Transistor Avalanche Disaster 

Tera Volt Even in the 80s a 2n3906 was an inexpensive transistor. They’re far cheaper today, of course, but even as a teen in the 80s I could easily afford to buy 2n3906 transistors in quantities of 100 without spending much.

In the 80s we could get 30 packs of transistors from Rat Shack for maybe $3. Which was more than they were worth but still not that expensive.

@@1pcfred And those parts from RS didn't meet spec...

@@Roy_Tellason why not?

@@1pcfred Breakdown voltage wasn't anywhere near the datasheet value for a 2N2222, for one example. Other characteristics didn't match up often either. I find that I'm doing better with parts salvaged from scrap equipment, though of course the idea of buying a bag of 100 2N3904s at around a penny each has some appeal, too. I don't think I'd want to buy them from China, though. Probably get 'em from Mouser, Digi-Key, one of those places...

Any circuit where a capacitor charges via a resistor, or discharges via a resistor by any significant voltage difference will have an exponential curve.

if you want a linear rise/sawtooth try a constant current supply. Unijunction transistor oscillators are fairly similar to Julian's design. Programable unijunction transistors can be programmed to breakdown at different breakdown voltages. down to about 1V. They do have a bit of reputation for being awkward to set up to oscillate/avoid them latching into conduction. I found it fairly easy to set the programming voltage from between a constant current module across a resistor to ground. and then then feed the timing capacitor from a slightly smaller constant current.

​@Mark1024MAK can you expand on this comment?

@@giornogiovanna729: you aware you used a racial slur there?

@@himselfe I would assume that your brain is made out of chinesium and other race-related metals.

himselfe Learn to take a joke.

@@asbestosfibers1325: you mean the guy that says racism doesn't exist and has never existed? Next you'll be telling me to Google Alex Jones and how 9/11 was a lizard false flag operation. 😂

@@cassidy5099: question is, why are _you_ getting butt-hurt over it?

Fortunately, failure with low voltage electronics is rarely dangerous. So fail often. Fail and learn :)

good idea.

That resistor and capacitor are forming the time constant for charging time so if you use 10M it will still oscillate but will produce only a short spike every few second or minute. Not sure if Jullian still have that Tek scope because I no longer watch his videos (kinda boring to me as always because of lacking technicalities) but it would have been nice to see how that oscillation waveform looked like. I know he wouldn't bother what is good and what is just below average type video making. At what voltage this reverse breakdown occurs or what would it be if we change the transistor or what voltage it stops will be your own imagination or experiment if you still need to know. Its just unbelievable that I unsubscribed his channel probably two years back and it still now remains the same quality as before. But yes I can not blame as he has got 160+K subscribers so peoples are enjoying what he is capable to deliver.

It is very saw tooth like. The voltage ramps up then snaps down quickly.

I may need the scope when I build the next stage - trying to get bridge rectifiers to operate as XOR gates :)

Julian Ilett bless your little cottons😁 I will be looking out for it sometime soon I hope. Reply appreciated 🏴󠁧󠁢󠁥󠁮󠁧󠁿👍

Yes, I will do more on that subject. I'm starting to enjoy these audio circuits more and more :)

Julian Ilett - in inverse parallel to the number of You Tubers putting their hands over their ears?🤣

This

The voltage on the emitter ramps up then snaps down. During that snap down action the base is about 0.,7V above the collector.

Throw in a mold and add clear pour epoxy or suitable medium and make a nice desk convo starter/paperweight! :D

search for "mini breadboard" on eBay :)

Mark W - ‘tis 😀

Not yet :)

they were featured in postbag 113

@@joinedupjon awesome thanks!

Nothing veiled on this channel - just wanted to see if it would work :)

Kerry D Wong tried 8 types and wrote it up on his blog - the NPN that oscillated at the lowest voltage was SS9018 @ 8.2V* which would therefore work off one fresh 9V battery. * though it varied between batches. but I think there must be types that oscillate at lower Voltage.

At a lower voltage probably - and energy is proportional to V²

He has loads of those Ruideng or Minghe CC/CV buck/boost converters, not sure why he's using PP3's instead of anyone of those attached to his solar battery. Even at this time of year, there'd surely be enough power left for that.

I don't have one - I do use a polyfuse when experimenting though and I'm more careful if moderately expensive parts are involved (who cares about one transistor?)... Then again I don't have an oscilloscope either. I've used an Arduino's analog input to graph waveforms in one project.

He got the oscilloscope from the factory free of charge. But none PSU factory sends him a lab power supply. :-)

No one wants to watch a dry lab procedural. Hijinx and banter and down-to-earth mistakes make good videos. Videos $ell. Julian makes circuits with batteries because Uncle Fumble makes circuits with batteries and goes on RU-vid to find out where his idea went wrong. Uncle Fumble doesn't have an oscilloscope, so Julian rarely brings it out.

Also a lab supply would not have saved his pile of transistors.

I've just designed a new audio breakout board of my own :)

Yes! They famously tortured their circuits too!

Great Stuff, Like it.

I was taught any circuit with a gain of more than +1 will osc. . . no matter how hard you try to stop it!

@@ihateemael That is only true when you are not trying to build an oscillator. If you try to make one, no amount of gain will cause oscillation.

I was just interested in the one transistor idea :)

It's in the transistor. It stays on as the voltage on the cap drops, to a point.

@@Ellipsis780 So then there must be more going on then the avalanche effect. The avalanche effect has no hysteresis according to the characteristic IV curves.

@@power-max avalanche does have a hysteresis. With current flowing, the carriers can bash more carriers lose. With no current, there is no bashing of new carriers. duckduckgo.com/?q=avalanche+diode+oscillator&t=hl&ia=web

appears 1 is just running the 'traditional' transistor oscillator 1x9 = 9... the other 2 are in series 2x9 = 18

Ollie B - £6😳 I got one for free when I bought something else 😀

It wasn't much of a party - there were no controls for premiere time - it just set itself to 2 minutes - weird :)

No.

Eddie - It’s the amount of current that flows when the transistor is reverse biased that kills it. What happens is, the capacitor charges up, as the transistor is not conducting (as it is reverse biased). But at some point, the voltage across the capacitor, which is the same voltage across the transistor, gets to the “break down” voltage of the transistor. The semiconductor junction in the transistor can no longer withstand the voltage. So suddenly it allows a high current to flow. The capacitor then dumps nearly all of the charge on it through the transistor (the transistor having “broken down”, it keeps passing current even when the voltage drops below the break down voltage). This current only stops when the capacitor is nearly empty of charge. If the peak current is too high, it will damage the semiconductor junction, hence killing the transistor. A suitable low value resistor in series between the transistor and the capacitor would have limited to peak current, and thus should have saved the transistor.

They're mini breadboards :)