Correlation between measurement methods and real audio evaluation
Wonderfull Eva!
That's the problem.... i think engineers must turn the first street left, and the second rigth... stop in one human Sciences University and study a little some "human Perception"...or some books can help too.
PMA asked...how can you explain audibility.
You can measure also:
When sounds good... your heart beats increase
Your arterial pression increase, despite your muscles relax.
Your eyes have one "apperture", one black point, iris... this one turns bigger diameter.
Sometimes, hear some musics, you are transported to places, dreaming when not really sleeping...
All the noises...the life noises disappear...you go deep into the music...as hipnotized.
If the music remember someone you lose... some loved person, you will cry.
The sound, the music playing never turn you tired.
The music do not pass you the idea that is loud.
Have you, PMA and friends, felt deep passion.... to loose the voice when meeting the lady?.... had your legs shaking?.... if positive, you can understand...if never felt that!.... do not try to understand...not possible, without reference, cannot understand!
regards,
Carlos
Wonderfull Eva!
That's the problem.... i think engineers must turn the first street left, and the second rigth... stop in one human Sciences University and study a little some "human Perception"...or some books can help too.
PMA asked...how can you explain audibility.
You can measure also:
When sounds good... your heart beats increase
Your arterial pression increase, despite your muscles relax.
Your eyes have one "apperture", one black point, iris... this one turns bigger diameter.
Sometimes, hear some musics, you are transported to places, dreaming when not really sleeping...
All the noises...the life noises disappear...you go deep into the music...as hipnotized.
If the music remember someone you lose... some loved person, you will cry.
The sound, the music playing never turn you tired.
The music do not pass you the idea that is loud.
Have you, PMA and friends, felt deep passion.... to loose the voice when meeting the lady?.... had your legs shaking?.... if positive, you can understand...if never felt that!.... do not try to understand...not possible, without reference, cannot understand!
regards,
Carlos
Pavel,
I have been studying your circuit. It is extremely clever, and I'm profoundly impressed with it. The distortion specification is astonishing, and the absence of H4 and beyond quite a revelation. Thank you for publishing it; it is a significant contribution to the art.
It would be nice to drive it with a tube, but almost nothing has the low Zout required. Driving emitters is tough going......
Have you done this with bipolar devices? Unless you move to very high Iq it is difficult to eliminate self-oscillation in mosfets. Any thoughts on this?
Your comments on classical music lovers are interesting. My own dear Mother is a classical music devotee to the exclusion of everything else, in fact. She can only appreciate music written by a composer dead at least 150 years. I reacted against this attitude as a young man and while I enjoy a lot of classical music (particularly Antonin Dvorak and Babrich Smetana, incidentally), my present love is blues, trance and contemporary acoustic. I even get a kick out of Greg Brown!! (There's one for you to work out, 'Intellectual Country'!!) There is excellence in every genre, and yes, the thrill of a live orchestral performance is quite something, but generally I'm told the concert hall acoustics and the orchestral quality here in Australia does not quite match Europe. However, I concede your point that the working standard should remain the live performance - this is realism after all - but there are many recordings which to me sound better than a live performance because I'm not sitting next to some wheezy, coughing octogenarian at one side of the hall....... (Don't worry, I'm not far off 60 myself.....!)
Cheers,
Hugh
I have been studying your circuit. It is extremely clever, and I'm profoundly impressed with it. The distortion specification is astonishing, and the absence of H4 and beyond quite a revelation. Thank you for publishing it; it is a significant contribution to the art.
It would be nice to drive it with a tube, but almost nothing has the low Zout required. Driving emitters is tough going......
Have you done this with bipolar devices? Unless you move to very high Iq it is difficult to eliminate self-oscillation in mosfets. Any thoughts on this?
Your comments on classical music lovers are interesting. My own dear Mother is a classical music devotee to the exclusion of everything else, in fact. She can only appreciate music written by a composer dead at least 150 years. I reacted against this attitude as a young man and while I enjoy a lot of classical music (particularly Antonin Dvorak and Babrich Smetana, incidentally), my present love is blues, trance and contemporary acoustic. I even get a kick out of Greg Brown!! (There's one for you to work out, 'Intellectual Country'!!) There is excellence in every genre, and yes, the thrill of a live orchestral performance is quite something, but generally I'm told the concert hall acoustics and the orchestral quality here in Australia does not quite match Europe. However, I concede your point that the working standard should remain the live performance - this is realism after all - but there are many recordings which to me sound better than a live performance because I'm not sitting next to some wheezy, coughing octogenarian at one side of the hall....... (Don't worry, I'm not far off 60 myself.....!)
Cheers,
Hugh
PMA said:
Hi PMA !
At which outputlevel/load are these ?
And it would be much more interesting to see the levels at 20khz,
as EC is said to be most valuable at high freqs.
Are these measured or from sim ?
The numbers sound good, with tripledarlington (12v into 4ohm,
1-20khz, V+/- = 24v) i get only down to 0.08% and 0.04% (2nd,3rd),
4th = 0.02%, bjts = bc546b/556b , mje15030/31, mjl3281a/1302a.
Values are sim'd, outputstage only.
Mike
Hugh,
Thank you for your post.
I have not tried the circuit for bipolar devices, but I am sure that it is possible to use it for double EF - after several minor modifications.
Could you please get into more details about mosfets "self-oscillations"? Can I measure it by 150MHz scope (as I usually do) or RF measuring antenna?
Regarding music, you will hardly believe it but I do listen to all kinds of music
. And I also do like classical music very much.
Thank you for your post.
I have not tried the circuit for bipolar devices, but I am sure that it is possible to use it for double EF - after several minor modifications.
Could you please get into more details about mosfets "self-oscillations"? Can I measure it by 150MHz scope (as I usually do) or RF measuring antenna?
Regarding music, you will hardly believe it but I do listen to all kinds of music
. And I also do like classical music very much.
MikeB said:
Both simulated and measured. Unfortunately I only have images from measurements of the older version of the circuit. Anyway I am attaching one of them (for 8 Ohm load). The distortion did not depend much on amplitude. I plan to measure final tuned version of the circuit.
Attachments
Pavel,
I see oscillations on both a 25MHz and a 100MHz CRO on a sliding bias PP bipolar/mosfet CFP of my own design (commercial). It appears as a fuzzy outline on the entire audio waveform. No such problems exist with Class A SE follower circuits using mosfets, and over about 200mA the problem seems to disappear. I draw the conclusion it is a tendency of mosfets to oscillate at lower currents; they can be tamed to some degree with 47pF caps across gate/drain and largish gate resistors to tame this charge migration across the gate dielectric.
Does that answer your question, or do you want more information? I'm sure you know more about this than I do.....
Cheers,
Hugh
This is a clever CCT. I begin to understand what it does. But I have a question. What is the 4 diodes do? I think the CCT should work without those diodes.
Also in the NP's explenation, the origin of this cct is not for EC in the first place. It is more for "clever bias" cct. Is that 4 diodes to be put in heatsink to get temperature compensation?
Also in the NP's explenation, the origin of this cct is not for EC in the first place. It is more for "clever bias" cct. Is that 4 diodes to be put in heatsink to get temperature compensation?
Hugh,
thank you very much for the explanation. I have not noticed this problem at the circuit discussed in this thread. I assume you have noticed base stopper resistors R12 and R13, they are probably doing their job.
On the other hand I encountered similar trouble that you have described when I was developing the class A BJT push-pull CFP output stage. Especially transition from class A into AB had been sensitive to oscillations, b-c capacitors at the drivers and output Zobel cured the problem. The oscillations were local, unmeasurable at the output, but just power BJT's.
Regards,
Pavel
thank you very much for the explanation. I have not noticed this problem at the circuit discussed in this thread. I assume you have noticed base stopper resistors R12 and R13, they are probably doing their job.
On the other hand I encountered similar trouble that you have described when I was developing the class A BJT push-pull CFP output stage. Especially transition from class A into AB had been sensitive to oscillations, b-c capacitors at the drivers and output Zobel cured the problem. The oscillations were local, unmeasurable at the output, but just power BJT's.
Regards,
Pavel
other mosfets
Great design Pavel,
I want to use IRF510 and IRF9510 in the output stage, what (resistors) do I have to modify in the schematic to use these mosfets?
I have virtually no experience with Mosfet amplifiers.
So I do not know how or what I have to optimize in the schematic to use these mosfets.
Thank You in advance:
Nexus
Great design Pavel,
I want to use IRF510 and IRF9510 in the output stage, what (resistors) do I have to modify in the schematic to use these mosfets?
I have virtually no experience with Mosfet amplifiers.
So I do not know how or what I have to optimize in the schematic to use these mosfets.
Thank You in advance:
Nexus
