Where's my voltage gain?!

[raiken]

To calculate the gain, substitude 1.7K for 1.7 volts ; and 1K for 1 volt. Also, if you should find any of this published in his books later on, you will now know where it came from.

-g

With all due respect, if you find that published in any book, you need to get another book! This is completely wrong. You cannot calculate gain in this manner for a vacuum tube.

RA

[R.G.]

With all due respect, if you find that published in any book, you need to get another book! This is completely wrong. You cannot calculate gain in this manner for a vacuum tube.
RA

Randall, it may be that mooreamps is onto something fundamentally new, and that math and physics don't work the same for him as for other people.

I think the right thing to do is ask him to document the math and physics behind what he says. Maybe there's something subtle there that we haven't seen before. I know this stuff is not rocket science, but still, there might be something.

What'a'ya say, moore old boy? Enlighten us benighted thirsters for knowledge. Tell us how you got to there.

[Old Tele man]

...and, the answer is: ???

[mooreamps]

Randall, it may be that mooreamps is onto something fundamentally new, and that math and physics don't work the same for him as for other people.

I think the right thing to do is ask him to document the math and physics behind what he says. Maybe there's something subtle there that we haven't seen before. I know this stuff is not rocket science, but still, there might be something.

What'a'ya say, moore old boy? Enlighten us benighted thirsters for knowledge. Tell us how you got to there.

I thought I already did, in the above posts. It's really nothing more than what I already posted. I also posted on another thread my formula for calculating the first order noise figure on a first input gain stage ; figuring the thermal noise contribution of the cathode resistor multiplied by the input input impedance. I try to share these finding, but these comments from Mr. R. don't help. Maybe he's not happy I am going in so deeply into the amplifier theory. I wasn't trying to. I just wanted to start building tube amps. Now I've power brakes and noise brakes, and fixed bias preamp gain stages, double-bridge dc rectifiers, :| Not what I was intending to do.....



But the math and physics are elementary correct, and fully applicable. If they were not, my colleges at work would have said something when it was presented to them. I do plan to write another paper on this, and submit it to Space Systems Company, since the electrical theory should apply the same to a FET based amplifier. I do this because we bias a FET gain stage in the same manner as a vacuum tube. Hense, a parametric LNA that can be used inside a spacecraft with just an exceptional input receive low noise figure.

I will say the theory is already based on other doctoral papers I've read written by others on the electrical thermal noise contribution of diodes. But I don't believe anyone had drawn the line linking cathode (or drain if you will) resistor noise figure as a function of input impedance.

-g

[R.G.]

I thought I already did, in the above posts. It's really nothing more than what I already posted. ... Maybe he's not happy I am going in so deeply into the amplifier theory. I wasn't trying to. ...
But the math and physics are elementary correct, and fully applicable. ... I will say the theory is already based on other doctoral papers I've read written by others on the electrical thermal noise contribution of diodes. But I don't believe anyone had drawn the line linking cathode (or drain if you will) resistor noise figure as a function of input impedance.

Cool! Lots of neato words!

We're all ears. 'Splain 'dis to us. Like I said, I always did like to learn something new.

So tell us *exactly* how Randall is wrong and you're right. Please use numbers and equations. I may not be too good with rocket science math, but I could manage to keep up with the math in good old simplistic, falling-off-a-log EE, so maybe there's hope for me to understand.

We're waiting. I have my Pentel 0.5mm and quadrule pad all warmed up and ready. Lead us through it.

Be gentle.

I almost forgot. Back at work, we always revered guys who had a deep enough understanding of the principles involved to be able to lead us through the theory and through the math and numbers to the bottom of a problem. These fellows were, in our vernacular, able to follow the concepts through the theory, through the equations, through the numbers, and right down to the bare metal. I did so admire someone who could do that. I'm hoping you'll be able to share with us and lead us from the concept right down to the bare metal.

[Merlinb]

I thought I already did, in the above posts. It's really nothing more than what I already posted. I also posted on another thread my formula for calculating the first order noise figure on a first input gain stage ;
I do plan to write another paper on this, and submit it to Space Systems Company,
But I don't believe anyone had drawn the line linking cathode (or drain if you will) resistor noise figure as a function of input impedance.

Ah, you mean this post:
Critique My Schematic
Or maybe this thread:
http://music-electronics-forum.com/t16677/

The problem is, Gary, your theory appears to be deeply flawed, at least with regard to valves (I dunno, maybe you're a whizz with BJTs and you're confusing the two). Let me help you:

Noise in triodes included noise due to grid current, but this decreases with negative bias and is entirely negligible compared with the shot and flicker noise in the anode current, at least over the audio bandwidth and provided there are no extraordinarily high source resistances (I mean tens of megohms).

The shot noise current (squared) can be expressed as:
i(shot)^2 = 2.576.k.Tk.gm.df
Where:
k = Boltzmann’s constant (1.38*10^-23)
Tk = cathode temperature (about 1000 Kelvin)
gm = transconductance
df = bandwidth

The flicker noise current (squared) can be expressed:
i(flicker)^2 =K.(Ia^2 / f). df
Where:
K= 10^-13 to 10^-12 for most preamp triodes
Ia = Anode current

These two noise currents can be represented as one noise voltage at the input by dividing by gm^2:
v(tube noise)^2 = [(2.576.k.T)/gm + K. Ia^2 / ( f .gm^2)] df

Now, the source resistance Rs (e.g, grid leak) also contributes Johnson noise to the input:
v(source noise)^2 = 4.k.T.Rs.df

And so does an unbypassed cathode resistor:
v(cathode resistor)^2 = 4.k.T.Rk.df
Where:
T = Ambient temperature (say 300K).

Let’s lump it all together as one noise voltage source:
v(noise)^2= [(3.55*10^-20)/gm + K. Ia^2 / ( f .gm^2) + 4.k.T(Rs+Rk)] df

As an example, a 12AX7 at 1kHz with Ia = 1mA, gm = 1.5mA/V, Rs = 100k, Rk = 1k, T = 300 kelvin, df = 20kHz

v(noise)^2= [2.37*10^-17 + 4.44*10^-16 + 1.66*10-20*(100000+1000)] * 20000
v(noise)^2= [2.37*10^-17 + 4.44*10^-16 + 1.67*10-15] * 20000
v(noise)^2= 2.14*10^-15 * 20000
v(noise)^2= 4.28*10^-11
Implying 6.54uV of Johnson, flicker and shot noise at the grid of the tube. Of all this noise, 78% is due to the source resistance and only 0.78% is due to the unbypassed cathode resistor!

Gary, I think your paper might need revising.

[R.G.]

Merlin, I think you got some metal showing down there! Good roundup!

[raiken]

If they were not, my colleges at work would have said something when it was presented to them. I do plan to write another paper on this, and submit it to Space Systems Company...
-g

Perhaps your "colleges at work" were either too polite, or too busy laughing, to say anything when you presented your half-baked theory?

I once had a "green" engineer working for me who reminded me a lot of you. He was a fresh Auburn grad who thought he knew everything, so he was always coming up with some off-the-wall "theory" that was only known to him, and he couldn't understand why everyone else had missed it. My favorite was the time he came to me with a plan to save the company a lot of money by replacing all the inductors in our products with series RC networks. He said that he had figured out that you could get the same phase shift with an RC that you would get with an RL, and capacitors were so much cheaper than inductors, it would save us a fortune. He claimed he had presented this theory to all his EE professors at Auburn and none of them could dispute it. Two things happened that day: (1) I lost all respect for Auburn's EE professorial staff, and (2) I fired the guy.

Here's some advice: If you value your (alleged) job at this Space place, don't write a paper on this subject. And if you decide to write one anyway, at least use a spell checker - it's "colleagues" not "colleges".

RA

[JHow]

...at least use a spell checker - it's "colleagues" not "colleges"[RA

Alas, a spell-checker won't catch that sort of mistake. It just looks for reconizable words - it doesn't consider context.

their / there / they're
affect /effect
its / it's

[raiken]

Alas, a spell-checker won't catch that sort of mistake. It just looks for reconizable words - it doesn't consider context.

their / there / they're
affect /effect
its / it's

Good point! Perhaps a proofreader is in order...

RA

[R.G.]

I once had a "green" engineer working for me who reminded me a lot of you. He was a fresh Auburn grad who thought he knew everything, so he was always coming up with some off-the-wall "theory" that was only known to him, and he couldn't understand why everyone else had missed it.

That story reminds me of a saying you hear sometimes here in Texas - "All hat and no cattle."

I am baffled as to why someone would make a technical pronouncement that they can't at least begin to back up with references outside their own head. Especially when it's obvious that the forum for the pronouncements contains people who have current, pertinent technical backgrounds and who are not the least inhibited about asking for technical backup on things like that.

I wish they wouldn't do that. I feel embarrassed for them, and frankly think they must have some kind of backup they haven't stated. So I always ask for more info. Sometimes I get it, sometimes I don't.

[Enzo]

Throughout my "career" I have always enjoyed the little anecdotes in the trade magazines. Like the last page in each issue of EDN. I don;t recall the details, but I remember one story where the senior engineer had the junior tech come into his office to show off his new plan to make their industrial process much more efficient. He asked the senior engineer's thoughts on the improvements proposed. After a moment, "Well, this 18 degree F cooling water might be a bit hard on the pumps, don't you think?" It stuck with me.

[raiken]

I'll never forget my "green days" as a fresh-out-of-college engineer. I graduated thinking I knew my stuff, and one day I was assigned a problem to fix an overheating 7805 type regulator on a production board. I measured the heat and quickly found a nice sized heatsink to leep it cool, and proudly brought my solution to the engineering manager. Another grizzled old engineer was in there and he did a quick calculation and said, "no way that regulator should be dissipating that much heat". He then proceeds to go out to the floor and look at the layout and said, "Oh, here's the problem, the output of that 5V regulator is shorted to this other regulator output. It doesn't need a heatsink. Fix the problem, don't mask it".

I felt like an idiot, but that lesson stuck with me all my engineering career.

I always try to make sure I have some sound technical backing before jumping to conclusions or proclaiming theories, and I'm not afraid to ask others for assistance if necessary. There is no shame in admitting ignorance, everyone has their strengths and weaknesses. You can either learn from your mistakes, or repeat them and look like a fool. There is nothing wrong with standing firm on your convictions, but you'd better be ready and able to back them up when subjected to peer review, and you'd better be able to humbly admit it when proven wrong. Ignorance and arrogance are a particularly bad combination.

RA

[vintagekiki]

I'm designing/building a pre-amp, and the 1st gain stage is using the triode of a ecf82/6u8-a, as used in V1 of the normal channel of the Vox AC10.

Complete schematics or link of your building please. If it is possible to specify the actual measured value tubes DC voltage (Ua, Uk etc)

[raiken]

Complete schematics or link of your building please. If it is possible to specify the actual measured value tubes DC voltage (Ua, Uk etc)

Here's the AC10 he references, normal channel, V1, but with LED biasing:

http://www.schematicheaven.com/voxamps/ac10_2.pdf