Amp-man,
A perfect Real time amplification is not possible as long as there is an inherent inductance and capacitance inside the circuit. I've ask the forum before and the answer is like that.
C and L will perform phase shift, the bigger the frequency, the more apparent they are.
You can do this experiment. Take 2 ch osciloscope and feed a signal generator into an amp. Ch 1 connects to the signal generator and ch2 connects to amp output (with scale offcourse, so the display in the osciloscope is the same). Put the osciloscope to dual trace (display both traces).
Swep the frequency from 0hz to 100khz. You will see some phase shift there, the input is not exactly on the same place as the output.
This applies to all analog audio power amps. Dont know about the digital amp.
What if we eliminate all the capacitors? Still cannot do it, because even transistors have inherent capacitance inside it, the pcb tracks will make inductance, etc. What we can make is to make the shift as little as possible in the range of 20-20khz.
Amp-man, why do you change your footage to that "LITTLE KNOWLEDGE IS DANGEROUS THING"? What happened to the previous footage?
A perfect Real time amplification is not possible as long as there is an inherent inductance and capacitance inside the circuit. I've ask the forum before and the answer is like that.
C and L will perform phase shift, the bigger the frequency, the more apparent they are.
You can do this experiment. Take 2 ch osciloscope and feed a signal generator into an amp. Ch 1 connects to the signal generator and ch2 connects to amp output (with scale offcourse, so the display in the osciloscope is the same). Put the osciloscope to dual trace (display both traces).
Swep the frequency from 0hz to 100khz. You will see some phase shift there, the input is not exactly on the same place as the output.
This applies to all analog audio power amps. Dont know about the digital amp.
What if we eliminate all the capacitors? Still cannot do it, because even transistors have inherent capacitance inside it, the pcb tracks will make inductance, etc. What we can make is to make the shift as little as possible in the range of 20-20khz.
Amp-man, why do you change your footage to that "LITTLE KNOWLEDGE IS DANGEROUS THING"? What happened to the previous footage?
lumanauw said:
The answer clearly dependes on what the questioner means
by CFB, that is, which one of the definitions.
As for the rest of your post, I don't think I can explan more
clearly than I did without getting mathematical, so please
read the application notes I suggested if you don't believe
me.
peranders said:
Ah, ah, careful now Per-Anders. It is not the current between
(if a curent can be between at all) the inputs, but the current
flowing in the inverting input only that is interesting. This is
not the same as the current flowing in the non-inverting
input, since the latter is usually high-impedance. I guess one
could make a CFB amp with both inputs low-impedance and
the currents equal, but that would be an unusual variant.
lumanauw said:
Hi
With a current feedback amplifier ,you can either make a voltage amplifier or a current amplifier...if you sample the output voltage you have a voltage amplifier if you sample the output current (via a series resistor) you have a current (or tranconductance ) amplifier.
What make a current feedback amplifier is not the end or the pourpose of the amplifier( to output a voltage proporcional to the input voltage or a current proporcional to the input voltage)but the way in wich the output sample (voltage or current) is injected in the input.In the current feedback amp because the inverting input has a low impedance the feedback is injected in current form.
And it make sense call this type of amplifier current feedback because it perform diferentely in tems of bandwith and slew rate than a conventional voltage feedback amplifier.
I am adding some more data to the following list that was posted earlier:
Sample output - Type of input - Name
1.Voltage - Voltage - Voltage feedback
2.Current - Voltage - Current feedback
3.Voltage - Current - Operational voltage feedback
4.Current - Current - Operational current feedback
( assuming "type of input" means - how the sampled signal appears at the input )
Effect on impedances:
1. Input Z goes up - Output Z goes down
( Voltage series NFB )
2. Input Z goes up - Output Z goes up
( Current series NFB)
3. Input Z goes down - Output Z goes down
( Voltage shunt NFB )
4. Input Z goes down - Output Z goes up
( Current shunt NFB )
Call it what you want , but this is how they relate to each other.
Cheers.
Sample output - Type of input - Name
1.Voltage - Voltage - Voltage feedback
2.Current - Voltage - Current feedback
3.Voltage - Current - Operational voltage feedback
4.Current - Current - Operational current feedback
( assuming "type of input" means - how the sampled signal appears at the input )
Effect on impedances:
1. Input Z goes up - Output Z goes down
( Voltage series NFB )
2. Input Z goes up - Output Z goes up
( Current series NFB)
3. Input Z goes down - Output Z goes down
( Voltage shunt NFB )
4. Input Z goes down - Output Z goes up
( Current shunt NFB )
Call it what you want , but this is how they relate to each other.
Cheers.
Tube_Dude said:
If you would have read the previous posts you would have understood that originally the term current feedback was used to denote feedback developed from the output current. Your definition is a quite recent one, coming from the development of opamps with a low impedance input node.
Now, you are free to use whatever of the two definitions you want, but if you say yours is THE right one, you close your eyes for at least 50 years of development, literature and teaching. Sounds to me like putting your head in the sand.
Also, I have no idea why you think that "it make sense call this type of amplifier current feedback because it perform diferentely in tems of bandwith and slew rate than a conventional voltage feedback amplifierw". I might as well say "it makes sense to call this type of amp a current feedback amp because it uses one resistor less in the feedback path" Both statements are logically nonsense.
Jan Didden
From your post it seems that i have discovered and invented the definition of current feedback...but it was not me (unhappily!!)
The concept is widely used in books..apllications notes...trade marked comercial amplifiers manufacturers...etc
So if you don't understand it or you don't want to use it...please feel free!!
janneman said:
Ok i will put in another way...in a car anallogy!
The Diesel engines have more than 50 years of devlopment...but today new technologies have been explorered as the commom rail diesel engines (CDI).
When we say a commom rail diesel engine we are being more especific than to say simply a diesel engine..
The some with amplifiers..if you say a current feedback voltage amplifier you are more especific and with a better explanation of the type and the perfomance of the amplifier!
Ok, let me try this:
You said:
What make a current feedback amplifier is not the end or the pourpose of the amplifier( to output a voltage proporcional to the input voltage or a current proporcional to the input voltage)but the way in wich the output sample (voltage or current) is injected in the input.In the current feedback amp because the inverting input has a low impedance the feedback is injected in current form.
That for me is your definition of CF. There are many, many here who do not agree. They have every reason to say that THEIR definition is the right one, because they learned it at school or whatever.
By denying that there are several definitions, you keep up the confusion.
You car analogy is flawed. A better analogy is to say that someone decided to call the new hybrid cars with combined electrical and diesel drive a diesel car. Then if you talk about a diesel car, which one do you mean? The original diesel car or the new hybrid?
If you talk to me about a CF amp, I'm not sure which one you mean (well, I do now).
I don't know how to solve this.
And also, amp_mans amplifier is for me clearly an VF amp, because although the return node looks rather low (the 120 Ohms resistors in one of the first diagrams), his feedback network is modelled as a conventional voltage divider giving voltage output. You see what a mess this new definition gives?
The old one was clear and could not be mistaken: if you short the output, and the feedback goes to 0, that's VF. If you disconnect the load, and the feedback goes to 0, that's CF. No difference of interpretation, clear, everybody knows what it is.
But I realise I am fighting a lost battle.
Jan Didden
You said:
What make a current feedback amplifier is not the end or the pourpose of the amplifier( to output a voltage proporcional to the input voltage or a current proporcional to the input voltage)but the way in wich the output sample (voltage or current) is injected in the input.In the current feedback amp because the inverting input has a low impedance the feedback is injected in current form.
That for me is your definition of CF. There are many, many here who do not agree. They have every reason to say that THEIR definition is the right one, because they learned it at school or whatever.
By denying that there are several definitions, you keep up the confusion.
You car analogy is flawed. A better analogy is to say that someone decided to call the new hybrid cars with combined electrical and diesel drive a diesel car. Then if you talk about a diesel car, which one do you mean? The original diesel car or the new hybrid?
If you talk to me about a CF amp, I'm not sure which one you mean (well, I do now).
I don't know how to solve this.
And also, amp_mans amplifier is for me clearly an VF amp, because although the return node looks rather low (the 120 Ohms resistors in one of the first diagrams), his feedback network is modelled as a conventional voltage divider giving voltage output. You see what a mess this new definition gives?
The old one was clear and could not be mistaken: if you short the output, and the feedback goes to 0, that's VF. If you disconnect the load, and the feedback goes to 0, that's CF. No difference of interpretation, clear, everybody knows what it is.
But I realise I am fighting a lost battle.
Jan Didden
It always results in a mess when somebody starts using a
well established term for something else and manages, by
marketing or whatever other means, to get a lot of people
to use it. However, done is done, and it is too late to do
anything about it in the case of voltage/current feedback, so
we have to live with the double meaning and confusion for
the forseeable future I am afraid.
Computers has caused a similar confusion. We all knew
what kilo, mega and giga was, until computer people (that
is people like myself
) started using these terms for
1024, 1024^2 and 1024^3. To mess things up even more,
even the computer industry uses the terms in both ways.
If we talk about internal RAM memory in a computer, then
a gigabyte is 1024^3 bytes, but harddisk capacity is measured
in the "normal" way so there a gigabyte is 1000^3 (or maybe
it is even messier as 1024^2*1000, I am not sure). Don't even
ask me how many cycles per second a GHz is in a CPU, ask Intel
or AMD, since I don't know.
Actually, I read a few years ago that there is now an ISO
standard saying that the prefixes for 1024 etc. should be
called kibi, mibi and gibi in order to avoid confusion, but so
far I have never seen it in use.
well established term for something else and manages, by
marketing or whatever other means, to get a lot of people
to use it. However, done is done, and it is too late to do
anything about it in the case of voltage/current feedback, so
we have to live with the double meaning and confusion for
the forseeable future I am afraid.
Computers has caused a similar confusion. We all knew
what kilo, mega and giga was, until computer people (that
is people like myself
) started using these terms for1024, 1024^2 and 1024^3. To mess things up even more,
even the computer industry uses the terms in both ways.
If we talk about internal RAM memory in a computer, then
a gigabyte is 1024^3 bytes, but harddisk capacity is measured
in the "normal" way so there a gigabyte is 1000^3 (or maybe
it is even messier as 1024^2*1000, I am not sure). Don't even
ask me how many cycles per second a GHz is in a CPU, ask Intel
or AMD, since I don't know.
Actually, I read a few years ago that there is now an ISO
standard saying that the prefixes for 1024 etc. should be
called kibi, mibi and gibi in order to avoid confusion, but so
far I have never seen it in use.
janneman said:
Yes it is...but not only for me...the textbook definition also.
As i told you in a previous post...if you and some other people don't agree ...what can i do?...Change the world ?
Who is keeping up the confusion? The people that try to express some news views about technologies or the people that (as allways) are afraid of changing his points of vue!
Of coarse....as i espected your analogy is the right one!!!
Me too!!!
Tube_Dude said:
Ah, the textbook!! Which one is that? I thought my book
was a textbook too, but then I am obviously mistaken since
it says something different.
Obviously, there are still various terminologies around also
at the universities. Just as an example from some googling,
they seem to use an entirely differen terminology in Dublin
http://www.electronics.dit.ie/staff/ypanarin/Lecture Notes/K235-1/3 Feedback.pdf
Who's censoring the internet???
I've done some more googling and according to the brief text
excerpts one gets in the hit list, there are still universities and
other organizations who use/teach the old terminology. For
some very strange reason (yes, I know it has to be pure
coincidence) almost none of these documents are available, they
are either not found, blank or hang my browser. At least I found
evidence that there are still american universities who teach
their students the old terminology:
http://www.ee.siue.edu/~alozows/courses/ece327/notes.pdf
(Yes, it is AFAIK a rather unknown university so maybe they
are still 30 years behing in Edwardsville, IL?)
NB, I am not proposing to use the old terminology, I have myself
adopted the new meaning of the term CFB, but it is obviously
somewhat stupid to claim that there is only one right meaning
of the term. We could just as well discuss forever whether a
certain animal should be called horse or cheval. Some would
even prefer to call it Pferd or cavallo, and janneman and
Tube_dude would probably have their "own" names for it.
Me, I know you are all wrong. I have known since childhood
that it is a häst.
I've done some more googling and according to the brief text
excerpts one gets in the hit list, there are still universities and
other organizations who use/teach the old terminology. For
some very strange reason (yes, I know it has to be pure
coincidence) almost none of these documents are available, they
are either not found, blank or hang my browser. At least I found
evidence that there are still american universities who teach
their students the old terminology:
http://www.ee.siue.edu/~alozows/courses/ece327/notes.pdf
(Yes, it is AFAIK a rather unknown university so maybe they
are still 30 years behing in Edwardsville, IL?)
NB, I am not proposing to use the old terminology, I have myself
adopted the new meaning of the term CFB, but it is obviously
somewhat stupid to claim that there is only one right meaning
of the term. We could just as well discuss forever whether a
certain animal should be called horse or cheval. Some would
even prefer to call it Pferd or cavallo, and janneman and
Tube_dude would probably have their "own" names for it.
Me, I know you are all wrong. I have known since childhood
that it is a häst.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.