Re: deep thoughts
...correct...
No...Sawzall said:
No...Sawzall said:
You are.....Sawzall said:
Sawzall said:
...correct...
then explain why...
If additional gain, in and of itself, does not enhance slew rate....
And gain has no relationship with rate of change...
AND slew rate is a measure of speed of change (or the ability of an amp to change voltage specifically, as noted by its units of measure V/sec).
How would having gain in the output stage enhance slew rate - in an amp that was not starved for gain in earlier gain stages, or having an output stage unable to follow the earler stages.?
If additional gain, in and of itself, does not enhance slew rate....
And gain has no relationship with rate of change...
AND slew rate is a measure of speed of change (or the ability of an amp to change voltage specifically, as noted by its units of measure V/sec).
How would having gain in the output stage enhance slew rate - in an amp that was not starved for gain in earlier gain stages, or having an output stage unable to follow the earler stages.?
Re: then explain why...
http://diyaudio.com/forums/showthread.php?postid=168561#post168561
Sawzall said:
http://diyaudio.com/forums/showthread.php?postid=168561#post168561
Sawzall, suppose you have a fast output stage, faster than the driver.
If the driver has 25 V/us which slew rate would the output of the output stage have if the gain was 4?
Correct, 100 v/us.
That's what the whole thread is all about but Mike is proving something else which I don't know what it is.
Now, is an amplifying output stage good? That's an another story.
If the driver has 25 V/us which slew rate would the output of the output stage have if the gain was 4?
Correct, 100 v/us.
That's what the whole thread is all about but Mike is proving something else which I don't know what it is.
Now, is an amplifying output stage good? That's an another story.
Thanks peranders...peranders said:
peranders said:
I am not sure i understand...but i think you are referring to the fact that two-pole compensation got mixed up in this along the way?
peranders said:
http://diyaudio.com/forums/showthread.php?postid=168411#post168411
No, you don't, ..... That's the problem.mikek said:
Ok, Mike we know that now. But I and many with me don't agree. Just as simple as that.
oK peranders.....what do i appear not to understand?peranders said:
peranders said:
...and...what is it exactly that you don't agree with....?
Hi Mike
I think the main reason that output stages with gain are not everyone's favourite is because they are sometimes a PITA to get stable.
A second reason may be the fact that the humble emitter follower is one of the most linear applications of a transistor one could imagine, even though some output stages with gain would allow additional local feedback.
Just one thing to remember:
It is very seldom THE piece of equipment that excels in exactly one discipline that becomes a classic !!! So achieving high slew-rate ALONE is worth nothing !
I am however always open for new things and so I did once perform simulations with a circuit that has an output stage with gain.
Simulations looked very promising but I wasn't able to try it out in real hardware so far.
Even though the theory behind technical stuff is most often the interesting part for an engineer, it is the real-world behaviour that counts in the end.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=7058
Regards
Charles
I think the main reason that output stages with gain are not everyone's favourite is because they are sometimes a PITA to get stable.
A second reason may be the fact that the humble emitter follower is one of the most linear applications of a transistor one could imagine, even though some output stages with gain would allow additional local feedback.
Just one thing to remember:
It is very seldom THE piece of equipment that excels in exactly one discipline that becomes a classic !!! So achieving high slew-rate ALONE is worth nothing !
I am however always open for new things and so I did once perform simulations with a circuit that has an output stage with gain.
Simulations looked very promising but I wasn't able to try it out in real hardware so far.
Even though the theory behind technical stuff is most often the interesting part for an engineer, it is the real-world behaviour that counts in the end.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=7058
Regards
Charles
Endless loops do me no good.
Mikek, the endless loops of referring back to a post that does not explain does not help me - in the next five posts forward from your link basically just points me off to a bunch of URL's. Lots of heat, no enlightenment there.
Thank you for your reply Peranders.
MikeK, to look at it another way, if you had the driver stage that had the 100 v/us, and the output stage was our humble emitter follower with its less than 1 gain. Would the amp not still have roughly the same slew rate without the stabilty issues that outputs with gain can have?
Mikek, the endless loops of referring back to a post that does not explain does not help me - in the next five posts forward from your link basically just points me off to a bunch of URL's. Lots of heat, no enlightenment there.
Thank you for your reply Peranders.
MikeK, to look at it another way, if you had the driver stage that had the 100 v/us, and the output stage was our humble emitter follower with its less than 1 gain. Would the amp not still have roughly the same slew rate without the stabilty issues that outputs with gain can have?
phase_accurate said:
hi Charles,
thanks for well considered points. I think you'll find we are in agreement on one point at least....i don't think CE output stages are very stable either...which is why i suggested generating the gain slightly before the output stage-proper.....here:
http://diyaudio.com/forums/showthread.php?postid=168411#post168411
I mean linear in terms of input to output voltage transfer (beeing unity with large accuracy). If this wasn't the case all those amps with emitter follower output stages, that take feedback before the output, would have quite high THD.
Regards
Charles
Re: Endless loops do me no good.
What is the hardest thing to design: A fast driver or a amplifying output stage?
I'm in again. Couldn't help it.
Yes, accordning to me.Sawzall said:
What is the hardest thing to design: A fast driver or a amplifying output stage?
I'm in again. Couldn't help it.
phase_accurate said:
An emitter follower is a gain stage with 100% local feedback. Feedback, as is known, makes a stages more linear. Naturally the EF stage is very linear.
Collorary: An EF output amp with zero feedback is an oxymoron.
Jan Didden
(Not sure about the meaning of oxymoron, but I like that word).
O.K. O.K. O.K.......
I meant take-off of the overall feedback loop. Can we agree on this ?
Regards
Charles
I meant take-off of the overall feedback loop. Can we agree on this ?
Regards
Charles
Lots of things have a "voltage transfer" close to unity into a loudspeaker load. Like a coil of wire or a relay or a forward-biased diode. Would you want to listen to your favourite musice through a forward biased diode?
I feel compelled to douse the slew rate debacle. For most loudness of music with typically efficient speakers, not much more than 5V/us is needed. We are debating a red herring.
I feel compelled to douse the slew rate debacle. For most loudness of music with typically efficient speakers, not much more than 5V/us is needed. We are debating a red herring.
Jan,
How do you define "feedback"? Is a resistor a 100% feedback device...like a transistor you change the voltage across its terminals and a proportional current flows. Usually extremely linear too.
How do you define "feedback"? Is a resistor a 100% feedback device...like a transistor you change the voltage across its terminals and a proportional current flows. Usually extremely linear too.
- Status
- Not open for further replies.