The cap was a 16v 470uF.
I used a 47K linear pot.

The LED didn't light up (but is confirmed to be working) and the voltage referenced to ground from the collector of Q1 varies from 0.2V to 0.4V depending on the setting of the pot.

I've had a quick read up on Avalanching but the information was rather vague, I'm still none the wiser of how this is supposed to work (need to read more theory ) Anyone got any ideas what's going on here?

[R.G.]

Anyone got any ideas what's going on here?

Yep. That would be this:

OK, this took me a while. I think that's happening is that the transistor is set up to break over the transistor's base-emitter junction and base-collector junction in series. The break over happens at the zener voltage of the B-E junction plus the forward junction of the B-C. Apparently for some transistors this so fills the broken B-E junction with charge carriers that it latches into avalanche mode and conducts until it sweeps out all the charge carriers. Perhaps the cap tossing in a lot of current helps keep it conducting until it is discharged. Normally the junction would just zener, and stay there, not discharging the cap. But the capacitor forcing through a lot of current makes for hot spots which keep conducting until the cap is discharged.
I think. 
I suspect - and what you've been posting - that the transistor's characteristics have a big effect on when, where and how it will go into avalanche oscillation. I also have some suspicions of the longevity of the transistor used like this. Could be fine though.

As I said, avalanche was poorly understood last time I read about it, and transistors vary a lot in how they do that. It's not a normally specified item for run-of-the-mill transistors. And transistors have gotten much better for what is considered normal operation since the 2N2222 was current. The "better" may include higher reverse breakdown for the base-emitter, and that could be all it takes for the circuit to not work.

Notice that for this to work, the voltage across the cap has to be bigger than the reverse breakdown voltage of the base-emitter, plus the base-collector junction, plus the LED start voltage, which I alluded to earlier with "put a resistor across the LED. As it sits I can see 1.5V to 4V on the LED, depending on color, 0.5V on the collector-base, and 5V and more on the reverse breakdown of the base-emitter. That can easily add up to more than 9V, and nothing will happen.

My comment on the resistor was aimed at pulling the LED voltage below it starting to nearly zero on the hope that you could start the base-emitter breakover and it might start the LED and initiate avalanche. It's a theory.

I know a PUT would work, as would a PNP/NPN setup to act like a PUT, but that last violates the "one transistor" objective.

In general, you have to have a breakover/recovery mechanism going on to force a relaxation oscillator like this to work. Avalanche mode is one way. So is the thyristor action of a PUT. Normal linear operation of a transistor, even operating reverse or as a zener for the base-emitter junction will never give a sustained oscillation. So the operation of this circuit is purely relying on stuff that is unspecified and accidental to the transistors.

[slacker]

Isn't the 2N2222 known for being a bit weird? I'm sure I've seen a few "quirky" designs that say they won't work with other transistors.

[arma61]

Hi there

still blinking at the same rate I left it 4 hours ago !

also checked the transitors, the one on the breaboard is a 2N2222A, while the old one I build years ago is a 2N2222 no letter at the end just an "U9" marked vertical on the body.

@Manny, may be you can lower the value of the pot, I have a 10K B on the breadboard now, and 22k on the old one,

[R.G.]

still blinking at the same rate I left it 4 hours ago !

I was more worried about four months. 

[deadastronaut]

Hi there

still blinking at the same rate I left it 4 hours ago !

also checked the transitors, the one on the breaboard is a 2N2222A, while the old one I build years ago is a 2N2222 no letter at the end just an "U9" marked vertical on the body.

@Manny, may be you can lower the value of the pot, I have a 10K B on the breadboard now, and 22k on the old one,

4 hours!... no fires or led popping then!..

great so it works,but it depends on the  value of the tranny i guess...maybe this is why manny's isnt having it!....hmmmm....

at least we know that it can be done!... (but carry a spare  )....

maybe its down to the cap voltage value too?...the tranny can take up to 40v unless i read it wrong!.....so maybe a 50v cap would be more suitable...even though its all running at 9v.....just thinking aloud!!!!!!....hmmmmm....

[merlinb]

By coincidence I happened to find this a couple of days before this thread started!  Funny how these things happen...

[deadastronaut]

coincidence indeed.. ..i notice thats running on 12v.... wonder what tranny that is though?..

[merlinb]

I just got this version to work! It didn't flash with any of the other transistors I tried (BC182/184/183/548). However, when I short out the LED I can see on the scope that the circuit will oscillate with many other NPNs, but as soon as the LED is added it stops, except with the BC337...

[deadastronaut]

great, , nice one merlin...what other npn  trannies will work with this..

i have 2n3904's 5088's....

[Gurner]

I'm curious - if the tranny is acting like a zener....then why not use a zener? (which bit have I not grasped!)