Scott, you can read my mindscott wurcer said:
. Have the same question to Jan. R's between pin2's and pin5's as well (and hence no GND currents). And if we allow for some feedback this dual could be used as fully diff. gain block, too, SuSy-style. Or as a nice ZF xfomer input and such.EDIT: oops, I see pin5 differential would be sketchy, we'd need something to have a stable DC op. point.
- Klaus
The pin3's can be used as differential inputs or single-ended of either phase. The common mode is undefined if you put the gain resistor between pin5's only. There is nothing to stop you from putting a resistor to ground on each and a gain adjust between them. You could also use common-mode feedback to the output. I think the JC-80 had both input and output servos for offset and common-mode ?
janneman said:
In this one current through an output device is almost constant:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=147916
and the previous version had a source-drain voltage almost constant as well. As the result, temperature, transconductance, capacitances' variations were minimal. It was the best amp I ever heard, though I paid for the quality by electricity (1.5 KW dissipation).
EUVL said:
Interesting application. I would discard C103, C105 & C108, since there is a servo anyway. Maybe connect the servo more upfront. Just thinking out loud here.
I guess you mean what gain in this mode, ol? The gain isn't the determining factor for sound quality, it's the input and output signal levels. You should keep Vin below a volt if possible, and Vout below 3 or 4 volts. Above that, it still sounds good
but measured distortion goes up.I also like to run them close to max supply (+/-16.5) to maximize undistorted levels.
jd
janneman said:
Hi Jan,
Interestingly, with a class A amp the output stage power dissipation actually goes down as output power increases. So temperatures still tend to swing with the program material, but somewhat in the opposite direction to that of a class AB amp (which tends to increase with output power until about one-third max power, then derease somewhat for higher power levels).
If one is willing to deal with a bit of gm doubling, one can over-bias a class AB amplifier to the point where the output stage power dissipation changes less as a function of output power.
Cheers,
Bob
Hi stinius,
I was only pointing out some observations to further support Bob's statement. I am well aware that you will never get more energy out of a system than you put into it. Perpetual motion anyone?
Hi John,
An equation is only one way to describe the situation. There are other perfectly valid ways to say or show the same thing. Personally, I know the equations I work with often enough. For the rest, I look what I need up. I find there are fewer errors made that way. Besides, math is sort of a shorthand way to express ideas. A dangerous way if you don't carefully consider absolutely everything when attaching numbers and expressions for your unknowns. The sad thing about talking about things in a math context is that you frequently loss the idea in the numbers. Another consideration is that some times you only need a trend or ballpark for an answer. Equations tend to focus some people too closely and they can't see the forest for the trees.
-Chris
I was only pointing out some observations to further support Bob's statement. I am well aware that you will never get more energy out of a system than you put into it. Perpetual motion anyone?
Hi John,
I found out about some of this stuff 44 years ago as well. What I can say with some certainty is that your knowledge of this is more extensive these days than it was when you first learned some of this stuff. Do the math, I was only 6 years old then.
Yes there are.
An equation is only one way to describe the situation. There are other perfectly valid ways to say or show the same thing. Personally, I know the equations I work with often enough. For the rest, I look what I need up. I find there are fewer errors made that way. Besides, math is sort of a shorthand way to express ideas. A dangerous way if you don't carefully consider absolutely everything when attaching numbers and expressions for your unknowns. The sad thing about talking about things in a math context is that you frequently loss the idea in the numbers. Another consideration is that some times you only need a trend or ballpark for an answer. Equations tend to focus some people too closely and they can't see the forest for the trees.
-Chris
Easy is better than complex. Chris, you are making it complex. This happens when techs try to do what engineers are trained to do. It is easier, in this case, to use the engineering equations.
Someone: Dimitri, Cordell, Wurcer, please bail out this situation. It is on a page of a favorite book owned by most of you.
Someone: Dimitri, Cordell, Wurcer, please bail out this situation. It is on a page of a favorite book owned by most of you.
Hi John,
I am not making anything complex. Just responding to your comment about equations where none are really required. I thought mentioning some of the negatives associated with using an equations was valuable to anyone contemplating that route.
Errors of omission in some instances. In other words, when can we use a simplified equation and when will that lead us astray? Any time someone decides to plug numbers into an equation, they stand the risk of ignoring effects that may be important. Since the numbers worked out (perhaps 10 times by three people), there is a tendency to accept the result that pops out without questioning whether that answer is valid or not. We humans, being lazy, will tend towards an easy answer without considering the situation any further.
Understanding the subject is far more valuable than plugging numbers into an impressive looking equations.
Techs are the people who recognize and correct the errors that engineers make. Technicians take a concept to working reality.
Anyway, that tone you have was a bit insulting, and quite uncalled for.
-Chris
I am not making anything complex. Just responding to your comment about equations where none are really required. I thought mentioning some of the negatives associated with using an equations was valuable to anyone contemplating that route.
Errors of omission in some instances. In other words, when can we use a simplified equation and when will that lead us astray? Any time someone decides to plug numbers into an equation, they stand the risk of ignoring effects that may be important. Since the numbers worked out (perhaps 10 times by three people), there is a tendency to accept the result that pops out without questioning whether that answer is valid or not. We humans, being lazy, will tend towards an easy answer without considering the situation any further.
Understanding the subject is far more valuable than plugging numbers into an impressive looking equations.
Oh John, this is far too easy to ignore.
Techs are the people who recognize and correct the errors that engineers make.
Technicians take a concept to working reality.Anyway, that tone you have was a bit insulting, and quite uncalled for.
-Chris
Chris, if you are not an engineer or a physicist, you are not in a position to judge. If you are, please don't make MY life more difficult by parsing with me. My name is still on the head of this thread, and it has gotten over 1 million hits, partially because of this. Please don't drive me away. It is physically painful, at this time, for me to contribute, here. Let us go forward, rather than in circles.
- Status
- Not open for further replies.