Sort of, what is sure is that whatever mod you ll do to this half baked design you ll hardly make worse than the original schematic.
The current original MF-A1 (model 2023) has this coil.
I don't know whether "Bernhard" will use it in his friendly test setup.
Unfortunately, this does not change the current wrangling over the stability issue. The root of the problem lies in the utopian demand for THD s of less than 0.0005% marks under full load.
short answer, sorry (I'm fed up a little)
@m0rten,
Thank you very much for your trust and cooperation.
The layout already looks great, there are still at least six small components missing, namely the limiter - I'll come back to that later.
So wait a moment with further activities, I would like to have the absolute stability issue and other small details clarified here on this board and especially in this thread.
The permanent bashing is getting on my nerves too much.
Thank you very much for your trust and cooperation.
The layout already looks great, there are still at least six small components missing, namely the limiter - I'll come back to that later.
So wait a moment with further activities, I would like to have the absolute stability issue and other small details clarified here on this board and especially in this thread.
The permanent bashing is getting on my nerves too much.
@wahab,
Why are you blowing the same hackneyed horn? Is it fun and does it promote your own satisfaction?
It would be constructive if you could explain to the readers (newcomers) why the two-stage (but symmetrical between the rails) **** has very specific limitations.
And perhaps where the fundamental differences to the EF lie.
Basically the whole program.
I'm so shocked now that I'm about to throw my original MF-A1 out the window - the topology is so bad compared to D. Self's “Blameless baptism”.
regards,
HBt.
PS
The Bernhard charge is in the room, can you show us “how we keep proposal 98, an iteration or evolution step, stable while maintaining the utopian THD”?
So let's stop the stopwatch at this point and jump back to my posting #28.
Maybe this will help the opposing team to play by the rules in the future.
This thread is not about the famous three-stage concept, with the EF-PP-OPS and the simple entry stage.
Maybe this will help the opposing team to play by the rules in the future.
This thread is not about the famous three-stage concept, with the EF-PP-OPS and the simple entry stage.
I'm talking about your Call for ...
As a good role model, you should of course lead the way.
But now it turns out that your interests lie in completely different areas.
Of course.
You wanted to bring down this thread from the start and nothing else.
I take my leave for the time being,
HBt.
PS
A short break seems to be necessary.
Basically still:
the comparison between apples and pears is simply not valid, either stewed apples or stewed pears, stewed apples is not stewed pears and vice versa. But both remain stewed fruit.
Apparently we had agreed on a type of fruit - and now it's time to prepare a wholesome dish.
The list of ingredients and the recipe are ready, now all that remains is to settle the usual and stupid argument about the amount of sugar.
Didn't you ask me to do a simulation ? Now you don't like the results ?
Why don't you post your ac analysis screenshot, together with the schematic for reference ? Some problem with that ?
Where did you get the idea that I would built a prototype of your schematic ?
To the soldering irons, diy-warriors! - was meant as a motivational call directed to you and your team of course.
It seems hard for you to accept that the A1 output stage is inferior to EF. According to my simulation, EF is still 20 dB better.
Your choice is ideological.
Why don't you post your ac analysis screenshot, together with the schematic for reference ? Some problem with that ?
Where did you get the idea that I would built a prototype of your schematic ?
To the soldering irons, diy-warriors! - was meant as a motivational call directed to you and your team of course.
It seems hard for you to accept that the A1 output stage is inferior to EF. According to my simulation, EF is still 20 dB better.
Your choice is ideological.
"Schonungslos" - relentless
Then devote yourself from now on to the complete, factual analysis of the circuit from the initial post (let's just start with that again).
This analysis also includes expert answers to all the why questions raised.
Stand-alone statements are completely worthless, as is insisting on the EF number or the BJT thingy at this point ... You know what I'm saying.
If you want to expose the 2N3055 as a piece of crap, you must also correctly explain the differences in use as an EF or common emitter ... and so on, just the full program.
But as I said before: this game is over. Furthermore, this is not a rhetorical question and answer session.
HBt.
Then devote yourself from now on to the complete, factual analysis of the circuit from the initial post (let's just start with that again).
This analysis also includes expert answers to all the why questions raised.
Stand-alone statements are completely worthless, as is insisting on the EF number or the BJT thingy at this point ... You know what I'm saying.
If you want to expose the 2N3055 as a piece of crap, you must also correctly explain the differences in use as an EF or common emitter ... and so on, just the full program.
But as I said before: this game is over. Furthermore, this is not a rhetorical question and answer session.
HBt.
The answer to this stupid continuous nagging EF is better than an emitter circuit, especially with the same BJT type from the box “not fast enough”, is so trivial that I can't bring myself to say it.
As soon as this trump card no longer works, one take the utopic THD out of the box, and if this nonsense no longer works, you move on to the stability of such a strongly feedback system. Open up a side issue with the buzzwords signal path, number of PN transitions or AC input coupling /cap ...
I'm completely fed up, I'm sorry.
If the eternally nagging opposing party is unable to force the supposedly unstable 98, i.e. oscillating disaster, to calm down (in the simulator) - and unfortunately I'm assuming that it will, I won't leave M0rton and other interested parties out in the cold.
But perhaps you will see it as a challenge - and accept it. It is indeed a challenge.
As soon as this trump card no longer works, one take the utopic THD out of the box, and if this nonsense no longer works, you move on to the stability of such a strongly feedback system. Open up a side issue with the buzzwords signal path, number of PN transitions or AC input coupling /cap ...
I'm completely fed up, I'm sorry.
If the eternally nagging opposing party is unable to force the supposedly unstable 98, i.e. oscillating disaster, to calm down (in the simulator) - and unfortunately I'm assuming that it will, I won't leave M0rton and other interested parties out in the cold.
But perhaps you will see it as a challenge - and accept it. It is indeed a challenge.
Side note
The simple circuit (or stage) from posting #157 has a real bandwidth of > 1MHz, under 5Wrms load conditions - and also a fairly large dynamic input resistance.
The characteristic values that we inevitably need for our utopian THD in a two-stage VAS (i.e. A1), in a gnfb system, have not yet been addressed or even defined at this point.
Everything is simply missing; apart from the destructive MC12-supported nagging, there is no lack of that.
I'm out of the number from now on,
HBt.
The simple circuit (or stage) from posting #157 has a real bandwidth of > 1MHz, under 5Wrms load conditions - and also a fairly large dynamic input resistance.
The characteristic values that we inevitably need for our utopian THD in a two-stage VAS (i.e. A1), in a gnfb system, have not yet been addressed or even defined at this point.
Everything is simply missing; apart from the destructive MC12-supported nagging, there is no lack of that.
I have understood you and your claqueurs.
I'm out of the number from now on,
HBt.
This has already been explained, a transimpedance stage has high output Z while a EF has low output Z, in the case
of the MFA1 that s even worse as the 2N3055/MJ2955 has to provide both a current gain and a voltage gain
and doing both at the same time will yield very poor linearity, indeed using the VAS as final stage is a very
bad idea.
With a total of 4 transistors in the chain that make about 10^8 current gain, a blameless has 10^10 current gain,
whatever the way this gain will be used, either as voltage ampification or as current amplification.
No need to do so, even if it s more difficult phisicaly to upgrade than the 70s era Dual amps that were fully modular.
Current mirrors would improve much this amp, but that s not enough, the 2N3055/MJ2955 should be replaced by a pair
of BDX66/67 or BDV64/65 to increase the whole chain s power gain, still, that would be inferior to a blameless using
the same component count but with the usual IPS + enhanced VAS + EF2 arrangement instead of the MFA1 s weird
IPS + enhanced VAS configuration.
The whole theater put forward with the remarkable but regularly introduced sideshows - it must be a joke.
Purely to pass the time, of course. I no longer assume that there could be a misunderstanding or a lack of specialist knowledge ...
In the so-called
It should also be noted that the designer determines the dynamic and static output resistance of this A1-like output stage. A corresponding negative feedback is also present at the end.
With the greatest of respect, but I can't go any further into this nonsense - it's too stupid for me.
Goodbye,
HBt.
PS
The criticized working basis, i.e. circuit, from my posting #98 works perfectly (with my models) both in the modern MC12 and in my ancient PSPICE, completely without oscillation.
However, one point is really important with the A1 concept (including the offshoots): it must not be led into any form of saturation at any point. I.e. in our case P=20W: Vout=12.65V & Iout=1.58A (all RMS) marks the end of the line.
An input-side band limitation and a limiter should be used. How these are correctly dimensioned depends, among other things, on the selected overall amplification factor (and we are not talking about current gain here!), which should not actually be less than 24dB for the A1.
What is the situation? We have an idea with a THD_1k of significantly less than 0.001% under full load.
It's hard to call something like that scrap.
Purely to pass the time, of course. I no longer assume that there could be a misunderstanding or a lack of specialist knowledge ...
If we want to draw a correct picture, then the following is true:
In the so-called
- "Emitterschaltungt" (common emitter), the dynamic output resistance (viewed from the collector electrode) is very large compared to the load (or also called working resistance)
- "Kollektorschaltung" (emitter follower) the case is completely different, here, there is already a maximum local negative feedback through the load, this load now looks into the emitter electrode - and sees a small dynamic output resistance ... always in comparison to something!
This statement is also nothing more than a pure generalization without specifying the working environment and working conditions. And therefore worthless and even incorrect.
It should also be noted that the designer determines the dynamic and static output resistance of this A1-like output stage. A corresponding negative feedback is also present at the end.
I beg your pardon?
With the greatest of respect, but I can't go any further into this nonsense - it's too stupid for me.
Goodbye,
HBt.
PS
The criticized working basis, i.e. circuit, from my posting #98 works perfectly (with my models) both in the modern MC12 and in my ancient PSPICE, completely without oscillation.
However, one point is really important with the A1 concept (including the offshoots): it must not be led into any form of saturation at any point. I.e. in our case P=20W: Vout=12.65V & Iout=1.58A (all RMS) marks the end of the line.
An input-side band limitation and a limiter should be used. How these are correctly dimensioned depends, among other things, on the selected overall amplification factor (and we are not talking about current gain here!), which should not actually be less than 24dB for the A1.
What is the situation? We have an idea with a THD_1k of significantly less than 0.001% under full load.
It's hard to call something like that scrap.
For me, the thread is dead and I ask that you refrain from spreading total nonsense. I'm logging off here, but will be happy to help via the PM function.
If M0rton still wants to finish the PCB project, I'd love it and would be delighted.
Warning
Possible instabilities cannot be ruled out in any project. There are exceptions, but A1 ideas are not one of them.
If M0rton still wants to finish the PCB project, I'd love it and would be delighted.
Warning
Possible instabilities cannot be ruled out in any project. There are exceptions, but A1 ideas are not one of them.