Re: Re: Reply
TO UPDATE and FOR BETTER PERFORMANCE
AND I KNOW WHAT I AM DOING VERY WELL, SO DONT REMIND ME SUCH THING .
TELL ME IS THERE ANY WAY THAT what i am doing is truth not a fake thing, Since i dont have a camera to post the images on this forum.
peranders said:
TO UPDATE and FOR BETTER PERFORMANCE
AND I KNOW WHAT I AM DOING VERY WELL, SO DONT REMIND ME SUCH THING .
TELL ME IS THERE ANY WAY THAT what i am doing is truth not a fake thing, Since i dont have a camera to post the images on this forum.
Voltage feedback does not change with load, always reflecting a fixed ratio of output/input. Of course current will vary with load impedance, but load voltage remains a fixed factor of input voltage.
Current feedback senses load current and creates a feedback signal (usually voltage) in proportion to this current. With no load connected, there can be no feedback signal since there is no load current. This means that power delivery actually increases with increasing load impedance, like an OTL tube amp. The reverse is true with voltage feedback.
I see clear evidence that the feedback regime operates with the load disconnected. Gain is invariant whether load is connected or not. Ergo, it's voltage feedback, and furthermore there appears no combination of the two feedback modes.
I don't like the lack of a thermal compensation, and I'm fed up with animated bloody gifs.
You are to be commended for your willingness to contribute, Ampman, but your manners and contempt for others are appalling. It's nice to be mild and inoffensive with your words, lest you be forced the indignity of eating them.
Cheers,
Hugh
Current feedback senses load current and creates a feedback signal (usually voltage) in proportion to this current. With no load connected, there can be no feedback signal since there is no load current. This means that power delivery actually increases with increasing load impedance, like an OTL tube amp. The reverse is true with voltage feedback.
I see clear evidence that the feedback regime operates with the load disconnected. Gain is invariant whether load is connected or not. Ergo, it's voltage feedback, and furthermore there appears no combination of the two feedback modes.
I don't like the lack of a thermal compensation, and I'm fed up with animated bloody gifs.
You are to be commended for your willingness to contribute, Ampman, but your manners and contempt for others are appalling. It's nice to be mild and inoffensive with your words, lest you be forced the indignity of eating them.
Cheers,
Hugh
Lumanauw,
Since we have the ambiguity in terminology I will, for lack of
better names, refer to old vs. modern definitions of current
feedback.
The Hiraga amp is as far as I can see current feedback in
the modern sense. The two schematics, a and b, that you
posted next illustrate voltage and current feedback in the
old sense.
As other examples, amp_mans amplifier is current feedback in
the modern sense. AKSAs recent post used current feedback
in the old sense.
I am not claiming either definition is right or wrong. They are
both logical in a sense. The old definition is current feedback
in the sense that the feedback is proportional to the load
current (and then we can feed back either an error voltage
or an error current). The modern definition is current feedback
in the sense that the output voltage is proportional to the
error current flowing in to/out from the negative input, ie. the
feedback input, (and then this current may be proportional
to either the voltage over the load or the load current). It is
unfortunate to have this clash of terminologies, but diyAudio
is hardly a standardizing organisation and I don't think we
would even be able to enforce our own forum standard on
terminology. I have personally decided to go along with the
modern terminology since that is how most people seem
to use it and is also consistent with how op amp manufacturers
use it. The question then is what to call voltage/current feedback
in the old sense. Voltage/current drive or voltage/current sense
seem appropriate, with a personal preference for the first of
these.
It should also be noted that while op amps are characterized
as voltage or current feedback in the modern sense, they
could not be classified as either in the old sense, since the
old definition relies on the feedback network arrangement
which is exterior to the op amp. A complete audio amplifier
on the other hand could be classified according to both
terminologies.
Since we have the ambiguity in terminology I will, for lack of
better names, refer to old vs. modern definitions of current
feedback.
The Hiraga amp is as far as I can see current feedback in
the modern sense. The two schematics, a and b, that you
posted next illustrate voltage and current feedback in the
old sense.
As other examples, amp_mans amplifier is current feedback in
the modern sense. AKSAs recent post used current feedback
in the old sense.
I am not claiming either definition is right or wrong. They are
both logical in a sense. The old definition is current feedback
in the sense that the feedback is proportional to the load
current (and then we can feed back either an error voltage
or an error current). The modern definition is current feedback
in the sense that the output voltage is proportional to the
error current flowing in to/out from the negative input, ie. the
feedback input, (and then this current may be proportional
to either the voltage over the load or the load current). It is
unfortunate to have this clash of terminologies, but diyAudio
is hardly a standardizing organisation and I don't think we
would even be able to enforce our own forum standard on
terminology. I have personally decided to go along with the
modern terminology since that is how most people seem
to use it and is also consistent with how op amp manufacturers
use it. The question then is what to call voltage/current feedback
in the old sense. Voltage/current drive or voltage/current sense
seem appropriate, with a personal preference for the first of
these.
It should also be noted that while op amps are characterized
as voltage or current feedback in the modern sense, they
could not be classified as either in the old sense, since the
old definition relies on the feedback network arrangement
which is exterior to the op amp. A complete audio amplifier
on the other hand could be classified according to both
terminologies.
Commentable thoughts
I completely agree with Christer that one has to use modern terminology regardin CFBamp to understand it fully in terms of modern phenomena and from what i think is that Christer has a better understanding of Current Feedback according to modern terminology then all of u have!
I completely agree with Christer that one has to use modern terminology regardin CFBamp to understand it fully in terms of modern phenomena and from what i think is that Christer has a better understanding of Current Feedback according to modern terminology then all of u have!
Commentable thoughts
HEY CHRISTER PLZ accept my Generous Friendship in terms of Current Feedback Amplifier!😀
HEY CHRISTER PLZ accept my Generous Friendship in terms of Current Feedback Amplifier!😀
Re: Commentable thoughts
I don't claim to have a very good understanding of the modern
CFB concept, there are many many on this forum who understand
it far better than I do. I understand the basic principle of how it
works, but there are many subtle issues I am still trying to
understand better. Stability issues are non-trivial, and more
difficult to understand than for VFB amps, for instance.
However, as I pointed out earlier, I learnt to use CFB in the
old way and then realized, after a lot of pain, that it is actully
used in a different way nowadays. Most others seem to have
learnt the concept in either of the two possible meanings
without knowing there are different definitions. I thus seem
to have a better understanding than most others why this
concept causes so much confusion. That does not mean I
understand CFB itself better than others.
amp_man_1 said:
I don't claim to have a very good understanding of the modern
CFB concept, there are many many on this forum who understand
it far better than I do. I understand the basic principle of how it
works, but there are many subtle issues I am still trying to
understand better. Stability issues are non-trivial, and more
difficult to understand than for VFB amps, for instance.
However, as I pointed out earlier, I learnt to use CFB in the
old way and then realized, after a lot of pain, that it is actully
used in a different way nowadays. Most others seem to have
learnt the concept in either of the two possible meanings
without knowing there are different definitions. I thus seem
to have a better understanding than most others why this
concept causes so much confusion. That does not mean I
understand CFB itself better than others.
Christer,
It confuses me too. So, like you said, what will the "modern terminology" call the B schematic?
From its terminology, the feedback current is injected to the transistor (via emitors in this case). What if we use single differential pairs, not N+P transistor. How do you inject feedback to perform such as called "Current Feedback Amplifier"? Wouldn't it make us conclude to a funny conclusion, that it is "impossible" to make Current Feedback Amp with single differential?
It confuses me too. So, like you said, what will the "modern terminology" call the B schematic?
From its terminology, the feedback current is injected to the transistor (via emitors in this case). What if we use single differential pairs, not N+P transistor. How do you inject feedback to perform such as called "Current Feedback Amplifier"? Wouldn't it make us conclude to a funny conclusion, that it is "impossible" to make Current Feedback Amp with single differential?
lumanauw said:
I don't if there is an agreed upon name for the B schematic.
According to the book Pjacobi referred to earlier it should
obviously be called current control. I have also seen people
refer to it as current drive, especially in the loudspeaker
forum.
As for the second question, differential pairs are used only
in VFB amps as far as I know. Maybe you could do some trick
to use it also in a CFB amp, but you would probably miss the
point of CFB then. The fundamental difference between VFB
and CFB lies in the very different types of input stage. A VFB
amp has two high-impedance inputs and outputs a voltage
proportional to the voltage difference between the inputs.
This is typically achieved by using a differential pair. A CFB
amp has a high-impedance positive input and a low-impedance
negative input (this is very important). As I explained earlier
in the thread (and this is very confusing when first encountering
this concept) the negative input is actually an output!! The
input stage is a buffer connected s.t. the positive input is
a high impedance input going into a unity gain buffer and
the negative input of the amplifier is the output of this buffer,
which has a low output impedance. This means that the input
stage tries to keep the negative input at the same voltage as
the positive input. Since the buffer does not have zero impedance,
it will not be able to keep exactly the same voltage so a current
will flow in or out through the negative input (ie. the buffer
output). The output voltage of the amplifier is proportional to
this current!!
I again highly recommend reading the following application notes:
Intersil AN9420.1 and Texas Instruments SLVA051 and SLOA021.
half knowledge is dangerous
The old current feedback = the feedback of the sensed current of the load.
Purpose could be to make sure that the current through the load will be proportional to the input voltage of the amp.
This is the case when the load is not purely resistive for instance.
Like in one of that schematics posted here, where a small resistor is connected between load and ground.
I learned that from Nelson 🙂 when I built my first dynamic current source, and I had to sense the load current and not the output voltage of the amp.
The old current feedback = the feedback of the sensed current of the load.
Purpose could be to make sure that the current through the load will be proportional to the input voltage of the amp.
This is the case when the load is not purely resistive for instance.
Like in one of that schematics posted here, where a small resistor is connected between load and ground.
I learned that from Nelson 🙂 when I built my first dynamic current source, and I had to sense the load current and not the output voltage of the amp.
Christer,
We can certainly make so called CFB according to fig. B (old terminology) with single differential. But cannot make CFB according to (modern terminology) by using single differential. So if people ask you "Can you make CFB audio power amp with single differential?" what will be the answer?
The way I look amp-man input stage is rather different. As well as for Hiraga amp example.
I took it this way. Read on Nelson Pass article on Mosfets. It is stated clearly with drawings, that there are 3 ways to use transistors. The only one that gives inverted output is collector output.
In this amp-man particular schematic, I consider there are 2 inputs that the results are superpositioned to each other. One is input signal from the base. This results in inverted signal in the collector.
The second is the signal injected to its emitor. This is similiar to common base operation of transistor, thus it will gives the same phase signal in the collector as in the emitors (where it comes in).
Superpositioning this 2 together in the collectors, results a steady gain factor. Which ever is superior or inferior, the rest of the circuit will automaticly try to balanced it again.
This is similiar to the operation of differential, but in differential pair, the signal fed to the emitor is fed by other emitor of the transistor pair, because the CCS that fed this differential have high impedance. The signal from the other emitor is from its base (have input from the feedback network, concerning the drops are the same 0.6 volts for the same differential transistors, so whatever happens in one base, the other base will try to match it).
We can certainly make so called CFB according to fig. B (old terminology) with single differential. But cannot make CFB according to (modern terminology) by using single differential. So if people ask you "Can you make CFB audio power amp with single differential?" what will be the answer?
The way I look amp-man input stage is rather different. As well as for Hiraga amp example.
I took it this way. Read on Nelson Pass article on Mosfets. It is stated clearly with drawings, that there are 3 ways to use transistors. The only one that gives inverted output is collector output.
In this amp-man particular schematic, I consider there are 2 inputs that the results are superpositioned to each other. One is input signal from the base. This results in inverted signal in the collector.
The second is the signal injected to its emitor. This is similiar to common base operation of transistor, thus it will gives the same phase signal in the collector as in the emitors (where it comes in).
Superpositioning this 2 together in the collectors, results a steady gain factor. Which ever is superior or inferior, the rest of the circuit will automaticly try to balanced it again.
This is similiar to the operation of differential, but in differential pair, the signal fed to the emitor is fed by other emitor of the transistor pair, because the CCS that fed this differential have high impedance. The signal from the other emitor is from its base (have input from the feedback network, concerning the drops are the same 0.6 volts for the same differential transistors, so whatever happens in one base, the other base will try to match it).
I think most people here miss the fact that a "normal" amp has
voltage in- voltage out (without feedback)
but we (I at least) mean an amp with CURRENT in (between inverting and non-inverting input) generates a voltage out (without feedback).
voltage in- voltage out (without feedback)
but we (I at least) mean an amp with CURRENT in (between inverting and non-inverting input) generates a voltage out (without feedback).
- Status
- Not open for further replies.