How exactly does this component that we call a transistor amplify the signal?

[tubegeek]

When I teach AMPLIFYING DEVICES, here's an outline of what I think is too important to skip over, even at the most rudimentary level.

• An amplifying device has the potential to create an output signal larger than the input signal: this is AMPLIFICATION.
• "Larger" means MORE POWER - in other words, MORE ENERGY
• Energy can neither be created nor destroyed, more energy must come from somewhere
• Therefore all amplifiers must have a POWER SUPPLY.

Welcome to the science of ELECTRONICS!- up till now, we have been studying ELECTRICITY. The dividing line is active devices, i.e., amplifiers.

• All amplifiers have these three features: an INPUT SIGNAL, a POWER SUPPLY, and an OUTPUT SIGNAL.
• The relationship between these three features: The input signal CONTROLS the power supply, which CREATES the output signal. Hence the output signal can have more power than the input signal.

• The three amplifying device categories encountered in audio are the VACUUM TUBE (VALVE), the TRANSISTOR, and the INTEGRATED CIRCUIT (in order of their invention)
• An integrated circuit is an extension of transistor technology by which many transistors (as well as other components like diodes and capacitors) can be combined very densely into a very small space. This difference is significant enough that it is reasonable to give integrated circuits a separate category.

As an aside, amplifying devices are not always configured to produce a larger output signal: there are unity-gain and logic applications, for example, which use amplifying devices but do not increase the power output from a circuit.

Also: a transformer does not amplify. If a transformer's output is a larger voltage than its input, its current output will be smaller (and vice versa), so the power output (V * I) will be approximately equal to the input. In real circuits output power will actually always be somewhat less than input power due to imperfections in real transformers. Transformers are very important, especially in audio, but they are NOT active devices.

[merlinb]

Also: a transformer does not amplify.

Not sure I agree with this...

[tubegeek]

It's a passive device. You can have voltage gain through it but you can't have amplification in the sense of greater power out.

I don't THINK these are my definitions but I'm not sure where I got them at this point.... I'd be happy to be corrected if I'm wrong.

I just looked on Rane's web site, their Rane Audio reference entry on GAIN offers this extra-confusing explanation:

"gain The amount of amplification (voltage, current or power) of an audio signal, usually express in units of dB (i.e., the ratio of the output level to the input level). For example, amplifying a voltage signal by a factor of two is stated as a voltage gain increase of 6 dB. [Historical Usage Note: originally the terms 'gain/loss' were restricted to power use only, and 'amplify/attenuate' were used for voltage and current -- although I can find no historical explanation for this arbitrary split, and no existing standards can be found that continue to make such a distinction. It is interesting to add that conformity with such a narrow definition of 'amplification' says that the original manufacturers misnamed their products: they should have been called a 'gainifier' -- not an 'amplifier.' According to the true believers a 'power amplifier' is a contradiction since you cannot 'amplify' power, only 'gainify' it. 'Power gainifier' is the correct term, according to them.]"

Anybody have any help here?

[R.G.]

Also: a transformer does not amplify.

Not sure I agree with this...

I'm pretty sure it's another of those one-man's-ceiling-is-another-man's-floor things.

Transformers can increase voltage, or current, but not both at the same time. They cannot provide power gain, except for the ever-more-abstract cases of magnetic amplifiers, which are not transformers in the sense of this forum in the first place.

In fact, transformers always produce a power loss, as some of the input power goes into waste heat in making the transformer operate.

Transformers are much like levers. A lever can "amplify" your force, or the distance of a motion, but not the power you put in to move it, and it always produces a net loss of power depending on the mechanical losses of the various pivots, sliding parts, and so on of the level mechanism.

In the immortal words of William Jefferson Clinton, "it all depends on what your definition of the word "is" is."   

For most of this forum's denizens, it's probably best to know it for what it really is - an electrical lever - and not have them get bound up in the definition of "amplify". It's hard enough to have to keep explaining that "gain" does not mean "distortion".

[tubegeek]

I'm just glad this controversy is making it possible for me to get away with the dreaded "MARQUEE"tag!

In all seriousness, the difference of opinion on this is why I stake out the definitions in the first few bullets, I do realize that without them, it becomes ambiguous.

On the other hand, if Merlin is disagreeing with me, I'm not so comfortable with my choices. I'm hoping that your "Not sure I agree with this..." isn't the same thing as "I'm sure I disagree with this..."

[R.G.]

I was trying to be polite.

I completely agree with you. Transformers do not amplify. They may step up voltage or current, but not both, and that's not what I call "amplifying", nor is it a common way to think of it in the EE biz.

[acehobojoe]

That makes sense to me, the OP. It's all about the relationship of voltage to current.

[merlinb]

I'm hoping that your "Not sure I agree with this..." isn't the same thing as "I'm sure I disagree with this..."

I have always thought of the word 'amplify' as being general; applying equally to voltage, current, power, or any individual quantitiy you care to choose, so I am happy to think of a transformer as being able to amplify voltage or current.
But maybe I just use the word wrongly. I know some people get really irritated when I talk about an attenuator as having 'gain', or when I say that an impedance can be 'bootstrapped' to a smaller value as well as a larger one. To my mind all those words just mean 'multiplication', so it's simply a matter of maths. A factor of 0.1 is just a number, and so is a factor of 10, so I don't see the need to attach different names to numbers greater or smaller than 1. Similarly, I don't see the need to attach the word 'amplify' only to watts. Watts so special about those units?  

that's not what I call "amplifying", nor is it a common way to think of it in the EE biz.

This isn't the first time I have noticed that the definitions adopted by the IEC and other organisations are sometimes illogical or needlessly exclusive. Sometimes they're simply ignored by working EEs. The pages of Wirelss World used to be filled with letters from people wondering on what planet the IEC was living!

[tubegeek]

Merlin, you are definitely not alone - the use of "amplify" in a more general sense is what I have found in a couple of places (Rane Audio Reference, Gibilisco's "Teach Yourself Electricity and Electronics," to name literally the first two I checked.)

I had thought it was a much more cut-and-dried distinction (based on the introduction of active devices into circuits) than it apparently is in common practice.

I hate it when I am both pedantic and inaccurate but it seems I am learning something anyway.

I suppose to be clearer I should say "A transformer cannot amplify POWER, and certainly is a passive device, not an active one." That would satisfy my own lust for power

I'm glad I took my class notes out for this test-drive, nobody I work with would have challenged it but sadly that's a low bar.

[bancika]

If we take it all the way we can conclude that there is no amplifier, passive or active. We are never amplifying something out of nothing, amplification would imply that it's perpetuum mobile 

One way (maybe correct, maybe not, I like it as a non-EE) to look at the amplifier stage, transistor or tube, is that amplifier just modulates the power supply using the input signal. What we get at the output is supply voltage that swings up and down as a response to input voltage magically interacting with the amplifier. So from that standpoint, it's really not amplifying anything.

[samhay]

bancika - I was just thinking the same thing.

Transistor - thought shalt break the 1st law of thermodynamics.

[R.G.]

I have always thought of the word 'amplify' as being general...
To my mind all those words just mean 'multiplication'
...
This isn't the first time I have noticed that the definitions adopted by the IEC and other organisations are sometimes illogical or needlessly exclusive. Sometimes they're simply ignored by working EEs.

In the immortal words of William Jefferson Clinton, "it all depends on what your definition of the word "is" is."