# Amplifiers, transistors and opamps

1. Aug 22, 2015

### fog37

Hello Forum,

I have recently learned about transistors and their ability to control (amplify) other signals. BJT transistors amplify current. There are also FET transistors. Some devices are termed current-controlled and others are called voltage controlled.

Are there amplifiers that amplify voltage (voltage amplifiers) and current amplifiers? Is there such a distinction? If current is amplified, then also voltage seems to be automatically amplified (V=IR).

Is the generic simbol for amplifier a triangle? Isn't that the symbol for operational amplifier?

thanks,
fog37

2. Aug 22, 2015

### LvW

It is a common misconception that the BJT would be a current-controlled device. Surprisingly, this false statement even can be found in some textbooks.
Instead, you can trust some papers from leading US unversities explaining that the BJT is voltage controlled - and the base current is an unwanted but unavoidable byproduct.
Furthermore, it is very easy to proove on a circuit level (without going deep into the physics of the pn junction) that it is the voltage Vbe that controls the collector current.
I think, everybody who knows and understands the working principle of a pn junction (pn diode) cannot come to the conclusion that the pn junction within the transistor behaves completely different.

(PS: This question was intensively discussed already here in the forum (starting with reply#13) : https://www.physicsforums.com/threa...imagine-transistor-saturation-current.824127/

Last edited: Aug 22, 2015
3. Aug 22, 2015

### anorlunda

A transistor is nothing at all like a resistor, so V=IR does not apply. Also, a transistor conserves energy so that power = VI (over all three terminals) can not increase.

4. Aug 22, 2015

### fog37

Hello LvW.

I have read that a BJT is a voltage controlled device too but most sources call it a current controlled device.

Could you give me your definition of voltage-controlled and current-controlled? In the BJT (voltage controlled) a voltage between base and emitter causes a large collector current to flow. So we have a voltage controlling a current. As you mention, the base current is very small and unwanted. Ideally it would be zero.
So, in general, there is a controlling circuit component and a controlled circuit component. How do the concept of voltage controlled and current controlled apply?

Voltage controlled device: the magnitude of either the current or voltage or resistance of the device is controlled by the voltage of a different device (controller).
Current controlled device: the magnitude of either the current or voltage or resistance of the device is controlled by the current of another different device (controller).

Do you have some practical examples of both types of device?

thanks
fog37

5. Aug 23, 2015

### meBigGuy

Saying a BJT transistor is voltage controlled or current controlled makes no difference. The math for thinking of it as voltage controlled (Ie = Ies (e ^(Vbe/VT) -1) is pretty sweet, but Ic = beta*Ib isn't so bad either. You can increase the base voltage by forcing current or increase the base current by increasing voltage. Both will increase the collector current. Either view works. Most analog designers I know think of it as Voltage controlled. But, don't get hung up on one view over the other. They both have their uses.

Trying to separate voltage and current is impossible since they are very interrelated . Did the voltage cause the current, or did the current cause the voltage?
But generally, voltage amplifiers have high input impedance and low output impedance so the voltage becomes an easy way to express the performance.
Current amplifiers have low input impedance, and high output impedance, so vice versa.

If you look at amplifiers, there are voltage amplifiers, current amplifiers, current feedback amplifiers, and transimpedance amplifiers. They all have their applications and uses. Voltage amplifiers are probably the best known and generally understood. I suggest you search for current feedback amplifiers. And look at current-to-voltage and voltage-to-current converter circuits.
I find the most useful method is to enter something like "current to voltage amplifier" in google and then look at the images for one that seems useful.

AS you will quickl see, taking a standard op-amp and hooking it up differently can implement a lot of different "amplifications"

https://en.wikipedia.org/wiki/Current-feedback_operational_amplifier

6. Aug 23, 2015

### LvW

"Many errors, of a truth, consist merely in the application of the wrong names of things." - Spinoza

I think, this sentence applies in particular to the subject of our discussion.
I agree to meBigGuy that - for practical applications and designing circuits - it makes no difference if we believe in current-control or voltage-control (note the term "believe"!) as long as we are following the correct design steps (and corresponding formulas).

However, the situation changes if we
(a) ask for physical principles behind the formulas, or
(b) start thinking what we are really doing (and why !), or
(c) critically review the properties of the circuits we have designed.

Examples:
for (a): This involves carrier physics, which I dont like to discuss at the moment (to involved for such a forum). On the other hand, we all know the "mystic" value of dVBE/dT=-2mV/K which tells us that - for a temperature caused increase in the collector current Ic - we have to reduce the base-emitter VOLTAGE by 2mV in order to keep Ic constant.

for (b): For BJT amplifier stages we are biasing the base terminal with a voltage divider which should be as low-resistive as possible (as allowed).
Why this design goal?
Because we try to make the base DC potential as "stiff" as possible which means: This potential should depend as less as possible on the base current (which has a very large tolerance). This clearly indicates that we are trying to use a DC VOLTAGE for biasing. However, we must find a trade-off between this "stiffness" and the resulting input impedance which shouldnt be too small (loading of the signal source). Such a trade-off is easier to realize using, for example, the method of bootstrapping.

for (c): Because of several reasons it is always recommended to use an emitter resistance Re for negative feedback. This method changes/modifies many properties of the BJT stage - one of these changes is the signal input resistance that is increased (desired effect, of course).
From feedback theory we know that such an increase for the input resistance is to be observed only in case the feedback signal is a VOLTAGE. If the feedback signal is a current, the input resistance would be decreased.
From this it follows, that it is the base-emitter VOLTAGE that is influenced by feedback .
______________________
As another problem, the situation is even more complicated because of our technical language which sometimes is confusing (see Spinozas sentence).
As an example, we are using the term "current source" for a voltage source with an internal source resistance that is much larger than the connected load resistance. But in fact, it is NOT a current source. No current without voltage. Technically, there is nothing like a "current source". Each current needs a driving voltage (better: an E-field) - otherweise the charged carriers forming the current cannot move.

As to your question regarding examples for current-control:
In most cases, we are using this term if the controlling terminal has a very low input resistance (some ohms), for example current-feedback amplifiers which have a high-resistive (non-inv.) input node and a second low-resistive (inverting) input node.
As another example serves the classical current mirror where the current in the second (output) transistor is (nearly) identical to an externally injected current into the first (input) transistor. By the way: This equality of both currents is possible only because both base-emitter VOLTAGES are identical - another proove for voltage control!.

LvW

Last edited: Aug 23, 2015
7. Aug 26, 2015

### fog37

Thank you.

To keep things clear in my mind, how should I think of a current amplifier and a voltage amplifier? Is a current amplifier a device with low impedance that lets current pass through it and amplifies it?

Is a voltage amplifier always a large impedance device such that an applied voltage at the input is converted into a larger voltage at the output ports?

thanks,
fog37

8. Aug 26, 2015

### meBigGuy

9. Aug 26, 2015

### donpacino

what I am about to say assumes a continuous gain across different loads and ideal amplifiers (which takes into account what mebigguy said)

a voltage amplifier will always amplify voltage.
ex. input: 5 V
amplifer: gain of 5 V/V

output voltage with no load: 25V output current: 0A
output voltage with 10 ohm load: 25 V output current: 2.5A
output voltage with 500 Kohm load: 25V output current: 50uA

the same goes for current. Keep in mind REAL amplifiers wont be linear or perfect.

10. Aug 26, 2015

### sophiecentaur

Absolutely. If there is an equation that relates volts to volts, current to current or any other combination then, if you want to, you can use that to generate a name (Transistor = Transfer Resistor???) Just goes to prove my signature about categorising things.

11. Aug 26, 2015

### meBigGuy

Actually, Sophie, you tend to categorize things more than anyone else on this forum (along with being the most negative poster). I always considered your signature to be the ultimate irony and have resisted dozens of situations when I wanted to tell you to read it yourself. I finally could not resist commenting on it. If no one "likes" this post I will consider myself suitably chastised.

12. Aug 27, 2015

### sophiecentaur

I have to admit to being negative and even grumpy at times. I do try to keep it under control.
There is a difference between attempted 'precision' and categorizing for its own sake - and then the resulting confusion when the chosen category causes a logical conflict. I read so many worried posts that have arisen just because someone has been told a 'rule' and then has found a exception.
I will try to practice what I preach, but I still see the message as a valid one. Could you give some example on this thread where you feel I have contravened my rule?

13. Aug 27, 2015

### fog37

Hello,

I have done some reading and wikipedia has a good description on what an amplifier is. The goal of an amplifier is, given an input signal, to produce an exact but enlarged copy of it, the output signal. This is done by modulating a power source such that the modulation re-creates the same shape as the input signal.
There are the following types of amplifiers:

1. Voltage amplifier: An input voltage is amplified to a larger output voltage. The amplifier's input impedance is high and the output impedance is low
2. Current amplifier – input current produces a larger output current. The amplifier's input impedance is low and the output impedance is high
3. Transconductance amplifier – This amplifier responds to a changing input voltage by delivering a related changing output current
4. Transresistance amplifier – This amplifier responds to a changing input current by delivering a related changing output voltage
Some amplifiers can be realized with just a few resistors and a single transistors. In other cases operational amplifiers are used which have multiple transistors. At the end of the day transistors seem to be the main ingredient of an amplifier...

There is always a source connected to the amplifier (to its input) and a load connected to the amplifier output, correct?

In example above (reported below) for a voltage amplifier, the output impedance is
ex. input: 5 V
amplifer: gain of 5 V/V
output voltage with no load: 25V output current: 0A
output voltage with 10 ohm load: 25 V output current: 2.5A
output voltage with 500 Kohm load: 25V output current: 50uA

the impedance of the load affects the output current while the output voltage from the amplifier remains constant. But the load can change. The load is what we connect to the amplifier to receive the amplified voltage. The load (with its impedance) is not supposed to be part of the amplifier circuitry, correct? All the voltage amplifier does is to provide a potential difference. A transmission line has an impedance. I guess a device can have two impedances (an input and an output). I am familiar with the process of matching those impedances to the source or to the load to avoid energy reflections....

Thanks,
fog37

14. Aug 27, 2015

### donpacino

one thing to add, amplifiers can also have a gain with a magnitude of less than one, which will attenuate the signal (make it smaller)

15. Aug 27, 2015

### sophiecentaur

One important thing about the 'enlargement' process is that the output signal should have more Power - or at least, more available. That may not involve increasing Voltage or Current, which could often be achieved using a passive method (e.g. transformer). This is mentioned in the above post.
Having said that, amplifiers can also be used to 'buffer' between two stages, so that the input circuits are not affected by variations in the output circuits.

16. Aug 27, 2015

### meBigGuy

Sophie, I don't think your post in this thread was an issue, it just pushed my (admittedly oversensitive) button since you mentioned your sig. The router thread was a classic example of you classifying the OP and making assumptions about router complexity (but mostly not being supportive or educational). No one liked my post here, so maybe others don't agree. I can be over-reactive, I know. You quite consistently post useful and educational information. My primary goal when I post is to try to move my knowledge and the thread's content forward in an useful and educational way. If I think someone is not prepared, I *try* to point him to something that will demonstrate that in a useful way or ask questions that will make him realize what he doesn't know. (I'm sure there are plenty of contrary examples in 1600 posts)

Regarding this thread and amplifiers:
The load is technically not part of the amplifier, but can affect amplifier performance.
For example, real voltage amplifiers have an effective output impedance, so increasing the load (driving lower resistance) will cause more voltage drop within the amplifier (and more power dissipation). Similar effects exist with the source impedance. Amplifier performance usually is specified for a range of load and source impedances.
Regarding amplifier design: given the load range, feedback techniques may be used to reduce the effects of loading.

17. Aug 27, 2015

### sophiecentaur

Why are you using an unrelated post to continue with this personal stuff? Do you want me and others to take you seriously? If I were you, I would try to calm down about this. It cannot be good for your blood pressure.
Send me a PM if you want to continue this.

18. Aug 27, 2015

### jim hardy

hmmmm

indeed ,,

i = eqv/kt and q&k are constants.

I've come to really look forward to posts by you guys, both of you are knowledgeable astute and eloquent.

Lavoisier on classification:

Please don't stop with your quips , Sophie , i've never seen one that was mean spirited or ad-hominem.
If we can't laugh at our mistakes, what will we do for fun when we get old ?

old jim

19. Aug 28, 2015

### LvW

"Trying to separate voltage and current is impossible since they are very interrelated . Did the voltage cause the current, or did the current cause the voltage?"

Jim Hardys post has pointed my attention again to the above quote (from meBigGuy). I think, this question deserves a separate discussion. Of course, they are "interrelated" - but is this something like a "chicken-egg" case? I dont think so: Voltage can cause current but not vice versa!

20. Aug 28, 2015

### sophiecentaur

Nice words Jim - thanks. But I know I can be ` grumpy old sod at times.