Noise Reduction in high Gain circuits - Metal Film?

[mooreamps]

Measure what? The 3dB improvement in signal-to-noise ratio? Please detail your procedure.

Thermal noise is distinct from shot noise, which consists of additional current fluctuations that occur when a voltage is applied and a macroscopic current starts to flow. For the general case, the above definition applies to charge carriers in any type of conducting medium (e.g. ions in an electrolyte), not just resistors. It can be modeled by a voltage source representing the noise of the non-ideal resistor in series with an ideal noise free resistor.

The power spectral density, or voltage variance (mean square) per hertz of bandwidth, is given by

{v_{n}^2} = 4 k_B T R

where kB is Boltzmann's constant in joules per kelvin, T is the resistor's absolute temperature in kelvin, and R is the resistor value in ohms (Ω). Use this equation for quick calculation:

sqrt{{v_{n}^2}} = 0.13 sqrt{R} nV}/\sqrt{Hz}}.

For example, a 1 kΩ resistor at a temperature of 300 K has

sqrt{{v_{n}^2}} = sqrt{4 * 1.38 * 10^{-23} * {J}/{K} * 300K * 1 {k ohm} = 4.07 nV/\sqrt per Hz}}.

Also, if you get a 3dB improvement by halving the cathode resistor, does this mean that you get a 6dB improvement if you reduce it by a factor of four? Enquiring minds want to know!

Randall Aiken

Yes, by measuring the noise content on the signal itself, on an oscilloscope.

-g

[raiken]

Thermal noise is distinct from shot noise, which consists of additional current fluctuations that occur when a voltage is applied and a macroscopic current starts to flow. For the general case, the above definition applies to charge carriers in any type of conducting medium (e.g. ions in an electrolyte), not just resistors. It can be modeled by a voltage source representing the noise of the non-ideal resistor in series with an ideal noise free resistor.

The power spectral density, or voltage variance (mean square) per hertz of bandwidth, is given by

{v_{n}^2} = 4 k_B T R

where kB is Boltzmann's constant in joules per kelvin, T is the resistor's absolute temperature in kelvin, and R is the resistor value in ohms (Ω). Use this equation for quick calculation:

sqrt{{v_{n}^2}} = 0.13 sqrt{R} nV}/\sqrt{Hz}}.

For example, a 1 kΩ resistor at a temperature of 300 K has

sqrt{{v_{n}^2}} = sqrt{4 * 1.38 * 10^{-23} * {J}/{K} * 300K * 1 {k ohm} = 4.07 nV/\sqrt per Hz}}.

Okay, so you plagarized that verbatim from this Wikipedia link:

Johnson?Nyquist noise - Wikipedia, the free encyclopedia

That still didn't answer the question.

Besides, thermal noise in a resistor has absolutely nothing to do with the 3dB gain in signal-to-noise ratio you get from paralleling triodes. But then again, neither does halving the cathode resistor when you parallel them. You are clueless.

Randall Aiken

[ampman1961]

Greetings everyone, New member here and a big hello to Mr. Aiken.
Billy Yates

[MooreCowbelle]

Heh everybody! hey there Billy! I came over from yonder to see what all the dust storm was about - looks like Mr. Moore (no relation!) has got himself in it again. Let me tell you, it takes a LOT of BS to get good Mr. Aiken riled.

On topic: apply as much math and Phd. science to it as you want, but in the end it is mostly subjective. Experiment with various resistor types in various positions and use your ears to hear the difference.

[ampman1961]

Hello Mr. Cowbelle.
It's a wonderful subject that can be dealt with in a manner that doesn't require long mathematical equations & theory, although it helps quite a bit if you can speak and consume that language. Plagiarism is not an option here folks.
I'll compare this to two types of Navigation, Celestial, and Dead Reckoning.
Both can be quite accurate if used properly and will provide very useful results.
Properly being the operative word here. Celestial is a fine art that takes a certain amount of skill and experience. Dead Reckoning isn't quite as rigid and can be applied with minimal skill, as long as certain laws are adhered to.
Both can get you there but it still requires and extra component that is much more subjective...the human ear, which is by far my favorite tool.
I was never good with math.

What Mr. Cowbelle said is very true.
Trial and error can provide dramatic and oftentimes gratifying results if approached with some firm footing. Start off with some decent grade components and see which ones work best for YOU.
That was easy, and I didn't even cut and paste something from Wikipedia to explain it.
Billy Yates

[mooreamps]

Okay, so you plagarized that verbatim from this Wikipedia link:

Johnson?Nyquist noise - Wikipedia, the free encyclopedia

That still didn't answer the question.

Besides, thermal noise in a resistor has absolutely nothing to do with the 3dB gain in signal-to-noise ratio you get from paralleling triodes. But then again, neither does halving the cathode resistor when you parallel them. You are clueless.

Randall Aiken

But, that's how it's done. Therefore, with all due respect, sir ; perhaps I would recommend you prototype the circuit , and then take the measurements yourself.

very respectfully,
gary moore

[Robert M. Martinelli]

The effect paralleling devices has on noise is clear, but, as to the reasons behind this behavior, (with all due respect for other opinions), I agree with what Randall said (FWIW).

As he said, "The noise level is 3dB lower for paralleled triodes because the noise sources of the two triodes are uncorrelated and add as the square root of the sum of the squares of the individual noise sources, while the signal is correlated and adds directly, giving a 3dB improvement in signal-to-noise ratio".

This technique takes advantage of the white noise "random" behavior (that renders the noise sources "uncorrelated", using Randall's words).

The same technique can be (and is) used with other "active" devices (e.g. op-amps), and the noise reduction it allows is proportional to 1/(Sqrroot(number of devices)). This means that paralleling 2 devices yields a -3dB noise reduction, paralleling 3 of them yields a -4.7 dB noise reduction, and, quite obviously, paralleling four devices yields a -6 dB noise reduction ("halves" the noise).

JM2CW

Best regards

Bob

[pdf64]

Whatever the theoretical rationale behind it is, Gary's original assertion (that fixed bias is inherently less noisy than cathode bias) should be simple enough for others to replicate. I hope to give it a go over the next week and report back, it would be useful for others to try it out also.
My initial idea is to use an AA battery to supply the fixed grid bias, and adjust the dc plate supply so that the standard 100k plate, 1k5/22uF cathode, 1M grid ref have identical operating conditions. Feed in a small (10mV? f=?) signal from a 50ohm source, scope the plate signal, and compare the noise on it with the 2 bias methods.
Can anyone suggest a better way of going about this?

[Mach Schnell]

But, that's how it's done. Therefore, with all due respect, sir ; perhaps I would recommend you prototype the circuit , and then take the measurements yourself.

very respectfully,
gary moore

YGBFKM



It is better to remain silent and be thought a fool than to open one's mouth and remove all doubt.

[Jelle]

But, that's how it's done. Therefore, with all due respect, sir ; perhaps I would recommend you prototype the circuit , and then take the measurements yourself.

very respectfully,
gary moore

[ame="http://www.youtube.com/watch?v=3xC6oVvwtIo&feature=related"]YouTube- Mike Osborn Beer Fest gm-18 video clips #4[/ame]

[ame="http://www.youtube.com/watch?v=rSGryXLG3Ds&feature=related"]YouTube- Model gm-5 ; 5 watt Guitar Amplifer with Power Braking #1[/ame]

[raiken]

This technique takes advantage of the white noise "random" behavior (that renders the noise sources "uncorrelated", using Randall's words).

The same technique can be (and is) used with other "active" devices (e.g. op-amps), and the noise reduction it allows is proportional to 1/(Sqrroot(number of devices)). This means that paralleling 2 devices yields a -3dB noise reduction, paralleling 3 of them yields a -4.7 dB noise reduction, and, quite obviously, paralleling four devices yields a -6 dB noise reduction ("halves" the noise).

JM2CW

Best regards

Bob

Finally! Someone gets it!

Randall Aiken

[wizard333]

There should be a law against playing Little Wing in public unless you can do it right. That was yet another example of someone doing it very wrong.

[R.G.]

Finally! Someone gets it!

Randall Aiken

Randall - you'd be thought a visionary and perhaps a prophet if you just explained that the use of two devices aligned the electron flows and this caused an influx of quantum cosmic ray energy which resonated with the signal, giving a 15.2% push over the edge. Maybe put people on a waiting list for the newsletter which explained this.

[Steve Conner]

There should be a law against playing Little Wing in public unless you can do it right. That was yet another example of someone doing it very wrong.

Was that supposed to be Little Wing?! I didn't even recognize it, guess that's what happens when you do something abnormal like using fixed bias on your preamp tubes.

[raiken]

But, that's how it's done. Therefore, with all due respect, sir ; perhaps I would recommend you prototype the circuit , and then take the measurements yourself.

very respectfully,
gary moore

You're skirting the issue. Prototype what circuit? The point I'm raising is that the 3dB increase in signal-to-noise ratio is caused by the uncorrelated nature of the noise in the tube, not by your contention of halving the cathode resistor value when paralleling them.

In addition, in most circuits the cathode resistor is fully bypassed, so as a noise source, it is shorted to ground and contributes nothing.


Randall Aiken