How does a transistor amplify current or voltage?

[Ratchit]

Brownout,

And the ones before.

Not the Fairchild data sheet.

I never said I proved it. You should read more carefully.

In post #395 you wrote "We presented simulaitons and proof." sic

Facts and truths, proof positive. Correct charts, correct math, correct simulations. All in excellent agreement, all consistent, all bulletproof. You've done nothing to show otherwise.

But not proven until Jony130 did so.

Glad you see the light.

With the Jony130 proof.

Ratch

 

[BrownOut]

Not the Fairchild data sheet.

All the other charts, equations, sumlations, etc.

I never said I proved it. You should read more carefully.
In post #395 you wrote "We presented simulaitons and proof." sic

Thanks for making my point

But not proven until Jony130 did so.

Proven many times and in many ways. Correct charts, correct math, correct simulations. All in excellent agreement, all consistent, all bulletproof. You've done nothing to show otherwise. And you never will.

With the Jony130 proof.

I prefer Jack Daniels, 80 proof.

 

[Ratchit]

MrAl,

My friend, if i may call you that, as i can put this as gently as possible...

Yes, as an address of courtesy. Go ahead and lay it on me. I can take it.

you dont seem to understand transistor action in its entirety, only partly as you have demonstrated repeatedly.
I had hoped to help you to understand a bit more, but you insist on ignoring the facts and presentations by myself and others

Well, I think I do. I have not ignored the "facts", but I am not going to accept them without proof unless they are obvious.

You can not seem to get the idea out of your head that the transistor only works the way you see it to work in spite of three knowledgeable people providing you with an abundance of information to the contrary.

Unproven information is suspect unless obvious. Can you list a specific item?

At the very least, you should acknowledge that ANY three terminal element has N modes of operation, where you can control the element in a number of ways (i think it is 6 in total, lets see, be, bc, ce, and their reverses and that's only for voltage).

Yes, and you can move the steering wheel by twisting the tires when the car is on a hoist. But who does that except a mechanic?

In other words, treat the transistor as a black box with three terminals and FORGET anything else you were told in the past.

There are many ways to model a transistor, but forgetting something is not my style.

You have to realize also that *we* understand every point *you* have presented so far,

As do I when you present something, even if I don't always agree with it.

but you have yet to understand one simple and straightforward point: that the transistor can be controlled, at least to some degree, by varying it's collector current and NOTHING else. It doesnt matter if its in saturation for not, for the third time i say this now! Forget about it. Assume sat if you wish, assume active if that suites you better, take your pic, the collector current still controls!

It is a matter of perspective and the region of operation. Sure, there is a one to one correspondence between, say Ib and Ic in the active region. You cannot change one without the other. But it is usually assumed that in the active region, Ib is the functional control.

That's it. Either you understand somehow, someday, that the transistor can be controlled from its collector current or you never do. Do us a favor and take a pick or flip a coin.
I've seen this happen many times before with other electrical issues, and it takes a bit of time to sit back and think about it and that's when it becomes more clear.

I explained my outlook of what controls what in the above paragraphs. No thinking time needed.

So are you willing to try to understand how the transistor can be controlled from its collector current, or are you going to argue against this indefinitely, despite all facts to the contrary?

I already said that the collector and Ib have a one to one relationship to each other, but usually Ib is thought to control the Ic in the active region. Nothing more needs to be said about this.

Ratch

 

[Ratchit]

Brownout,

All the other charts, equations, sumlations, etc.

They were discredited by the Fairchild data sheet, until Jony130 proved it wrong.

We presented simulaitons and proof." sic Thanks for making my poin

So the "we" included you. You did try to prove Vce increased enough with Ic's increase while in saturation. The Fairchild data contradicted you, so you did not succeed. Jony130 did.

Proven many times and in many ways. Correct charts, correct math, correct simulations. All in excellent agreement, all consistent, all bulletproof. You've done nothing to show otherwise. And you never will.

Proven once by a good data chart provided by Jony130. You are right, I cannot disprove the data given by that good ON data sheet, and I never will.

I prefer Jack Daniels, 80 proof.

Only in moderation, I hope.

Ratch

 

[BrownOut]

They were discredited by the Fairchild data sheet, until Jony130 proved it wrong.

That's the dumbest, illogical thing I've ever heard in my life. It was wrong and irrelevant: it cannot discredit anything

So the "we" included you. You did try to prove Vce increased enough with Ic's increase while in saturation. The Fairchild data contradicted you, so you did not succeed. Jony130 did.

I never said I wasn't included. You're taking more nonsense. It was wrong and irrelevant: it cannot discredit anything. So you make the same dumb comment twice, makes it twice as dumb.

Proven once by a good data chart provided by Jony130. You are right, I cannot disprove the data given by that good ON data sheet, and I never will.

Proven many times and in many ways. Correct charts, correct math, correct simulations. All in excellent agreement, all consistent, all bulletproof. You've done nothing to show otherwise. And you never will.

 

[Ratchit]

Brownout,

That's the dumbest, illogical thing I've ever heard in my life. It was wrong and irrelevant: it cannot discredit anything

But until Jony130 showed it wrong, we did not know it was wrong. So it was logical to assume it was right. There is a time sequence of events involved.

I never said I wasn't included. You're taking more nonsense. It was wrong and irrelevant: it cannot discredit anything. So you make the same dumb comment twice, makes it twice as dumb.

On post #399 you said "I never said I proved it. You should read more carefully." So did you try to prove it?

Proven many times and in many ways. Correct charts, correct math, correct simulations. All in excellent agreement, all consistent, all bulletproof. You've done nothing to show otherwise. And you never will.

Proven once by a good data chart provided by Jony130. You are right, I cannot disprove the data given by that good ON data sheet, and I never will.

Ratch

 

[BrownOut]

But until Jony130 showed it wrong, we did not know it was wrong. So it was logical to assume it was right. There is a time sequence of events involved.

The facts that proved correct operation also proved that chat was wrong and/or irrelevant.

On post #399 you said "I never said I proved it. You should read more carefully." So did you try to prove it?

That is correct. I never said I proved it.

Proven once by a good data chart provided by Jony130. You are right, I cannot disprove the data given by that good ON data sheet, and I never will.

The correct charts, correct math, correct simulations were shown. All in excellent agreement, all consistent, all bulletproof. You've done nothing to show otherwise.

 

[Jony130]

Ratchit

Can you construct the curves shown in fig. 16 of the ON data sheet from the Ic vs Vce? Show us how if you can.

Yes "we" can. And to be honest it is not so hart do to.
And I can not understand why Ic vs Vce curve was no enough for you to admit that Ic current may cause BJT to come out of saturation.

 

[MrAl]

MrAl,


Yes, and you can move the steering wheel by twisting the tires when the car is on a hoist. But who does that except a mechanic?

The transistor is not a car. The transistor is an electronic component. Strange though, you seem to say that no one would want to move the steering wheel that way yet you mention at least one person that would anyway. Which is it?

There are many ways to model a transistor, but forgetting something is not my style.

You dont really have to forget, you simply have to take a fresh look at it, as you do with any black box.

It is a matter of perspective and the region of operation. Sure, there is a one to one correspondence between, say Ib and Ic in the active region. You cannot change one without the other. But it is usually assumed that in the active region, Ib is the functional control.

That's just the opposite. You can change one without the other. That's why i mentioned the black box and taking a fresh look at this. The key word there too is "usually". That's right, usually you do assume Ib has the control, but in this case it doesnt. Ic has control, and if you look at it as a black box you can possibly get control in other ways too.

Look at it like this:
I present you with a *new* device, *not* a transistor, but it too has three leads. That means we can measure three voltages 'across' and three currents 'through' the device. If we call the voltages v1, v2, and v3, and the currents i1, i2, and i3, we can relate one parameter to the other two algebraically as either of:
v1=v2+v3
v2=v1+v3
v3=v1+v2
and
i1=i2+i3
i2=i1+i3
i3=i1+i2

which means we can control at least one parameter with the other two, and and there are other relationships too if we can establish some inter-terminal resistances, which means some of the currents might depend on some of the voltages, or vice versa.
There could also be internal amplification, which would mean that perhaps v3 for example depended on v2 times a constant like A. Thus, v3=v2*A, and that means other things are possible too.
What this all means is that a three terminal device is a much more complex device that needs to be looked at very carefully, and what is more is that we can not assume that one terminal is the only terminal that can control something else about the device unless the device, under test, proves this to be true.

Let me describe another type of device, which i wont say what it is yet, but we will mark the terminals the same as a transistor, b, c, and e, but those are only letters this time. Thus lead 1 is marked 'b', lead 2 marked 'c', and lead 3 marked 'e'. Now i wont specify what kind of device this is yet, just a few measurements.
Ok, so now we have our new device which we'll call the bce device, and to make it a little simpler we'll connect the 'e' terminal to ground, 0v. Next we measure 1ma flowing into the 'b' terminal from a current generator we have set up to pump 1ma into the 'b' terminal. Next we measure 10ma flowing into the 'c' terminal from another different current generator. Next we measure across the 'c' and 'e' terminals and we measure 0.15 volts. Next we increase the current generator connected to the 'c' terminal to 200ma, and we note that the only thing that changed was the voltage between 'c' and 'e', as it went up to 9.6 volts. Thus, when we increased the current into the 'c' terminal we saw the 'c' to 'e' voltage rise by 9.45 volts, from 0.15 to 9.6 volts.
Any problem with this so far?

Here's yet another 3 terminal device:
Ground the 'e' terminal, pump 1ma into the 'b' terminal, pump 10ma into the 'c' terminal,
measure 150mv between the 'c' terminal and the 'e' terminal. Increase the current into the
'c' terminal to 200ma, measure 2.86 volts from 'c' to 'e' terminals.
Any problem with that?

Any problem with either of these two devices?

 

[Ratchit]

Jony130,

Yes "we" can. And to be honest it is not so hart do to.

Not hard maybe, but tedious and prone to error. Some of the points between the two curves do not synchonize, and the final Ic-Vce curve does not match what the manufacturer shows.

And I can not understand why Ic vs Vce curve was no enough for you to admit that Ic current may cause BJT to come out of saturation.

OK, lets talk about that. Back to post #363 where you do your calculations. First a straight forward calculation to get the base current of Q1 = 600 µamps. Now the idea is to follow the 600 µamp Ib curve up to where it flattens out, and you aver it comes out of saturation at about Vce=1 volt. OK so far? When the transistor is on the 600 µamp line, it is just as much in the active region as it is in the saturation region. I always thought that anything left of the line where all the base currents converge was saturation, and to the right of that line was active. If it is on the line, then it is in a metastate condition. Any thoughts on that? Next you somehow show that the transistor comes out of saturation at 120 ma, but the Ib curve flattens at about 85 ma. Furthermore, in fig 16 of the ON date sheet, it shows that at 600 µamp and Vbe 0.69 volts, the Ic is about 85 ma.

Next, you give a equation of I =3V*t/100uH = 30mA/µs , which can be interpreted to be I = 30ma/µs, which is a units mistake. So can you show me how you get 120 ma? Another question, did you calculate the 30ma/µs, or did you get it from the simulation? That Ic vs Vce is a beautiful graph. Where did you find it? Most of the data sheets don't even have a Ic vs Vbe graph. So if you can clear up those points I mentioned, then you can probably convince me that saturation cutout can be determined from the Ic vs Vbe graph.

Ratch

 

[Ratchit]

MrAl,

The transistor is not a car. The transistor is an electronic component. Strange though, you seem to say that no one would want to move the steering wheel that way yet you mention at least one person that would anyway. Which is it?

I was using that analogy to illustrate a point, not to prove anything. There are exceptions, but most people would like to control the tire direction, not the steering wheel.

You dont really have to forget, you simply have to take a fresh look at it, as you do with any black box.

Every new problem needs a fresh look.

That's just the opposite. You can change one without the other. That's why i mentioned the black box and taking a fresh look at this. The key word there too is "usually". That's right, usually you do assume Ib has the control, but in this case it doesnt. Ic has control, and if you look at it as a black box you can possibly get control in other ways too.

Yes, in saturation, Ib does not have much control.

That's just the opposite. You can change one without the other. That's why i mentioned the black box and taking a fresh look at this. The key word there too is "usually". That's right, usually you do assume Ib has the control, but in this case it doesnt. Ic has control, and if you look at it as a black box you can possibly get control in other ways too.

Look at it like this:
I present you with a *new* device, *not* a transistor, but it too has three leads. That means we can measure three voltages 'across' and three currents 'through' the device. If we call the voltages v1, v2, and v3, and the currents i1, i2, and i3, we can relate one parameter to the other two algebraically as either of:
v1=v2+v3
v2=v1+v3
v3=v1+v2
and
i1=i2+i3
i2=i1+i3
i3=i1+i2

which means we can control at least one parameter with the other two, and and there are other relationships too if we can establish some inter-terminal resistances, which means some of the currents might depend on some of the voltages, or vice versa.
There could also be internal amplification, which would mean that perhaps v3 for example depended on v2 times a constant like A. Thus, v3=v2*A, and that means other things are possible too.
What this all means is that a three terminal device is a much more complex device that needs to be looked at very carefully, and what is more is that we can not assume that one terminal is the only terminal that can control something else about the device unless the device, under test, proves this to be true.

Let me describe another type of device, which i wont say what it is yet, but we will mark the terminals the same as a transistor, b, c, and e, but those are only letters this time. Thus lead 1 is marked 'b', lead 2 marked 'c', and lead 3 marked 'e'. Now i wont specify what kind of device this is yet, just a few measurements.
Ok, so now we have our new device which we'll call the bce device, and to make it a little simpler we'll connect the 'e' terminal to ground, 0v. Next we measure 1ma flowing into the 'b' terminal from a current generator we have set up to pump 1ma into the 'b' terminal. Next we measure 10ma flowing into the 'c' terminal from another different current generator. Next we measure across the 'c' and 'e' terminals and we measure 0.15 volts. Next we increase the current generator connected to the 'c' terminal to 200ma, and we note that the only thing that changed was the voltage between 'c' and 'e', as it went up to 9.6 volts. Thus, when we increased the current into the 'c' terminal we saw the 'c' to 'e' voltage rise by 9.45 volts, from 0.15 to 9.6 volts.
Any problem with this so far?

Here's yet another 3 terminal device:
Ground the 'e' terminal, pump 1ma into the 'b' terminal, pump 10ma into the 'c' terminal,
measure 150mv between the 'c' terminal and the 'e' terminal. Increase the current into the
'c' terminal to 200ma, measure 2.86 volts from 'c' to 'e' terminals.
Any problem with that?

Any problem with either of these two devices?

My head hurts just thinking about all of the three devices. I hope this is over soon. But yes, I am familiar with the black box concept. Every engineering student is.

Ratch

 

[Jony130]

Ratchit,

Some of the points between the two curves do not synchonize, and the final Ic-Vce curve does not match what the manufacturer shows

Thats becaues Ic-Vce curve don't come from the same manufacture.
The collector saturation region come from ON semi and Ic-Vce form Rohm.

So can you show me how you get 120 ma?

Because we use the simulation model for 2N4401 I plot Ic = f (Vce) for Ib=const in LTspice (see attached curve)
And from Ic-Vce curve plot by LTspice it's clearly to see why in LTspice current stop increase when he reache 120mA.
And of course if we change IB then Ic current will stop increases by a different value of a Ic current (see attached file).

Another question, did you calculate the 30ma/µs, or did you get it from the simulation?

Why to use simulation for such a simple calculation?
I think that everyone knows this equation:
V = L * dI/dt so for constant voltage in the inductor equation look like this
V = L *ΔI/Δt
So if we want to know the rate of current change (dI/dt) in amps per second we simple calculate:
ΔI/Δt = V/L = 3V/100µH = 30KA/s = 30mA/µs

That Ic vs Vce is a beautiful graph. Where did you find it? Most of the data sheets don't even have a Ic vs Vbe graph

**broken link removed**

 

[MrAl]

Hello again Ratch,


Well, you didnt answer my question again. I simply asked if you had a problem with either of those two devices.

START QUOTE
Let me describe another type of device, which i wont say what it is yet, but we will mark the terminals the same as a transistor, b, c, and e, but those are only letters this time. Thus lead 1 is marked 'b', lead 2 marked 'c', and lead 3 marked 'e'. Now i wont specify what kind of device this is yet, just a few measurements.
Ok, so now we have our new device which we'll call the bce device, and to make it a little simpler we'll connect the 'e' terminal to ground, 0v. Next we measure 1ma flowing into the 'b' terminal from a current generator we have set up to pump 1ma into the 'b' terminal. Next we measure 10ma flowing into the 'c' terminal from another different current generator. Next we measure across the 'c' and 'e' terminals and we measure 0.15 volts. Next we increase the current generator connected to the 'c' terminal to 200ma, and we note that the only thing that changed was the voltage between 'c' and 'e', as it went up to 9.6 volts. Thus, when we increased the current into the 'c' terminal we saw the 'c' to 'e' voltage rise by 9.45 volts, from 0.15 to 9.6 volts.
Any problem with this so far?

Here's yet another 3 terminal device:
Ground the 'e' terminal, pump 1ma into the 'b' terminal, pump 10ma into the 'c' terminal,
measure 150mv between the 'c' terminal and the 'e' terminal. Increase the current into the
'c' terminal to 200ma, measure 2.86 volts from 'c' to 'e' terminals.
Any problem with that?

Any problem with either of these two devices?

END QUOTE


Those are two pretty simple circuits, if you cant understand them then you cant understand a transistor either. All i ask is if you have some sort of problem with either of those two circuits, that all. I'd like to help end this "debate", so do you have a problem with either of them?

 

[Ratchit]

Jony130,

Thats becaues Ic-Vce curve don't come from the same manufacture.
The collector saturation region come from ON semi and Ic-Vce form Rohm.

That is a good explanatory answer.

Because we use the simulation model for 2N4401 I plot Ic = f (Vce) for Ib=const in LTspice (see attached curve)
And from Ic-Vce curve plot by LTspice it's clearly to see why in LTspice current stop increase when he reache 120mA.
And of course if we change IB then Ic current will stop increases by a different value of a Ic current (see attached file).

So the model parameters in the simulation are different than the manufacture's data? I find that disturbing.

Why to use simulation for such a simple calculation?
I think that everyone knows this equation:
V = L * dI/dt so for constant voltage in the inductor equation look like this
V = L *ΔI/Δt
So if we want to know the rate of current change (dI/dt) in amps per second we simple calculate:
ΔI/Δt = V/L = 3V/100µH = 30KA/s = 30mA/µs

You are right, I am very familiar with that equation. I don't know why I did not discern that. I had a temporary lapse of preception, I guess.

Ok, you have made your point. If you use the correct model or curve, you can determine the point of departure from saturation. Does that mean that a real 2N4401 will follow the manufacture's data and come out of saturation at 85 ma?

Ratch

 

[Ratchit]

MrAl,

No problem comprehending the operating points of either of those two circuits.

Ratch

 

[MrAl]

MrAl,

No problem comprehending the operating points of either of those two circuits.

Ratch

Hi again Ratch,


Oh ok, good, then you have no problem with understanding that with those two different circuits when we increase the current through the c to e terminals (current control) we get a rise in voltage between c and e also in BOTH of those circuits. So why then do you have a problem with the transistor where we increase Ic (current control, almost the same as Ice) and see an increase in Vce, as that is a three terminal device too?
I just dont see how you can understand TWO devices that have an increase in Ice and Vce increases, yet with the TRANSISTOR when we increase Ice and get an increase in Vce you dont understand how that can happen?
I'd really like to know how and why you see the transistor as different than the other two devices. Care to explain?

 

[Ratchit]

MrAl,

Oh ok, good, then you have no problem with understanding that with those two different circuits when we increase the current through the c to e terminals (current control) we get a rise in voltage between c and e also in BOTH of those circuits. So why then do you have a problem with the transistor where we increase Ic (current control, almost the same as Ice) and see an increase in Vce, as that is a three terminal device too?
I just dont see how you can understand TWO devices that have an increase in Ice and Vce increases, yet with the TRANSISTOR when we increase Ice and get an increase in Vce you dont understand how that can happen?
I'd really like to know how and why you see the transistor as different than the other two devices. Care to explain?

Sure, easy. I never had any problem comprehending that Vce would increase with increasing Ic. Just the bulk resistance in the transistor alone would do it. I had trouble seeing that the increase would be enough for the Ic involved. I saw from the simulation that the voltage would have to be 0.69 volts to bring it out of saturation. The now discredited chart from Fairchild showed much less than that. When Jody130 presented an accurate chart of what would happen, then I concurred that the current increase was enough to bring the transistor out of saturation.

Ratch

 

[MrAl]

Hello again,


Ok somehow you're still talking about saturation, but that's ok i guess. So now you agree that the transistor can be Ic controlled (ie current controlled) like those other two circuits i posted that also have three terminals?

 

[Ratchit]

MrAl,

Ok somehow you're still talking about saturation, but that's ok i guess. So now you agree that the transistor can be Ic controlled (ie current controlled) like those other two circuits i posted that also have three terminals?

Yes, it can.

Ratch

 

[MrAl]

MrAl,



Yes, it can.

Ratch

Thats great