Several years ago, I believe that amplifier must have THD as low as posible and full power bandwidth is several kHZ above 20kHz is enough. 14 years ago a measure my amplifier have THD 0,006% at 1kHz, 120W on 8,2 Ohm resistor. I satisfied the result until I built my VSSA variant.
Now, I know slew rate is importance too
I tried VAS current on my VSSA as low as 9mA, but the sound better when VAS current is around 13 - 18 mA. More VAS current is more slew rate, but depend on topology and the compensation.
It is all about trade-off. And I choose more slew rate than low THD. I have built amplifier with high slew rate and low slew rate on same topology.
It is nothing wrong use low current VAS or high current VAS. It depend on your goal. Even, you can make an amplifier with high distortion if you like the sound and then you can call it: sound effect.
Now, I know slew rate is importance too
I tried VAS current on my VSSA as low as 9mA, but the sound better when VAS current is around 13 - 18 mA. More VAS current is more slew rate, but depend on topology and the compensation.
It is all about trade-off. And I choose more slew rate than low THD. I have built amplifier with high slew rate and low slew rate on same topology.
It is nothing wrong use low current VAS or high current VAS. It depend on your goal. Even, you can make an amplifier with high distortion if you like the sound and then you can call it: sound effect.
Several years ago, I believe that amplifier must have THD as low as posible and full power bandwidth is several kHZ above 20kHz is enough. 14 years ago a measure my amplifier have THD 0,006% at 1kHz, 120W on 8,2 Ohm resistor. I satisfied the result until I built my VSSA variant.
Now, I know slew rate is importance too
But Slew Rate was not the only variable that was changed in your example
The question is how much of each you would tolerate (when there is trade-off).
Think about amplifiers that have high THD, but most of the THD comes from second order. A purist may say that distortion is distortion. But don't you know that when people said "marvelous vocal/midrange!" it is almost for sure that they are talking about second order distortion. Topology like TSSA for example, will give you more 2nd order dominance than VSSA topology, no matter how you tailor the VAS current etc.
You may not like 2nd order distortion, but many do. AND, many people can NOT hear the "pure" sound (from low distortion amplifier) anyway because their amplifier is a magnitude better in distortion than their speakers!
But Slew Rate was not the only variable that was changed in your example
The question is how much of each you would tolerate (when there is trade-off).
Yes, it seem impossible to design 2 amplifier with same specification except the slew rate.
I have built 2 amplifier using blameless/lin topology. One have high slew rate and other low slew rate. You must tried your self
I ask some experienced member and amplifier guru but no one can give the answer (when there is trade-off). So, I must do my experiments and still learning.... For now, my goal is full power bandwidth is 200 kHz minimum...
Ideal amplifier have no distortion. It sound good or not depend on who listening it. Blue is an color, it can be beautiful or not and you like it or not, but it still blue
I got to post11 and three times the question was asked of the OP "what is a driver"
I see a driver as the last device before the output pair of a Power Amplifier.
But the OP comes back with
A driver transistor for a power amplifier needs an Ic >=500mA and preferably ~1500mA
This is the peak base current that the output pair/s could demand in exceptional operating conditions. That demand cannot be met by a 50mA, nor a 100mA, device.
I see a driver as the last device before the output pair of a Power Amplifier.
But the OP comes back with
The OP does not seem to know what he is asking nor what he is being told.
A driver transistor for a power amplifier needs an Ic >=500mA and preferably ~1500mA
This is the peak base current that the output pair/s could demand in exceptional operating conditions. That demand cannot be met by a 50mA, nor a 100mA, device.
Of course I know what I'm asking. But not everybody knows what I'm asking. I also knows what I'm being told.
If you said "A driver transistor for a power amplifier needs an Ic >=500mA and preferably ~1500mA", of course I can debate if I want to, but I'm not that kind. Even tho my profession has been close to theory, Math, Physics, etc., I don't like talking about theory (especially debating on whether something is current drive or voltage drive, which I found silly). I'm a down to earth person and I read between the lines. If you make a typo for example, and if my analysis is that it is a typo, no way I will argue about it. Because what is important for me is the real value even if between the lines.
It does not read like you do.
That why I wrote "I got to post11 and three times the question was asked of the OP "what is a driver"........................
But the OP comes back with
"Actually I don't think that the "driver" is the point,
The OP does not seem to know what he is asking nor what he is being told."
So I return to the first post with your question:
Then we can make some progress, if progress is what you want.
That why I wrote "I got to post11 and three times the question was asked of the OP "what is a driver"........................
But the OP comes back with
"Actually I don't think that the "driver" is the point,
The OP does not seem to know what he is asking nor what he is being told."
So I return to the first post with your question:
What is your definition of a To126 Driver?
Then we can make some progress, if progress is what you want.
Last edited:
What progress??? The question has already been answered, don't you know that?
My closest friend once told my other friends "If Jay ask you a question, never assume it is a simple question. If it is a simple question, he won't ask you."
Are you confused, Andrew?
Go back to "sx-amp and nx-amp" thread and observe the last post.
You said: "...Fortunately there is a wide margin..............." then Terry came with a question: "How critical is the actual inductance?" and you answered: "I said in response to the question."
So how critical is the actual inductance?
Don't you know what I mean here?Not 100% impossible. Don't you know that when I presented two FFTs, I have adjusted some variables such that they are apple to apple. How I adjusted this variable, it's complex and I don't have to disclose my procedures. You have the option to judge that the data is garbage or the opposite, up to you. It is as is
I have built 2 amplifier using blameless/lin topology. One have high slew rate and other low slew rate. You must tried your self
I have built many amplifiers, Bimo. If you found out that SR is more important than what the theorists said, then read my older posts. I have been ahead of you at least 10 years in that matter
Do you know what the problem is? Because many experts do not have good ears and many golden ears do not understand Ohm law. What is the point telling numbers if you don't know how the numbers sound? What you need is to correlate the numbers with listening perception. And guaranteed, you will find strong correlation between sound quality and measurement.
You know electronics more than I do Bimo, but I'm serious when I told people that in my day-to-day activities with audio people, until now I have never met anyone that has better ear than mine. I have met fanatics, but no one is more fanatics than me in this hobby. You can mention names in this field (Jakarta-Bandung), I probably know them
That is what I want to really know: "to correlate the numbers with listening perception".
What maximum THD, minimum slew rate, minimum S/N ratio, minimum PSRR, maximum IMD that man can not detect?
It just more easier to achieve the specification rather than imagine how it sound. It is accurate enough? No? Why? I don't know... anybody care?
To do that you must have good ears
If you don't have good ears, it is up to you, whether you want to trust in others ears or trust in your own ears
To do that you must have good ears
Unfortunately, now I can not measure the specification when I understand how to achieve the specification that I want in simulation.
Then, I correlate the sound perception to the simulation result.
May be you need to compare measurement with simulation up until you find trusted models. Once you find the trusted models then you can rely on simulation imo. To tell you the truth, I don't always use a single model for my transistors, because different model has different purpose. I even use models that is modified to suit my need (and my specific parts).
I think I have never found a simulation that can not represent the actual listening perception. Yes, once, but may be I made mistake in the simulation. Or twice, but I made beginners mistake of trusting any models I could find. Or may be more than twice, but long time ago when I knew less.
I'm developing a method to optimize a circuit in LTSpice. This thread is basically my effort to validate the result. I picked highly regarded circuits and optimize it. Some I have built some still in simulation. Sometimes I'm amazed with my findings
I don't think so. Layout and component choice is importance to get measurement result is as close as possible to simulation result. I have many years experience to make PCB layout in commercial product (international brand). It is not so easy to make good layout.
I think it still importance to measure the specification.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.