Working of BJT as an amplifier: concept

[shaan_aragorn]

How does a small change in base current result in a large change in collector current in a BJT? Please explain in terms of movement of electrons (in NPN). Most textbooks I have referred say that the reason is low resistance of emitter-base and high resistance of collecter-base. However i cannot see why that is the reason as I cannot picture the movement of electrons in a BJT. Please help.

 

[ranger]

Well I will try my best to explain it (with emphasis on electrons). Here goes:

Think of the base-emitter junction as a regular diode junction. In order for charge to flow, the junction must be forward biased with a voltage drop of about 0.7V (for silicon) across it. Now what does this actually mean? Well for a start it means current can flow from emitter to base. Furthermore it means that electrons in the emitter will flow over to the base-emitter junction and holes from the base will flow to the base-emitter junction. Because of the nature of holes and electrons, they combine at the junction. But here is where the magic is, the base has been doped in such a way that only few holes are available (about 1 hole for every 100 or so electrons). The more electons you put into the base, the less holes will be available for the electrons from the emitter (becuase most of the holes-electrons combinations are taken up by electrons entering the base), hence more emitter electrons will enter the collector region and there will be a stronger flow of charge. So what happens to 99% of the electrons? Well as you probably already know, the base layer is very thin. There also exist a depletion region between the base and collector regions. It is this depletion region that the electrons must cross to enter the collector region. Electrons then come under the influence of a strong electric field (which is between the base and collector region) that "sucks" the electrons. The electrons then leave the transistor through the collector terminal as current. In other words it all started with a forward biased diode junction (base-emitter junction) which provided the necessary holes and electrons, the electrons then combined with holes. The remaining electrons were then collected by the collector and emitted as current. The collector collects the electrons emitted by the emitter.

 

[shaan_aragorn]

Thanks for your efforts, but it leaves my question un answered: Why does a small change in base current result in a large change in collecter current? Again please lay emphasis on the flow of electrons.

 

[ranger]

Its actually small changes in the voltage applied across the base–emitter terminals which causes the current that flows from emitter to collector to change significantly. Its this applied voltage that causes the depletion region between the base and emitter to get thinner and charges start to cross it. A change in voltage causes a change in current, this is roughly like a resistor. But the voltage that controls the current flow in on an entire different region (base-emitter), yet its effects are seen on the collector side of the transistor. So its effects are transferred. Hence the name transistor.

 

[cabraham]

voltage/current = chicken/egg

Its actually small changes in the voltage applied across the base–emitter terminals which causes the current that flows from emitter to collector to change significantly. Its this applied voltage that causes the depletion region between the base and emitter to get thinner and charges start to cross it. A change in voltage causes a change in current, this is roughly like a resistor. But the voltage that controls the current flow in on an entire different region (base-emitter), yet its effects are seen on the collector side of the transistor. So its effects are transferred. Hence the name transistor.

Voltage does NOT *cause* current. All one need do is refer back to the basic rules regarding causality.

If "A" causes "B", i.e. A is the cause & B is the effect, two conditions must be met.

1) A must exist independently of B, i.e. A exists with or without B.

2) B can never exist without A.

A corollary to these two conditions can be stated as follows. A will ALWAYS chronologically precede B, and never vice-versa. The cause must always precede the effect. The effect can NEVER take place before the cause. There are no exceptions.

An example of A causing B is a bowling ball rolling down the alley, striking the pin, and knocking it down. The ball's energy was transferred to the pin, *causing* it to fall. The energy of the ball was established *independently* of the pin. Also, the ball had kinetic energy before the pin fell. Hence the ball *caused* the pin to fall.

Now, what if the bowling alley personnel wanted to play a joke on you, and tied some fishing line to the pin. With one pin standing, you roll the ball towards it, but someone yanks on the fishing line, knocking down the pin before the ball arrives. The ball then rolls through the space formerly occupied by the pin.

Would anyone argue with me when I say that since the pin fell *before* the ball arrived, that the ball was NOT the *cause* of the pin falling?

Take as another example the two poles of a magnet. There are always two, a north and a south, regardless of how many times you break the magnet into pieces. Does the north pole "cause" the south pole, or is it the other way around? They can only exist simultaneously and neither one can exist without the other. This relationship does not meet even one of the causality requirements, let alone both. Hence the relationship between the N and S poles on a magnet is NOT cause/effect. It is basically a chicken/egg thing.

In order to forward bias the base-emitter (b-e) junction on a bjt and set up an electric field in the b-e region, current must first exist before the voltage can change from zero to approx. 0.65 volts. The b-e junction capacitance requires a displacement current so that the voltage Vbe can change. In any capacitor the change in current will always "lead" or *precede* the accompanying change in voltage. Our friend "Eli the ice man" points this out for us since the 19th century. As the voltage Vbe increases from 0 to 0.65V, the base and emitter conduction currents increase simultaneously, or "in unison" with Vbe. Neither can change without the other. Hence the displacement component of base current takes place BEFORE a change in Vbe, whereas the conduction component is in UNISON with Vbe. Since the displacement current chronologically precedes the b-e voltage Vbe, it CANNOT be CAUSED by Vbe. An effect can never precede its cause. Since the conduction component of base current can never exist without Vbe, and Vbe can never exist without the former, neither one "causes" the other.

The voltage at the base-emitter DOES NOT CAUSE base current or any current. The base current DOES NOT CAUSE collector current. In order to produce collector current, Ic, all of the following quantities must be simulataneously present:

displacement current in base terminal, displacement current in emitter terminal, voltage across base-emitter terminals, displacement current in collector terminal needed to energize base-collector capacitance and provide base-collector voltage and associated electric field, base-collector voltage Vbc.

It takes EACH and EVERY of the above quantities to create or "cause" if you will, collector current Ic. The 3 quantities Ib (base current), Vbe (base-emitter voltage), and Ie (emitter current) can never exist independent of one another. All three of these are responsible for creating the electric field in the base emitter region.

The displacement components of the collector and base currents are both needed to create the base-collector voltage and the electric field in the reverse-biased base-collector junction.

These quantities cannot be separated. One doesn't "cause" the other. The displacement currents in all 3 terminals chronologically PRECEDE their respective voltages. This alone completely invalidates any possibility of a "cause-effect" relation. Yet another property that refutes the cause-effect claim is the mutual nature of the current and voltage in all terminals. Neither can exist without the other, just as in the magnetic poles case.

Arguing cause and effect is really nothing more than chickens and eggs. Best regards.

Claude

 

[ranger]

Voltage does NOT *cause* current. All one need do is refer back to the basic rules regarding causality.

I did not say that. What I was saying is that the voltage affects the depletion region, which will affect the amount of charge that can cross over.

In order to forward bias the base-emitter (b-e) junction on a bjt and set up an electric field in the b-e region, current must first exist before the voltage can change from zero to approx. 0.65 volts.

How would this current exist if there isn't a potential difference?

 

[waht]

Just going to say that, a slight increase in voltage in the base will try to maintain the same voltage drop of Base-Emitter junction. Therefore, a change in voltage at the base will result the same change in potential at the emitter if there is a resistor from emmiter to ground or its internal BE resistance.

Therefore an increase in voltage at the emitter will increase the current flow from collector to emmiter.

 

[cabraham]

I did not say that. What I was saying is that the voltage affects the depletion region, which will affect the amount of charge that can cross over.

How would this current exist if there isn't a potential difference?

It does indeed, but according to semiconductor producers, the b-e capacitance associated with a forward biased p-n junction depletion region is a strong function of current, and a weak function of voltage. For a reverse biased p-n junction, as in the b-c region, the capacitance of the depletion region is a strong function of voltage and weak function of current. Both quantities exert influence, but to a differing degree.

Please be more specific when you say "potential difference". I can only assume that you're referring to base-emitter potential difference Vbe. I was referring to *displacement current". This current Ib,disp, exists even when Vbe is at zero. In any capacitor, current exists even at the moment when voltage across the plates is still zero. The voltage exponentially rises, as the current diminishes. Of course, the external source providing this current must output a voltage, but that voltage is NOT Vbe. Now, regarding the conduction component of Ib, Ib,cond, of course it can't exist without Vbe. I've already stated that. But Vbe cannot exist without Ib,cond, nor without Ib,disp for that matter. In your original post, please reread it carefully, you strongly imply a pecking order, i.e. that voltage is more prominent and current is "caused by" voltage, which is not the case.

In order to control the depletion region and charge motion, you must change the value of the electric field. Electric fields store energy, W = 0.5*D*E, where D is the electric displacement, and E is the electric field intensity. Any change in this electric field value changes W, the energy. A change in energy requires some amount of time. The change in energy divided by the time is the average *power*. Hence a non-zero power is required to change the field. Power is the product of voltage and current, namely Ib,disp * Vbe, which must be non-zero. This can happen only if Ib,disp and Vbe are BOTH NON-ZERO. Thus in order to control the rate of charge flow through the depletion region, you must vary the electric field with time. This cannot be achieved with current alone or voltage alone. Both Ib and Vbe in unison are needed. Ditto for a FET. BR.

Claude

 

[rbj]

Thanks for your efforts, but it leaves my question un answered: Why does a small change in base current result in a large change in collecter current? Again please lay emphasis on the flow of electrons.

a while ago, i tried to hand-wave an explanation:

https://www.physicsforums.com/showthread.php?t=130412

(it's been 30 years since I've taken semiconductor device physics).

 

[berkeman]

I was referring to *displacement current". This current Ib,disp, exists even when Vbe is at zero. In any capacitor, current exists even at the moment when voltage across the plates is still zero.

In your original post, please reread it carefully, you strongly imply a pecking order, i.e. that voltage is more prominent and current is "caused by" voltage, which is not the case.

Power is the product of voltage and current, namely Ib,disp * Vbe, which must be non-zero. This can happen only if Ib,disp and Vbe are BOTH NON-ZERO. Thus in order to control the rate of charge flow through the depletion region, you must vary the electric field with time. This cannot be achieved with current alone or voltage alone. Both Ib and Vbe in unison are needed. Ditto for a FET. BR.

Claude

Claude, current flows because of a potential difference (voltage difference). Ranger is correct in his cause-effect description. F=ma, the force causes the acceleration. The acceleration does not cause the force.

 

[cabraham]

Claude, current flows because of a potential difference (voltage difference). Ranger is correct in his cause-effect description. F=ma, the force causes the acceleration. The acceleration does not cause the force.

Once again, examine the causality rules. Your statement is pure dogma. Ranger cannot be even remotely correct. If V "causes" I, why does I precede V in a capacitor? If I causes V, why does V precede I in an inductor?

As far as current flowing because of a potential difference, may I remind you that a superconducting loop can sustain a current indefinitely without a potential difference. I is non-zero while V is zero. A perfect insulator can sustain a non-zero voltage (ideal charged capacitor) with zero current. In one case I exists without V, and vice-versa for the other case. This absolutely precludes any possibility of a cause-effect relation. Neither I nor V is responsible for causing the other.

If a force is known to cause an acceleration, why does it follow that V must be the cause of I? Why doesn't I cause V? Once again, the rules of causality must be examined.

If I hold a ball, 3 feet above the ground and release it, it will drop. The two causality requirements, independence and sequence, must be met to concur that gravitational force caused the ball to accelerate. Rule 1 states that the g force must exist independent of the ball's acceleration, which it does. While I was grasping the ball, gravity was already pulling down on it. It pulls down always, independent of the ball. Rule 2 - the ball, however does not fall in a zero-gravity field. The g force chronologically PRECEDES the ball falling. All of the requirements have been met IN THIS CASE.

In an electric component, resistor, capacitor, diode, bjt, light bulb, etc., V does not exist without I, and I does not exist without V. Just like the poles of a magnet, one cannot exist independently. Which pole, north or south, is the cause of the other? Can anyone answer that one? The notion that V "causes" I is nothing more than a prejudice, acquired by many years of using power supplies that are "constant voltage" sources. Rules 1 and 2, which requires independence for the cause, and dependence for the effect, cannot be met by I nor V. To change the collector current Ic in a bjt requires a change in the base-emitter electric field. This requires a change in ENERGY. Energy change takes place over a finite time. The ratio of energy to time is power. This power can exist, i.e. have non-zero value, only if Ib and Vbe are both non-zero. Thus a change in Ic requires a change in both Ib and Vbe as well some finite amount of TIME. The change in Ib takes place AHEAD of the change in Vbe. Therefore Vbe cannot be the cause. The sequence requirement can't be met.

F = ma, and V = IR, are two distinct cases. Best regards.

Claude

 

[waht]

If a force is known to cause an acceleration, why does it follow that V must be the cause of I? Why doesn't I cause V? Once again, the rules of causality must be examined.

I think you have to examine the basics again.

Electrons only flow because they are repelled by a negative charge and pulled towards the positive one.

If there is no outside charge, or potential difference, the electrons don't move period.

Same with gravity, if there is no graviational potential, then a ball will float midair restless.

Therefore you get:

1. Potential difference V
2. Current flow I


As for the superconductor loops, you can have AC current loops circulating forever, because in AC you have an alternating voltage complementing the current 90 out of phase.

 

[berkeman]

Thanks waht.

 

[Manchot]

You can think of it using a water analogy. Imagine that you had a container of water with two holes in the bottom: one really big one (the collector), and one really small one (the base). There is also a faucet above the container which you can control. (How much water flows out of it is the emitter current, and how much the faucet is turned is the base-emitter voltage.) To get a certain flow out of the base hole, you keep turning the faucet until you get the desired flow. However, in doing so, you have also ensured that a large amount of the water is flowing out of the collector.

 

[ranger]

cabraham, why do you keep on insisting on causality rules? Also what's wrong with current and voltage begin out of phase? The poles of a magnet has nothing to so with this discussion.

 

[cabraham]

I think you have to examine the basics again.

Electrons only flow because they are repelled by a negative charge and pulled towards the positive one.

If there is no outside charge, or potential difference, the electrons don't move period.

Same with gravity, if there is no graviational potential, then a ball will float midair restless.

Therefore you get:

1. Potential difference V
2. Current flow I


As for the superconductor loops, you can have AC current loops circulating forever, because in AC you have an alternating voltage complementing the current 90 out of phase.

I have examined the basics quite thoroughly. Let's not forget that every quantity cannot be defined, or you'll get circular definitions. The one quantity that is postulated, or unproven is *charge*. The intrinsic force between two charges is called Coulomb force and is postulated based on empirical observation. Since this Coulomb force integrated over distance is equivalent to work or energy, we define a quantity, V, potential or voltage if you will in terms of charge and Coulomb force. The force is an inherent property of charge interaction, and voltage is defined as the ratio of work to unit charge, hence 1 volt = 1 joule/coulomb.

Now, to say that the force which moves the charge is due to the potential V, is a classic example of circular arguments, or circular definitions. Remember that charge, in coulombs, is not defined in any other unit. No electric quantity is more basic than charge. Charges don't move because there is a potential difference. Potential difference, or voltage, is DEFINED as the work per unit charge associated with the inherent property that charges exert forces upon one another. Charges move only because of proximity to other charges. This Coulomb force is inherent to charge and voltage is derived from that force integrated over the distance. Current I is charge per unit time, 1 amp = 1 coulomb / sec. Both I and V are defined in terms of charge, energy and time. Q, W, and T are the fundamental quantities, and I and V are both derivative.

I and V are both a result of Coulomb force, the associated energy over distance, and time.

Regarding the superconducting loop, a non-zero *dc* current will sustain indefinitely with zero volts. Of course, an ac superconducting loop with ac current must have a non-zero voltage, since the loop possesses non-zero inductance. A time-changing current can only exist if the voltage is non-zero.

To summarize, charges move due to proximity and interaction with other charges. The Coulomb force is intrinsic, fundamental, and not defined in anything more basic. Voltage and current are NOT FUNDAMENTAL electric quantities, but are defined as ratios of charge, energy, and/or time. Of course, if a charge moves due to the presence of other charges, a potential difference must exist. But in order to bring charges into proximity with other charges, work had to be done in transporting them. Moving charges is current, and the associated work per charge is voltage. Whenever you move charges, both I and V are produced concurrently. I and V are produced together. V does not produce I. I'm at a loss to make it any clearer. Best regards.

Claude

 

[berkeman]

And we are at a loss in trying to convince you. I'd recommend that we just let the issue lie for now, especially since it's a bit of a hijack for the original post (OP).

Welcome to the PF, BTW Claude.

 

[waht]

Good read Claude, but you said I precedes V, which caused a bit confusion.

I'm still baffled at this statement "V does not produce I "

But somehow I don't buy it.

Anyway, Berkman is right, we got carried away with this too much.