Why some amps use two triode in parallel?

[Alan0354]

Electrometer amplifiers are ‘Transimpedance’ amplifiers with a high value feedback resistor which is the dominant noise source. If you double the value of the feedback resistor the output voltage doubles but the noise only goes up by root 2 so you’ve increased the signal to noise ratio by 3dB. It is just another example of Merlin’s rule of thumb.

No we talk about two amps, not increase the gain of the single amp. Yes, transimpedance is just a big resistor with virtue ground. It's just that simple!!! But we( the industry) gone through all that. Bigger resistor slows the speed!!! It's all about compromise of speed, noise, sensitivity.....and last but not the least, $$$$$.

Also, transimpedance amp has a different way of calculating noise gain. You cannot even say in a blanket statement about increasing the resistor and get 3dB. The input capacitance with the feedback resistor forms the feed back gain. The higher the resistor, the higher the closed loop gain. The input capacitance become the gain setting element. So it is not that simple. This will be another different branch of material. If you care to, I can dig up some article for you in how to calculate noise gain for these transimpedance amps as this it way off of what we are talking. But the info is out there and I studied it. I designed a lot of transimpedance amp from detection of 100fA to over 500MHz amp using GaAs dual gate MESFET.

[Merlinb]

I thought I made it very clear that parallel the triode does not double the gain. All my calculations were based on gain=54 for single triode and 78 for parallel. Again, read my posts.

I think this whole discussion is dragging simply because (I think) English is your second language, so we are misunderstanding each other.

For voltage amp like you show, you can gain S/N like that. but the tricky part is the summing that you just said it happens. If it is tube amps, the summing part become tricky. You cannot sum two voltage and double the voltage!!!

Yes, I know. Let me say one last time: The 3dB rule of thumb for parallel amplifiers applies to the theoretically ideal case (in English scientific writing we use the word 'ideal' to mean 'absolutely perfect in every way'). We all know you will never achieve this in real life (but you can get close with some circuits). That is why my example had to contain 'ideal buffers', to prove the point of the theoretically ideal mathematics. Perfect buffers/mixers do not exist in reality, which is why in reality you would never achieve the 3dB imporvement.

[Dave H]

No we talk about two amps, not increase the gain of the single amp. Yes, transimpedance is just a big resistor with virtue ground. It's just that simple!!! But we( the industry) gone through all that. Bigger resistor slows the speed!!! It's all about compromise of speed, noise, sensitivity.....and last but not the least, $$$$$.

It was only an example to illustrate the 3dB rule. It was for a mass spectrometer. Speed didn’t matter. It's always a compromise. You don’t get owt for nowt. I did get a 3dB improvement but had to double the supply voltage.

There is a parallel amplifier example on page 11 of the pdf below.

an21f-2.pdf

[Alan0354]

It was only an example to illustrate the 3dB rule. It was for a mass spectrometer. Speed didn’t matter. It's always a compromise. You don’t get owt for nowt. I did get a 3dB improvement but had to double the supply voltage.

There is a parallel amplifier example on page 11 of the pdf below.

[ATTACH]27968[/ATTACH]

Mass Spectrometer is my field of experty. I spent 12 years in this very field and was the manager for 10 years out of this. I designed all the circuits in our TOF, Dynamic SIMS etal. That's where I designed all sort of transimpedance amps from femto amp electrometer amps to lazar diode amps. From a few hundred Hz to over 500MHz. The company is called Charles Evans and Associate, then sold to Physical Electronics of Minisota and later became Revera Inc. We had transimpedance amp of all different kinds from using over 2Gohms resistor to using Dual Gate MESFET to design discrete opamp.

Saying frequency is not important is absolutely wrong. It is EVERYTHING. Read about NOISE GAIN for transimpedance amps.

[Alan0354]

I think this whole discussion is dragging simply because (I think) English is your second language, so we are misunderstanding each other.


Yes, I know. Let me say one last time: The 3dB rule of thumb for parallel amplifiers applies to the theoretically ideal case (in English scientific writing we use the word 'ideal' to mean 'absolutely perfect in every way'). We all know you will never achieve this in real life (but you can get close with some circuits). That is why my example had to contain 'ideal buffers', to prove the point of the theoretically ideal mathematics. Perfect buffers/mixers do not exist in reality, which is why in reality you would never achieve the 3dB imporvement.

Please don't get personal about my English. We disagreed, it is a heated argument, but I never stood this low to start going outside of technical argument.

I understand exactly what you said. I am not interested in "IDEAL" argument that serves limited practical use. I prove my case and you can keep your IDEAL parallel amps.

[g1]

You are going to have to choose if you want to talk about theoretical or practical. You can't go back and forth as it suits you. Most theory is about ideal situations, what happens in practice is no longer theoretical.

[Alan0354]

No theory is about making things work. What is the point of making a theory that dose not serve real life use. I never go back and fore, I consistently disagree with this "ideal theory".

And what does this ideal theory help in the parallel triodes? So far, the parallel triode is loosing big.

[Dave H]

Saying frequency is not important is absolutely wrong. It is EVERYTHING. Read about NOISE GAIN for transimpedance amps.

You are taking what I said out of context again. The FP doesn't have to work at RF and frequency response isn't everything in THIS APPLICATION. I used a 10G resistor and it was fast enough. I even tried a 50G. It was lower noise but not quite fast enough.

[Alan0354]

You are taking what I said out of context again. The FP doesn't have to work at RF and frequency response isn't everything in THIS APPLICATION. I used a 10G resistor and it was fast enough. I even tried a 50G. It was lower noise but not quite fast enough.

As I said, double the resistance is not parallel two amps. I don't even see the connection. Of cause larger the resistor, the better the noise advantage on the first pass. That's why Victoreen makes 10ee11 ohms resistor

[Enzo]

Ideal theory tells us how something ought to work in the absence of other influences. By studying the ways the practical results vary from the ideal, we learn more about the systems overall. THAT is the point of it.

Alan, I don;t for one second believe Merlin was criticizing your English. I believe he was saying that some of the nuance in the technical discussion back and forth is lost due to the language difference. So sometimes a statement is made by one, and not completely understood by the other, but the discussion continues assuming there WAS complete understanding. I do think he was in part telling you that the two of you might agree more than you think you do.

[Dave H]

As I said, double the resistance is not parallel two amps. I don't even see the connection.

The connection is that they both reduce noise by the same principle. The signal is doubled but the statistical nature of the noise means it only increases by root 2

[Alan0354]

Ideal theory tells us how something ought to work in the absence of other influences. By studying the ways the practical results vary from the ideal, we learn more about the systems overall. THAT is the point of it.

Alan, I don;t for one second believe Merlin was criticizing your English. I believe he was saying that some of the nuance in the technical discussion back and forth is lost due to the language difference. So sometimes a statement is made by one, and not completely understood by the other, but the discussion continues assuming there WAS complete understanding. I do think he was in part telling you that the two of you might agree more than you think you do.

OK, I tone down my post a little. I am not really interested in this theory debate anymore. For one, take up too much of my time. My amp is calling me and complain because I ignored her the last two days.

[jazbo8]

man, what happened here? perhaps a case of not see the forest for the trees... please don't take it the wrong way, just trying to lighten the mood

Alan said: "And when you say -3db, at best is ideal, but this is real life circuit that you have to consider all the contributing sources."

I think no one is saying otherwise, Merlin & Gingertube said all along (paraphasing here) - when you parallel two identical ideal amplifiers with un-correlated noise figures (consisting of the sum of the noises sources that you have shown), the S/N ratio is improved by 3dB (2x). This improvement is for the ideal amplifiers only, and does not factor in the other components in the circuit. The amplifier can be valves, BJT, MOSFET, or opamps but the general rule applies.

OTOH, you are correct to point out that some builders mis-apply the above rule and/or some manufacturers were stupid enough to use it as a selling point and spread the myth of "parallel triodes for low-noise design".

[Alan0354]

Let's just agree to completely disagree. I understand completely the argument that the parallel amps are a theoretical ideal not taking into consideration of external effect in the real world, that it does not matter it won't work in most real world cases. Let me just close with this analogy:



Communism and Socialism is a Utopia. When people share, no body left behind, everyone enjoy life and live happily ever after. That is an IDEAL theory. You ignore a few minor details considered to be external effect:
1) Ego.
2) Greed.
3) Power grabbing.
4) Laziness.
BUT, if you ignore these minor external effects, Communism and Socialism really work!!!! Read Animal Farm and you get the idea.

Look at the real world history, because of the minor external effect that people in power want to stay in power, massacre of millions of intellectuals in Starling, Mow, Paupot............ Hunger wide spread, re-education camp( say death camp)...............

Then look at Europe, Britain and German were one of the proud and strongest country till the WWII days. German almost conquered the world with their innovations.....jets, atomic bomb........Britain had to transfer their most state of art technologies to US to develop because of the instability during WWII. Now somehow, they got into the idea of socialism, what are they good for now? Other than Mercedes and BMW that are selling because of the name associate with prestige, what product they can claim to be on top? The whole Europe is in financial crisis and guess why? It's the ideal theory of Socialism!!!

Then look at us, the United State of America. FDR made us the strongest in the world, we had Kennedy brought us to the moon. We had Reagan defeat the Soviet and won the Cold War. Then what happened since? The new generation grew up not experiencing the evil of Communism and Socialism, starting to admire the Utopia ( theory) of Communism and Socialism.........that people can all be equal, fair, happy, no hunger, no war, no killing.............Then we have obama. 6 years, $6 trillions........Millions loosing their health care all in the name of social justice. Ukraine just got invaded, and the Cold War is about to be revived all because of obama believe the world should be all equal and we have no right to be the dominant power in this world. All the red line drawn by obama and laughed it off pretty much invite Putin to invade Ukraine.....Because the little external effect of power grabbing.

Using IRS to intimidate opposition, NSA to email and phone tap of repoters..........All to hold on to the power and the end justify the means........

All for the ideal theory of Communism and Socialism .


Don't forget the latest news, the mayor of NY....Bill de Blasio, promised equality, socialism............he just closed the charter school for the poor in Harlem, the only hope for people in that part of the city to get out of the gato..........to please his supporters.

[Merlinb]

Now somehow, they got into the idea of socialism, what are they good for now? Other than Mercedes and BMW that are selling because of the name associate with prestige, what product they can claim to be on top? The whole Europe is in financial crisis and guess why? It's the ideal theory of Socialism!!!

Compared to the US, western Europe has the better healthcare*, education*, shorter working hours, longer life expectancy, lower rates of crime, more social equality... I'd say socialism is doing pretty well! (Not that we really are socialists in the ideal sense. But Americans do like to put politics into neat little boxes for some reason, which doesn't really work...)
Oh, and who do we have to thank for the financial crash? Was it the capitalists by any chance?

*'Better' in the sense of quality divided by cost.

Not quite sure what this has to do with parallel triodes...