Re: Re: voltage / current mode endless debate
A current source can also be modeled with a voltage source with a huge resistor......(Norton-Thevenin conversion)
...So....how often have you seen a nominal current source described as 'This-voltage-source-with-a-very-large-series-resistor' ??
Surely you don't think Cherry, Pass, Baxandal, etc are all wrong??
millwood said:
A current source can also be modeled with a voltage source with a huge resistor......(Norton-Thevenin conversion)
...So....how often have you seen a nominal current source described as 'This-voltage-source-with-a-very-large-series-resistor' ??
Surely you don't think Cherry, Pass, Baxandal, etc are all wrong??
Re: so ... what would you propose
So-called 'current' feedback is infact (deeply compromised!) voltage feedback, as the feedback network's transfer function is a voltage ....
.....and of equal importance, the foward path transfer function is a voltage also......before and after the feedback loop is closed...:
http://diyaudio.com/forums/showthread.php?postid=423522#post423522
It does not get plainer than this....
mirlo said:
So-called 'current' feedback is infact (deeply compromised!) voltage feedback, as the feedback network's transfer function is a voltage ....
.....and of equal importance, the foward path transfer function is a voltage also......before and after the feedback loop is closed...:
http://diyaudio.com/forums/showthread.php?postid=423522#post423522
It does not get plainer than this....
by the way, Texas Instruments has a new application note out in their Analog Applications Journal on current feedback filter design:
http://focus.ti.com/lit/an/slyt060/slyt060.pdf
jack
http://focus.ti.com/lit/an/slyt060/slyt060.pdf
jack
Re: Re: Re: Re: Re: Re: let us come to the point
How about a CB transistor stage?
Current in - voltage out, right!?
What do you see in most amplifiers called "current feedback" which in your opinion is a voltage feedback as the main issue in this thread is about?
A CB stage, right!?
BTW: I realize that we have also an amplifier topology which sense the current over a shunt resistor in series with the load, but such an amplifier "needs preferably" a high input impedance in the feedback node, is such one a "voltagae feedback" then by your definition Mikeks? (just the oposit way...)
Just curious when somebody is so "noisy" about something!
Another note: There's more then 4 diffrent amplifier in my opinion, input can be V or I, output can be V or I and the feedback input node can be V or I, theoretically that gives 8 all in all combinations to concider, right?
Anyhow, I don't know what to call it but I know how it works and what it use to be called in others textbook, enough for me.
Cheers!
mikeks said:
How about a CB transistor stage?
Current in - voltage out, right!?
What do you see in most amplifiers called "current feedback" which in your opinion is a voltage feedback as the main issue in this thread is about?
A CB stage, right!?
BTW: I realize that we have also an amplifier topology which sense the current over a shunt resistor in series with the load, but such an amplifier "needs preferably" a high input impedance in the feedback node, is such one a "voltagae feedback" then by your definition Mikeks? (just the oposit way...)
Just curious when somebody is so "noisy" about something!
Another note: There's more then 4 diffrent amplifier in my opinion, input can be V or I, output can be V or I and the feedback input node can be V or I, theoretically that gives 8 all in all combinations to concider, right?
Anyhow, I don't know what to call it but I know how it works and what it use to be called in others textbook, enough for me.
Cheers!
I think the manual about CFB active filters that jack posted is pointless…..since none of the Sallen and Key circuits presented measure the outgoing current of the circuit, there is no current feedback.
Furthermore all this talk about current feedback and voltage feedback is rather pointless since the discussion is about an already defined definition that none of the amp manufacturers either know, or they simply ignore it.
I think that the subject is mostly made up by marketing people that need new gimmicks in order to sell new amps.
If an amplifier used current feedback it would produce a current from the output terminals only proportional to the input signal. This would result in a voltage on the output that is proportional to the load of the amp.
Iout/Vin = 1/R1
Vout/Vin = (RL+R1)/R1
From this we see that the output current is independent of the load resistor RL. The feedback signal is a voltage, but the system generates an output current!
The terms Voltage feedback and Current feedback are related to the information of the feedbacksystem, and not the way the feedback signal is represented (voltage or current)
\Jens
Furthermore all this talk about current feedback and voltage feedback is rather pointless since the discussion is about an already defined definition that none of the amp manufacturers either know, or they simply ignore it.
I think that the subject is mostly made up by marketing people that need new gimmicks in order to sell new amps.
If an amplifier used current feedback it would produce a current from the output terminals only proportional to the input signal. This would result in a voltage on the output that is proportional to the load of the amp.
An externally hosted image should be here but it was not working when we last tested it.
Iout/Vin = 1/R1
Vout/Vin = (RL+R1)/R1
From this we see that the output current is independent of the load resistor RL. The feedback signal is a voltage, but the system generates an output current!
The terms Voltage feedback and Current feedback are related to the information of the feedbacksystem, and not the way the feedback signal is represented (voltage or current)
\Jens
It is a bit funny really. Long ago I brought this topic up in a
thread because my old textbook used the term current feedback
in the sense that Jens just described. Since I wasn't aware of
the conflicting terminologies at the time I was puzzled when
trying to understand what a CFB op amp was. At that time
nobody seemed interested and I was told not to pay attention
to old textbooks. Later the topic came up again when some
people used the term in the "old textbook sense" and there
were long discussions eventually leading to the recognition
of everybody that there are indeed conflicting terminologies.
Now we have the topic up for discussion again and the tendency
is, every new thread on this topic is longer than the previous
ones and for each thread, there is an increasing number of
people who insist on using the term "current feedback" in
the "old" sense. I wonder where the people went who told
me not to read those old books?
What do we know then?
1) Some, mostly old, textbooks uses the term current feedback
in the sense Jens described, but most textbooks call it
something else.
2) The op amp manufacturers uses the term current feedback
in a different sense (I am to lazy to give a precice description).
3) 1 & 2 means there is a terminological conflict which seems
to bother some people. I am sorry, that is real life. I think it
doesn't matter what subject you discuss or study, you will
find a lot of cases of conflicting terminology. Perhaps electronics
has less of such conflicts than most other subjects, but it would
be an illusion to believe there are no such conflicts.
4) The term "current feedback" per se tells nothing about
what it means. Any claim that it must mean something in
particular is based on further assumptions. By changing the
set of assumptions we can argue for a number of different
meanings of the term.
So, shall we agree to call current feedback in the old sense
"raspberry pie" and current feedback in the op amp manufacturers
sense "blueberry pie" and settle the dispute? I don't mind. From
a logic point of view the terms are just names anyway.
(Personally I have adopted the modern use of the term and
prefer to call the old type of current feedback for "current sense"
or "current drive" which both seem more informative to me.)
thread because my old textbook used the term current feedback
in the sense that Jens just described. Since I wasn't aware of
the conflicting terminologies at the time I was puzzled when
trying to understand what a CFB op amp was. At that time
nobody seemed interested and I was told not to pay attention
to old textbooks. Later the topic came up again when some
people used the term in the "old textbook sense" and there
were long discussions eventually leading to the recognition
of everybody that there are indeed conflicting terminologies.
Now we have the topic up for discussion again and the tendency
is, every new thread on this topic is longer than the previous
ones and for each thread, there is an increasing number of
people who insist on using the term "current feedback" in
the "old" sense. I wonder where the people went who told
me not to read those old books?
What do we know then?
1) Some, mostly old, textbooks uses the term current feedback
in the sense Jens described, but most textbooks call it
something else.
2) The op amp manufacturers uses the term current feedback
in a different sense (I am to lazy to give a precice description).
3) 1 & 2 means there is a terminological conflict which seems
to bother some people. I am sorry, that is real life. I think it
doesn't matter what subject you discuss or study, you will
find a lot of cases of conflicting terminology. Perhaps electronics
has less of such conflicts than most other subjects, but it would
be an illusion to believe there are no such conflicts.
4) The term "current feedback" per se tells nothing about
what it means. Any claim that it must mean something in
particular is based on further assumptions. By changing the
set of assumptions we can argue for a number of different
meanings of the term.
So, shall we agree to call current feedback in the old sense
"raspberry pie" and current feedback in the op amp manufacturers
sense "blueberry pie" and settle the dispute? I don't mind. From
a logic point of view the terms are just names anyway.
(Personally I have adopted the modern use of the term and
prefer to call the old type of current feedback for "current sense"
or "current drive" which both seem more informative to me.)
JensRasmussen said:
Nobody has been able to come up with any such new books
in the previous discussions, so that changes the situation
somewhat. I still use a textbook from 1968. However, please
note that there seems to be no general agreement on
terminology between the various textbooks. What is called
current feedback in some books is called a number of other
things in other books. At least the op amp manufacturers
seem to agree on what they mean by the term.
I have looked at the datasheets, and indeed AD844 claims to be a current feedback opamp. However, at the same time the datasheet tells us that the AD844 can be used as a substitute for “normal opamps”. This for me is a contradiction, as the word current-feedback to me either means something about the external feedback network, or some sort of internal compensation circuit.
If we talk about internal circuitry I don’t really care how the thing works internally, as long as it performs as an opamp. For me the statement in the datasheet is nice to know, but it does not change the fact that the opamp has to work as any other device.
Therefore I believe that the term current-feedback is kind of misplaced, because it to me means something about the external circuitry and not how the thing is made to perform the way it does.
In any case opamps are difficult to discuss because we don’t really know how they are build – because only a simplified schematic is presented.
As for discrete power amps, I have still to see a circuit here that conceptually works the way the opampcircuit I posted earlier. This is with good reason, because it would be useless as a voltage amplifier (What most people need to play music). I have seen voltage controlled current amplifiers used in setups used to measure Thiele & Small speaker parameters , but that’s a different story.
Just what I think
\Jens
If we talk about internal circuitry I don’t really care how the thing works internally, as long as it performs as an opamp. For me the statement in the datasheet is nice to know, but it does not change the fact that the opamp has to work as any other device.
Therefore I believe that the term current-feedback is kind of misplaced, because it to me means something about the external circuitry and not how the thing is made to perform the way it does.
In any case opamps are difficult to discuss because we don’t really know how they are build – because only a simplified schematic is presented.
As for discrete power amps, I have still to see a circuit here that conceptually works the way the opampcircuit I posted earlier. This is with good reason, because it would be useless as a voltage amplifier (What most people need to play music). I have seen voltage controlled current amplifiers used in setups used to measure Thiele & Small speaker parameters , but that’s a different story.
Just what I think
\Jens
If you look at figure 4 in the Analog Devices data sheet for the AD844 it is obvious that the internal architecture is that of a CFB amplifier.
Why are people finding the CFB architecture so difficult to comprehend? Just go back to first principles and look at how the signal passes through the amplifier. It's not rocket science!
And CFB architectures can work very well for audio. Just make sure you servo the dc offset.
Why are people finding the CFB architecture so difficult to comprehend? Just go back to first principles and look at how the signal passes through the amplifier. It's not rocket science!
And CFB architectures can work very well for audio. Just make sure you servo the dc offset.
Ok, here is a classic one:
http://www.bonavolta.ch/hobby/en/audio/hir20.htm
Now what would you call it? CFB or VFB?
http://www.bonavolta.ch/hobby/en/audio/hir20.htm
Now what would you call it? CFB or VFB?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.