Symmetric vs asymmetric noise source

Symetric noise source vs asymetric

Hi,
This is a symmetric noise source https://www.eeweb.com/extreme-circuits/symmetric-noise-source
And this https://www.learningelectronics.net/circuits/images/optimised+semiconductor+noise+sourcew.gif is it's asymetric version.
I wonder why one would need a symetric noise source, what are the pros?
The asymetric source produces 300mvpp but the symetric one only 150mvpp at the same consumption current as the articles state. The circuit values are the same so why is that difference?

Re: Symetric noise source vs asymetric

not sure what they mean by "symmetric". Maybe they mean it is not a clipped sinewave output...and therefore has less harmonic content?
But i am also not sure why this is a "noise source" as opposed to a simple analog oscillator? Does it put out a sine wave, or a true white noise output?

Re: Symetric noise source vs asymetric

biff44 said:

?But i am also not sure why this is a "noise source" as opposed to a simple analog oscillator? Does it put out a sine wave, or a true white noise output?

Click to expand...

Not an oscillator. It uses reverse biased transistors to generate noise. https://www.edaboard.com/showthread...-produced-by-reverse-NPN-Emitter-Base-voltage

Re: Symetric noise source vs asymetric

frankrose said:

Not an oscillator. It uses reverse biased transistors to generate noise. https://www.edaboard.com/showthread...-produced-by-reverse-NPN-Emitter-Base-voltage

Click to expand...

But what is this symmetric feature of the circuit? I haven't built it yet to see this on the scope. Is it maybe compared to the DC (i.e negative and positive noise)? And if so, why would one need such a feature?

Re: Symetric noise source vs asymetric

Good questions. I have no idea.

Negative noise? Positive noise? ...I don't think so. Even I don't know what are those.

This was a funny article at the 1st looking, then I couldn't identify my exact feelings. I felt some pain, but my knowledge is not enough to describe everything, especially the theory behind the statements, it can happen that every statements are true.... But take a look for this quote:

...the amplitude of the noise signal is asymmetric. This problem can be solved by using two transistors as two independent noise sources. / problem? but ok, independents, that is true I think.

One of these has a series resistor to earth, and the other has a series resistor to the supply line. / Still ok, supply and gnd referenced sources...however noise sources, so is it important?

Each of these noise sources produces an asymmetric noise voltage, with opposite asymmetry. / what? independents but they have opposite signs? ..sorry, opposite asymmetry? ..what does that mean for noise????

If these two voltages are combined, the amplitude of the result will be symmetric." / These are so demagogue statements, it describes noise as sum of voltages and I have no idea what is this asymmetric or symmetric noise. Google neither helped.

---

So sorry, I can't help you. My knowledge is not enough here.

Re: Symetric noise source vs asymetric

Hm... the answer is in this page https://www.elektormagazine.com/magazine/elektor-200012/16925
The waveforms are for this specific circuit compared to a non-symetric one.
Which one is the symetric and which the asymetric?

Now the question is, why on earth one would need a symetric noise source?

Re: Symetric noise source vs asymetric

I wouldn't say the "waveform" is not just an illustration. Yes, you cannot even separate which is which.
I found only one scientific article about symmetric and asymmetric noise, heavy physics content, I guess this circuit above is not related to its.
https://www.physik.uni-augsburg.de/theo1/hanggi/Papers/197.pdf

Re: Symetric noise source vs asymetric

My best guess is that they're assuming the amplitude of the noise is large enough that a small signal model of the BJTs is not accurate. If that's true, then the ideally gaussian PDF of the noise may be clipped or skewed to one side. The "symmetric" version of the circuit causes the distortion to affect both sides of the PDF equally. Note that even if the overall PDF is symmetric, it still won't be gaussian, so I'm not sure what the point is.

Re: Symetric noise source vs asymetric

looking at those two waveforms, i would clearly like the symmetrical one more.
The asymmetrical one looks clipped at the top...so if i were driving an analog circuit, and wanted to see things like harmonics and compression behavior in a high gain amplifier...the symmetrical one would give me a more "real life" test. It would equally hit both supply rails with the large signal output

Re: Symetric noise source vs asymetric

It is sad the article didn't mention any words you mentioned: clipping or Probability Density Function (PDF). Thus, asymmetric noise means asymmetric PDF of noise, where the spectrum is not white, and this is the problem.
BJTs have strong non-linear behaviour in this operation region, it causes distortion. In these arrangements, when the load resistors are above and below the BJTs the distortion is balanced between the output of the branches, it makes symmetric PDF and white, flat noise spectrum.

Re: Symetric noise source vs asymetric

frankrose said:

Thus, asymmetric noise means asymmetric PDF of noise, where the spectrum is not white, and this is the problem.

Click to expand...

Not necessarily true. Simply changing the PDF of the noise does not affect its PSD, which is what defines its "whiteness."