Challenge Thread: The fastest breadboard oscillator on the mudball

[BreakingOhmsLaw]

That is total chaos! I built the circuit with 3 x BC547c and wow! I have to raise the supply from 4V to about 4.8V before the world turns crazy but then the reward is big.
Also checked on the SA, there the baseline just rises well up to 200MHz.

That would make a nice source for a fairly random RNG.
I've tried reproducing the noise source from 'The Art of Electronics' where they use the parasitic Zener in a BJT but it would not work very well. I am guessing it needs to be a specific make and model of that BJT. Transistors may be built up in a completely different way depending on manufacturer and year of production, as Richi demonstrates on his great website:

https://www.richis-lab.de/2N3055.htm

[herc]

there is another danger in using a transistor as a zener replacement: they are not designed for that and start to age / degrade quickly.
what was a nice hardware RNG, might turn - unnoticed - into a non-random something. there is a paper about it:

[SilverSolder]

yes, interesting, isnt it ?

and by intuition i guess that this oscillator - though it needs more parts - has a much higher output power than the single transistor designs. what do you think ?

It is possible, but it would just be a guess.   There is more gain in the circuit overall so the amplitude might be higher, but other factors could cancel that out.  Build and test! 

[SilverSolder]

there is another danger in using a transistor as a zener replacement: they are not designed for that and start to age / degrade quickly.
what was a nice hardware RNG, might turn - unnoticed - into a non-random something. there is a paper about it:

I recently took apart a piece of equipment that used a matched pair of MAT-01 transistors (two matched NPN transistors in a metal can) mounted in a solid block of aluminum to stabilize the temperature, where one of them was used as a zener.  The equipment was probably 30 years old but unlikely to have been in constant operation - but it had definitely seen many hours of use.

Just for fun, I decided to measure the gain of the two transistors.   The one that had operated in zener mode had a gain that was less than 1/3 of its "partner" transistor...   so yes,  the "destructive effect" is real and easily measurable.

[herc]

you can simplify the multivibrator oscillator quite a bit: you dont need to feed the coil from the middle, and you dont need the coupling capacitors.

this is called the "Ruck-Zuck Oszillator" published by german ham operator Eugen Berberich in Funkamateur 10/2005. It is actually a VCO that works over wide frequency ranges by simply changing the supply voltage.

[ Specified attachment is not available ]

[SilverSolder]

you can simplify the multivibrator oscillator quite a bit: you dont need to feed the coil from the middle, and you dont need the coupling capacitors.

this is called the "Ruck-Zuck Oszillator" published by german ham operator Eugen Berberich in Funkamateur 10/2005. It is actually a VCO that works over wide frequency ranges by simply changing the supply voltage.

(Attachment Link)

That looks like a fun little circuit.  What does "ruck-zuck" translate to in English?

[herc]

"Ruck-Zuck" means "very quick" , "in a jiffy", "in no time (at all)"

[herc]

it was really Ruck-Zuck !
seems to have a lot of harmonics ...

small loop: around 500 Mhz, rather strong harmonics > 1.5 GHz

[SilverSolder]

Yes it looks like it is making square waves, LOL! 

[NiHaoMike]

you can simplify the multivibrator oscillator quite a bit: you dont need to feed the coil from the middle, and you dont need the coupling capacitors.

this is called the "Ruck-Zuck Oszillator" published by german ham operator Eugen Berberich in Funkamateur 10/2005. It is actually a VCO that works over wide frequency ranges by simply changing the supply voltage.

(Attachment Link)

Feed the coil from the middle with a current limited supply and you won't need the resistors. I remember reading about that variant of the circuit (plus a reverse diode for protection) operating in the FM band as a simple phone line tap that connected in series with the line.

[herc]

with a shortcut instead of a loop, the frequency shot up to 1.2 GHz

[0culus]

I finally received my 38 awg magnet wire and some rather high speed parts from digikey...I'll be experimenting and getting back in the game soon. 

[SilverSolder]

with a shortcut instead of a loop, the frequency shot up to 1.2 GHz

 

What transistors are you using for this?

[herc]

those transistors are BFR 93A from NXP ( 6 GHz) :  https://www.nxp.com/docs/en/data-sheet/BFR93A.pdf

i wonder how far up the harmonics can go ?

[0culus]

I finally received my 38 awg magnet wire and some rather high speed parts from digikey...I'll be experimenting and getting back in the game soon. 

Nevermind. The parts are too small for me to solder. Already ruined 5 transistors. Lucky they were dirt cheap.

[herc]

Nevermind. The parts are too small for me to solder. Already ruined 5 transistors. Lucky they were dirt cheap.

what did you order ? what size are they ?
i found SOT23 with 1.9mm pin distance is probably the limit for me. i used sticky tape to somehow fix the transistor before soldering.

[1uk3]

Nevermind. The parts are too small for me to solder. Already ruined 5 transistors. Lucky they were dirt cheap.

I recently had to solder magnet wire to a 40pin 0.5mm pitch IC (I messed up the footprint  )

Just tried the circuit with a pair of BC547 and can't get over 180MHz 

[SilverSolder]

Nevermind. The parts are too small for me to solder. Already ruined 5 transistors. Lucky they were dirt cheap.

I recently had to solder magnet wire to a 40pin 0.5mm pitch IC (I messed up the footprint  )

Just tried the circuit with a pair of BC547 and can't get over 180MHz 

I once managed to solder wire strands on to an 0.5mm pitch BGA package.   We need a new contest, "Dumbest and most time wasting hand soldering attempt!"

[0culus]

Nevermind. The parts are too small for me to solder. Already ruined 5 transistors. Lucky they were dirt cheap.

what did you order ? what size are they ?
i found SOT23 with 1.9mm pin distance is probably the limit for me. i used sticky tape to somehow fix the transistor before soldering.

I ordered both 18 and 22 GHz RF transistors. I forget the package, but they are tiny. I was able to develop a foolproof method of getting the soldering done. The trick is you must do all the legs at the same time or you'll just be fighting the wires fall off as you heat it. These have 4 legs (two are emitters) so I carefully wrapped the magnet wire across two of them and then again after filing off the enamel. This allowed me to get them soldered all at the same time and then snip the bridge. That said, I haven't managed to get anything oscillating yet.

[mnementh]

Nevermind. The parts are too small for me to solder. Already ruined 5 transistors. Lucky they were dirt cheap.

I recently had to solder magnet wire to a 40pin 0.5mm pitch IC (I messed up the footprint  )

Just tried the circuit with a pair of BC547 and can't get over 180MHz 

I once managed to solder wire strands on to an 0.5mm pitch BGA package.   We need a new contest, "Dumbest and most time wasting hand soldering attempt!"

Suck what...? That sort of soldering is what I do for FUN...

mnem
 

[BreakingOhmsLaw]

Suck what...? That sort of soldering is what I do for FUN...
mnem
 

If Louis Rossmann ever comes across this thread, he is going to curbstomp all of you 

[TheUnnamedNewbie]

another neat oscillator:

http://www.totalitaer.de/Rftechnik/mikrowellensender2.htm

(Attachment Link)

That's funny, I was just thinking if a multivibrator type circuit could be made to work at RF frequencies...   Looks like the answer is "Yes"!

You can use them up to 100's of GHz even, on-chip.

[SilverSolder]

Nevermind. The parts are too small for me to solder. Already ruined 5 transistors. Lucky they were dirt cheap.

I recently had to solder magnet wire to a 40pin 0.5mm pitch IC (I messed up the footprint  )

Just tried the circuit with a pair of BC547 and can't get over 180MHz 

I once managed to solder wire strands on to an 0.5mm pitch BGA package.   We need a new contest, "Dumbest and most time wasting hand soldering attempt!"

Suck what...? That sort of soldering is what I do for FUN...

mnem
 

0.5mm BGA being soldered to an adaptor.   Would have been easier to get the right adaptor in the first place! 

The wires are strands of fine desoldering braid...

[joeqsmith]

While I have no reason to suspect foul play on Joes part, I ask him to provide clear images while running and showing the counter for the sake of fairness.

I believe that Joe is where he is now, because he has shown good knowledge of RF engineering and particularly how to deal with parasitics (thinning wires, shielding etc.)

That said, i have parts on order, and i know i am not the only one. First to dethrone Joe shall be known as the kingslayer from hence forth!

Looking forward to it!     


I've been enjoying this book from George Southworth.  For the few of you who are actually interested in learning something, I recommend it.
https://worldradiohistory.com/BOOKSHELF-ARH/Technology/Fourty-Years-of-Radio-Research-Southworth-1962.pdf

[mnementh]

   0.5mm BGA being soldered to an adaptor.   Would have been easier to get the right adaptor in the first place!     The wires are strands of fine desoldering braid...

    That kind of soldering I do NOT do for fun; I ONLY do it when I'm being well-paid by the hour, NOT flat-rate. 

mnem
*tsssssssst...*