How do I know if a transistor is voltage controlled or current controlled? It seems to me that changing voltage, ceteris paribus, changes the current so what gives?

http://en.wikipedia.org/wiki/Bipolar...rs_in_circuits

Talking about exponential current gain also seems to indicate a change in R or V. I'm just trying to mock up a basic, basic audio amplifier using NPN transistors but it feels confusing given the fact it's a 3 terminal IC.

Edit: This is the NPN I've got. http://www.makershed.com/v/vspfiles/...ges/2n3904.pdf
 PhysOrg.com engineering news on PhysOrg.com >> NASA: Austin, calling Austin. 3-D pizzas to go>> Research suggests modular design competence can benefit new product development>> Army ground combat systems adopts tool for choosing future warfighting vehicles
 Look on the datasheet you provided about halfway down on page 2 where it specifies HFE Current Gain. Can you find any place in the spec. sheet referring to voltage gain? No. This is because bipolar transistors are current gain devices. The collector current is controlled by the base current. The voltage gain on the other hand is controlled by the resistors used in the collector and emitter circuits, not by any parameter of the transistor.
 Does the current gain get controlled by an increase in voltage or a reduction of resistance or both?

Current gain is just that, current gain, for dc and low frequency hFE = IC/IB. For high frequency, hfe = ic/ib. As far as "an increase in voltage" goes, which voltage do you mean?

Some devices are fabricated to exhibit current gain that is constant over several decades of current. One device which comes to mind is the MPS06/56 (npn/pnp). The more common device, the 2N2222A/2907A (npn/pnp) exhibits a current gain that rises with increasing collector current, peaks at some value in tens of mA range, then drops. Also, hFE rises with increased temperature, and drops at low temp.

A good text, peer-reviewed, on semiconductor devices for applications will give you a good start. I recommend university text books, and app notes from bjt OEMs. These are the experts. There are capable experts on this forum who have great knowledge, but there are hackers with partial knowledge who engage them in endless debate, confusing any issue. These bjt questions seem to flush out the self proclaimed experts and end up having the thread locked, then nobody benefits.

Just trying to help by telling you how it is.

Claude

 Quote by cabraham ...but there are hackers with partial knowledge who engage them in endless debate, confusing any issue. These bjt questions seem to flush out the self proclaimed experts and end up having the thread locked, then nobody benefits.
That's one way to kill contributions to a thread on the third post.

 Quote by the_emi_guy That's one way to kill contributions to a thread on the third post.
Unfortunately what I said has been proven true for over a decade. I wasn't trying to discourage contributions to this thread, but electronics just seems to have that appeal to a lot of people who have little to no formal education in the subject. Having an interest in something w/o formal education is not a bad thing. As long as one knows that they have limitations in their knowledge base, all is well. I have been taking guitar lessons for 2.5 years. I am an ok rhythm guitarist, and a not good at all lead guitarist. I'm working on it. I have a full time job, and study for the Ph.D. in EE, so practice time is limited.

Having said that, I would never argue with a highly skilled guitarist about how to play the instrument. Anybody has the right to play an instrument for fun w/o developing their chops to a pro level. But we need to know our limitations. I am not qualified to tell a guitar pro anything at all. My guitar knowledge is at the point I call "enough to be dangerous".

Unless a person has studied EE for years at the uni level, and practiced a decade or so doing serious hardware development, their opinions are just their opinions based on limited skill. I encourage comments, questions, and robust discussion, but I do not want to engage in demolition derby which locks the thread. Nobody benefits from that. The contrarians who preach heresy do not accept peer-reviewed text books as authority and have an inflated view of themselves. It makes communication difficult.

I'm only saying up front that questions like the one the OP posted tend to start flame wars. I wish to avoid that, but by all means, please do not misconstrue what I've said. Discussion is a good thing, but we need to agree on authoritative references, and which institutions are the experts. I hope I don't come across as trying to discourage people from posting. BR.

Claude

Recognitions:
Homework Help
 Quote by mearvk How do I know if a transistor is voltage controlled or current controlled? It seems to me that changing voltage, ceteris paribus, changes the current so what gives?
The voltage changes in small-signal BJT operation can be too small to be accurately measured. Being too miniscule and so changeable with temperature and from device to device means it is not repeatable and the BJT does not lend itself to predictable stable voltage-based design here.

 Quote by NascentOxygen The voltage changes in small-signal BJT operation can be too small to be accurately measured. Being too miniscule and so changeable with temperature and from device to device means it is not repeatable and the BJT does not lend itself to predictable stable voltage-based design here.
Good point. Also FWIW, a voltage source across the b-e junction of a bjt is destructive. With voltage control or voltage drive, a current is present in the b-e junction. That current times the source voltage is the power dissipated. He temperature increases due to this power, and the current increases because silicon is more conductive at increased temp. Voltage stays fixed w/ increased current so that power increases.

Increases power increases the temp, which makes the silicon more conductive. Current increases, as does power, increasing temp, increasing current, etc. We all know what thermal runaway means. We cannot control a bjt by controlling voltage directly. We can instead control its current.

A current source across the b-e junction establishes a fixed current and a voltage determined by the silicon junction characteristics. The product of current & voltage is the power, which results in a temperature rise. Silicon is more conductive at increased temp. But with current drive this is a stable condition. An increase in conductance of silicon means that the resistance decreases. With current drive, the voltage drops as temp increases. Thus the power drops, and thermal runaway does not happen.

The bjt device obviously needs both current and voltage to operate, and one w/o the other cannot happen (except in photo-transistors where I or V can exist alone). But only 1 of them is the directly controlled quantity with the other indirect, determined by material & physics laws. If we control the current we get thermal stability. If we control the voltage we get a pile of ashes.

There is a good reason why things are defined the way they are. Again, the importance of voltage is universally acknowledged. Nobody in the know overlooks voltage when analyzing a bjt. If I wish to drive a b-e junction from a 3.3V source, I scale the resistor properly based on VBE and the 3.3V source. But if my source is 1.2 volts,and I wish to drive the 3 b-e junctions of a triple Darlington bjt device, well, I am going to use a different source, one w/ a higher voltage.

I never neglect VBE because I know it is important, as is IB, as is IE, as is VCE, as is thermal coefficient, temperature, etc. We don't control a bjt with these other quantities like VBE or VCE, but we include them in our computations. I will elaborate if needed. BR.

Claude
 Recognitions: Gold Member Science Advisor Saying a transistor is 'really' current or voltage controlled is to fall into the trap of living life by classification. That is a false god when followed to the extreme, surely. What counts is its transfer function, as normally given and what circuit it's included in. If the be junction voltage varies by next to nothing, the base current will vary a lot and relationship between Ic and Vbe looks horrible. So it's gain is given in terms of current. But why worry, when you're in the early stages of electronics? Modern transistors have a lot of 'gain', however stated, and it is normal to use feedback in most practical circuits where gain is defined by other, linear components. I remember being taught about the transistor at device level before its use in a circuit and I can't help feeling that was the wrong way round. However, to be fair, your average transistor had an hfe of about 20 at the time.
 As Cabraham said, transistors can be oversimplified to the point the model is basically unable to explain some interesting phenomenon, or overcomplexified - I had to go through a year of solid state physics during which I was able to calculate the average velocity of the minority charge carriers, but it wouldn't help me build a circuit. A good start would be Ed Lipiansky's book: Electrical, Electronics, and Digital Hardware Essentials for Scientists and Engineers. I don't know that specific book, but I know the author's style.