Hi,

I am very confused in the fact that determine transistor is a current controlled device or voltage controlled device.I read the reference document in this forum said that transistor is a current controlled device, and I always believe that but rencently I have just read "How Do Transistors Work? " at http://amasci.com/amateur/transis.html
In the site they said that Bipolar transistors, like FETs, are voltage-controlled devices. This make me feel very confused.Could anyone please help me clarify this?

Thank a lot.

 

Ignore some of the crap you find on the internet. Everyone that can type is not an expert in electronics, neither are they speaking to beginners when they try to go into the most controversial aspects.

 

A bjt is current controlled. Search my post history & I provide links to threads which go into exhaustive debate. The keeper of the site you refer to is Bill Beaty, whom I exchanged info with. He's a smart person but does not have enough education to go telling the whole EE community, semiconductor industry, & academia how a bjt "really works". Like many contrarians, he has truth in his writings, but then takes an assumption that is not valid & leads us to a false conclusion.

Ten truths plus one lie is not reliable. A conclusion must be supported by all known laws of physics. I demonstrate in my postings, with help from other qualified EE/scientists, that a bjt needs both current & voltage to operate, but current, specifically emitter current is the quantity that is best used to control device behavior.

Claude

 

In the site they said that Bipolar transistors, like FETs, are voltage-controlled devices.

Bipolar transistors are not like FETs. They are different technology. A bipolar transistor is current controlled. A FET is voltage controlled.

 

Current and voltage both go together. If you had a class of devices that were considered current controlled devices and another class that were voltage controlled where would you draw the line when a new class of devices falls somewhere in between?

The physics of bjt is described using electron-hole pair generation, carrier concentration and mobility. Thus the model used is based on charge flow. Hence a bjt is considered a current flow device.

 

From a strict physics point of view the BJT is a voltage controlled device and they have a small-signal transconductance gain value ( Δ collector current divided by Δ base-emitter voltage). But that characteristic is typically only used in the calculations for the design of small-signal AC amplifiers.

However, since the input looks like a low-impedance, forward biased diode there is also a current associated with this base voltage. The collector current is proportional to this base current given by the value of Beta or Hfe. And from a large-signal (switching) and biasing point of view its much more convenient to use Beta than the transconductance value for calculating circuit operation, so that is how BJTs are typically described.

The point is that, although BJTs can be viewed as voltage operated devices, they are more commonly thought of as current operated devices and that is how you see them analyzed in most circuit applications.

 

For the most part, it is a meaningless distinction. Is a resistor a voltage-controlled device or a current-controlled device? Does the voltage across it result from the current flowing through it, or does the current flowing through it result from the voltage appearing across it?

This is why you have two base small-signal models for a BJT, one of which has a current-controlled current source and the other a voltage-controlled current source. They are 100% equivalent and interchangeable, but one makes some circuit design/analysis tasks easier while the other makes different tasks easier.

In order for the distinction to have meaning, you have to be working at a level at which the model does not use a one-to-one relationship between base-emitter voltage and base current and then look at whether the collector current tracks the voltage or the current better.

My understanding is that, while the charge-flow model is the easiest one to get your mind around, that a field-based description is just as doable, valid, and fundamental.

 

This is the guy that posted (today) that he couldn't tell if .8 volts from base to emitter would allow any current to flow. Which model would be more appropriate for him?

 

For the question in the other thread, he needs to either use Ebers-Moll model to directly find the collector current from the base-emitter voltage or he to find the base current and then apply the forward beta.

Without an external mechanism for limiting/determining the base current, you pretty much have to treat it as a voltage-controlled device (which is probably why he posted both of these questions). Hopefully the OP (others thinking about posting questions) will learn from this and realize that if you post a question without establishing the context behind why you are asking, then you are very likely to get answers that simply don't have any bearing on what you are trying to understand, but that's because you have forced the responder to assume a context (and they might not even be aware that they have done so) which may be completely at odds to the one the OP has in mind.

 

I have that problem regularly!

 

[HIJACK]
It really underscores the importance of teaching verbal and written communication skills to students (particularly engineering students, but students at all levels and majors). I'm as guilty as the rest when it came to seeing the required English and technical writing courses as a waste of time and, in part because of the attitude I had, I didn't get nearly as much out of them as I could have. I don't have any magic bullets to make high school and college undergraduates really grasp the value and importance; lot's of methods have been tried to get the point across and I don't think any have really met with success. I think most people (myself included) just aren't in a mindset to understand the importance of speaking and writing clearly until we have to deal with being on the receiving end of tons of stuff where the person can't speak or write clearly. (NOTE to the OP -- this is NOT directed at you or your post, at least not to the degree implied; it's a much more general and generic statement than that). It's then that we start realizing how much of what we hate in what we are given by others is contained in what we produce ourselves and only then that we start trying to get better.

I will forever regret 7th and 8th grade English classes. Those are the only two classes I ever ditched and I couldn't get along with either teacher and learned practically nothing, but those two years were all about grammar and sentence structure and parts of speech. I can't wait till my daughter gets to those classes (if they still teach any of that stuff by then) so that I can 'help' her with her homework and finally force myself to learn it myself!
[/HIJACK]

 

]From a strict physics point of view the BJT is a voltage controlled device and they have a small-signal transconductance gain value ( Δ collector current divided by Δ base-emitter voltage).[/B] But that characteristic is typically only used in the calculations for the design of small-signal AC amplifiers.

However, since the input looks like a low-impedance, forward biased diode there is also a current associated with this base voltage. The collector current is proportional to this base current given by the value of Beta or Hfe. And from a large-signal (switching) and biasing point of view its much more convenient to use Beta than the transconductance value for calculating circuit operation, so that is how BJTs are typically described.

The point is that, although BJTs can be viewed as voltage operated devices, they are more commonly thought of as current operated devices and that is how you see them analyzed in most circuit applications.

"From a strict physics point of view" the bjt is a charge controlled device. Any semiconductor physics test will affirm the same. The CC model & equations (current control) work fine until the speed of operation approaches the physical limitations of the device regarding storage time, rise, fall time, etc. The small signal transconductance model is very limited. It cannot accurately model the device in saturated (switching) mode. Only charge control is useful.

The CC model works fine when the speed of the input signals is well below the limitations of the device. The basic transistor equation is

1) Ic = alpha*Ie.

Auxiliary equations include:

2) Ic = alpha*Ies*exp((Vbe/Vt) - 1).

3) Ic = beta*Ib.

None of the 3 equations above account for time/speed limitations due to internal charge storage. Only charge control can provide accurate modeling when the device is operated at high speeds, as well as employed as a saturated switch. BR.

Claude

 

The CC model & equations (current control) work fine until the speed of operation approaches the physical limitations of the device regarding storage time, rise, fall time, etc.

I hope this clears things up for anhnha (the original poster.)

 

"From a strict physics point of view" the bjt is a charge controlled device. Any semiconductor physics test will affirm the same.

Not according to some of the posts in this thread.
So who are you going to believe?

 

Not according to some of the posts in this thread.
So who are you going to believe?

I believe the bjt producers, who model the device as current controlled at low speeds. At high speeds where the internal delays cannot be ignored, the charge control model is used.

Ljcox & Ratchit have participated in many threads besides bjt behavior. Those who insist bjt is voltage controlled have a belief that voltage is "causal" & current is an effect. They view the whole universe as follows: "voltage controls current in accordance with impedance". They start with voltage as an independent variable, then when that voltage is connected across an impedance, current is determined as V/Z.

Ask any contrarian (insists that bjt is VC despite OEM bjt producers claiming otherwise) the following question:

In a simple resistor, which is part of a larger network, not directly connected across a voltage source or current source, does the current through the resistor determine the voltage, or vice-versa, or is it chickens & eggs? Ask them. Let me make a prediction. The contrarians will claim that the voltage determines the current always.

They believe that a bjt device cannot be current controlled because there has to be a cause for the current, & that cause is voltage. First of all, Ic is controlled by Ie, not Ib. They insist that Ib is just a side effect of Vbe. This is wrong. The b-e junction has diffusion capacitance, which results in Ib leading Vbe in phase/time.

A change in Ib preceded the corresponding change in Vbe. A cause can never lag its effect. Causality states that cause always precedes effect. Vbe LAGS Ib, so it cannot be the cause of Ib. The following threads discussed this in detail & Ratchit acknowledged my positioin as valid, for which I commend him.

http://www.physicsforums.com/showthread.php?p=3434785#post3434785

http://www.electro-tech-online.com/...-does-transistor-amplify-current-voltage.html

I will clarify if asked. No need to type out lengthy posts when every point & counterpoint has already been made. Again, it will take time to read through the 2nd link, but we cover the most often presented arguments in favor of VC & the VC position is built on half truths, false assumptions, & arbitray statements/dogma. Again, I will discuss with anybody interested.

So who should we believe? The OEM. Check out Natl Semi/Tex Instr, Fairchild, Intl Rectif, etc. They call the bjt a CC device, & for switching at high speed, it is called a charge controlled device. I concur.

Claude

 

Thanks for all help.
Sorry for my poor English and limited knowledge.
I didn't have a proper education and I have learn many wrong things.Now I am trying to learn everything again from basis like diode, transistor,...

 

I believe the bjt producers, who model the device as current controlled at low speeds.
...

I think "current controlled" is the best generalisation, especially for beginners.

The most important factor controlling where the transistor is on it's conduction curve is the base current. Unlike a FET where it is gate voltage.

Someone could have a long successful career in electronics and never need to know much more than "a bipolar transistor is current controlled" and "a FET is voltage controlled". Leave the physics for physicists.