What is the list in order of Efficiency?
I'd be interested to know what the most efficient Pass Power amp is... I've only got a limited current capacity in my tiny apartment, and running a 5.1 surround system with all Aleph-X amps might be too much for it to handle.
Can anyone shed any light on which are Nelson's most efficient power amps?
-Erik.
I'd be interested to know what the most efficient Pass Power amp is... I've only got a limited current capacity in my tiny apartment, and running a 5.1 surround system with all Aleph-X amps might be too much for it to handle.
Can anyone shed any light on which are Nelson's most efficient power amps?
-Erik.
Hot Hot Hot
Yep, that's about what I expected. The Gainclone sounds like a viable alternative. But being the devoted Pass fanatic, I'm considering tackling a design based upon those being discussed in the Monolithic supersymmetry thread. There was one, here (second one down the page):
http://www.diyaudio.com/forums/showthread.php?s=&threadid=11205&perpage=15&pagenumber=4
or at the bottom of the page here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=11205&perpage=15&pagenumber=3
That might just work for my purposes.
-Erik.
moe29 said:check out the CHIP AMP forum for what you're looking for.
5 Aleph amps in a small apartment would generate WAY too
much heat
Yep, that's about what I expected. The Gainclone sounds like a viable alternative. But being the devoted Pass fanatic, I'm considering tackling a design based upon those being discussed in the Monolithic supersymmetry thread. There was one, here (second one down the page):
http://www.diyaudio.com/forums/showthread.php?s=&threadid=11205&perpage=15&pagenumber=4
or at the bottom of the page here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=11205&perpage=15&pagenumber=3
That might just work for my purposes.
-Erik.
moe29 said:check out the CHIP AMP forum for what you're looking for.
5 Aleph amps in a small apartment would generate WAY too
much heat
An Aleph3 (200 watt I believe) might still be acceptable (especially in winter).
I always wondered, isn't there some way you could use the excessive heat to produce electricity (or do something usefull, apart from frying eggs)??
Yeah, good idea... Who was that guy?
You read my mind, Nelson. Or, I suppose I read yours in this case. I've been poring over the push pull schematic in your supersymmetry patent, and wondering if there would be any good reason not to tackle a project with the extra parts to see what the differences are between it and the simpler designs?
Now, if I understand correctly, you have stated that single ended -biased by resistors- class A is the least efficient, followed by single ended class A biased with constant current sources. More efficient than that would be a push pull design or "complementary symmetry" design as I believe it's called. Would there be any further benefit to bias a complementary symmetry pair with constant current sources? I'm sort of under the impression that more is better, but I'd like to be corrected on that if there are diminishing returns at some point?
Hey, how about an "X'ed" complementary symmetry amp with Aleph constant current sources, folded cascodes, and some kind of reverse peltier pads on all the heatsinks to recycle the excess heat back into current for the CCS's -an idea I like to call "superselfbiasing"
-Erik.
Nelson Pass said:You guys are pretty good at reverse, or even anticipatory reverse
engineering. The Class AB X amp is presented in full on the
patent, with only the values missing.
You read my mind, Nelson. Or, I suppose I read yours in this case. I've been poring over the push pull schematic in your supersymmetry patent, and wondering if there would be any good reason not to tackle a project with the extra parts to see what the differences are between it and the simpler designs?
Now, if I understand correctly, you have stated that single ended -biased by resistors- class A is the least efficient, followed by single ended class A biased with constant current sources. More efficient than that would be a push pull design or "complementary symmetry" design as I believe it's called. Would there be any further benefit to bias a complementary symmetry pair with constant current sources? I'm sort of under the impression that more is better, but I'd like to be corrected on that if there are diminishing returns at some point?
Hey, how about an "X'ed" complementary symmetry amp with Aleph constant current sources, folded cascodes, and some kind of reverse peltier pads on all the heatsinks to recycle the excess heat back into current for the CCS's -an idea I like to call "superselfbiasing"
-Erik.
Re: Yeah, good idea... Who was that guy?
" Would there be any further benefit to bias a complementary symmetry pair with constant current sources? "
This would be a class AB amp
I'll let the zen master Pass answear the other questions
e.lectronick said:
You read my mind, Nelson. Or, I suppose I read yours in this case. I've been poring over the push pull schematic in your supersymmetry patent, and wondering if there would be any good reason not to tackle a project with the extra parts to see what the differences are between it and the simpler designs?
Now, if I understand correctly, you have stated that single ended -biased by resistors- class A is the least efficient, followed by single ended class A biased with constant current sources. More efficient than that would be a push pull design or "complementary symmetry" design as I believe it's called. Would there be any further benefit to bias a complementary symmetry pair with constant current sources? I'm sort of under the impression that more is better, but I'd like to be corrected on that if there are diminishing returns at some point?
Hey, how about an "X'ed" complementary symmetry amp with Aleph constant current sources, folded cascodes, and some kind of reverse peltier pads on all the heatsinks to recycle the excess heat back into current for the CCS's -an idea I like to call "superselfbiasing"
-Erik.
" Would there be any further benefit to bias a complementary symmetry pair with constant current sources? "
This would be a class AB amp
I'll let the zen master Pass answear the other questions
Re: Re: Yeah, good idea... Who was that guy?
Hmmm...
I was under the impression that it wasn't necessarily the manner in which an amp is biased that seperates class A from AB, but rather how much each side is biased. For instance, if both the top and bottom active devices are given only enough gate voltage to turn on for one half of the waveform, it would be class AB, but if they were biased with enough voltage to be on all the time they would be class A, right? I guess you've got me a little confused. I may be the one who is confused however. What I meant by 'biasing' (and I'm beginning to think that's the wrong word for it) with a ccs is to put one between the V+ and the upper active device, and one between the V- and the lower active device. Would that in particular make any difference in the efficiency of the amplifier?
Please remember, I'm a noob, and my understanding of electronics is still pretty basic. (eg lot's of books, much less practical application of the knowledge therein)
-Erik.
Bricolo said:
" Would there be any further benefit to bias a complementary symmetry pair with constant current sources? "
This would be a class AB amp
I'll let the zen master Pass answear the other questions
Hmmm...
I was under the impression that it wasn't necessarily the manner in which an amp is biased that seperates class A from AB, but rather how much each side is biased. For instance, if both the top and bottom active devices are given only enough gate voltage to turn on for one half of the waveform, it would be class AB, but if they were biased with enough voltage to be on all the time they would be class A, right? I guess you've got me a little confused. I may be the one who is confused however. What I meant by 'biasing' (and I'm beginning to think that's the wrong word for it) with a ccs is to put one between the V+ and the upper active device, and one between the V- and the lower active device. Would that in particular make any difference in the efficiency of the amplifier?
Please remember, I'm a noob, and my understanding of electronics is still pretty basic. (eg lot's of books, much less practical application of the knowledge therein)
-Erik.
the few things I know:
SE or PP with high enough bias for the whole waveform= class A (I still don't understand the benefict of PP for class A operation)
PP with bias (but not high enough to enable one of the devices to conduct during the whole waveform)= class AB (runs class A if the output power is low)
PP with no bias: class B
SE or PP with high enough bias for the whole waveform= class A (I still don't understand the benefict of PP for class A operation)
PP with bias (but not high enough to enable one of the devices to conduct during the whole waveform)= class AB (runs class A if the output power is low)
PP with no bias: class B
Makes sense
Ok, so we're on the same page on the subject of bias.
I think that for class A, the benefit of PP would be the ability to choose active devices with lower tolerances ~ half the power handling capacity of a device used alone in a circuit since the two transistors share the work load. This doesn't really explain to me how pp could be more efficient, though. Except that with two gaind devices you are doubling the amplification of a similar circuit with a single device while maintaining the fact that it's a single gain stage?
-Erik
Ok, so we're on the same page on the subject of bias.
I think that for class A, the benefit of PP would be the ability to choose active devices with lower tolerances ~ half the power handling capacity of a device used alone in a circuit since the two transistors share the work load. This doesn't really explain to me how pp could be more efficient, though. Except that with two gaind devices you are doubling the amplification of a similar circuit with a single device while maintaining the fact that it's a single gain stage?
-Erik
Aha! I found the passage where Nelson talks about efficiency!
I finally found the "Pass"age where I got the notion of efficiency being greater in PP
In Nelson's article DIY project, "The A40 Power Amplifier", under the section 'Beating the Heat' he writes:
"Class A amplifiers have different efficiency factors depending upon the design. The leaste efficient is the circuit if Fig. 4a, where the transistor is biased by a resistor and whose AC output power to the load is less than 20 percent of its idling dissipation. Fig. 4b shows the same configuration where a constant current source replaces the resistor, improving the linearity and efficiency of the circuit. The value of the constant current source must be equal to or greater than the maximum output current. For an 80W peak (40W, rms) into 8 ohms, therefore, the current must be at least 3.2 A, which practically speaking means a worst case dissipation of 200W per channel in the idling output stage.
Push-pull circuitry more or less doubles the efficiency of a class A output stage (Fig. 4c) because unlike the constant current sourced design, its idle current need be only one half the peak output current, or 1.6A in the example, for an idling dissipation of about 100W for a 40W amplifier."
So, given these points, I still ask the question whether there would be any benefit to use constant current sources to supply (bias) a class A Push-Pull stage?
Nelson? Any insights?
-Erik.
Bricolo said:
I finally found the "Pass"age where I got the notion of efficiency being greater in PP
In Nelson's article DIY project, "The A40 Power Amplifier", under the section 'Beating the Heat' he writes:
"Class A amplifiers have different efficiency factors depending upon the design. The leaste efficient is the circuit if Fig. 4a, where the transistor is biased by a resistor and whose AC output power to the load is less than 20 percent of its idling dissipation. Fig. 4b shows the same configuration where a constant current source replaces the resistor, improving the linearity and efficiency of the circuit. The value of the constant current source must be equal to or greater than the maximum output current. For an 80W peak (40W, rms) into 8 ohms, therefore, the current must be at least 3.2 A, which practically speaking means a worst case dissipation of 200W per channel in the idling output stage.
Push-pull circuitry more or less doubles the efficiency of a class A output stage (Fig. 4c) because unlike the constant current sourced design, its idle current need be only one half the peak output current, or 1.6A in the example, for an idling dissipation of about 100W for a 40W amplifier."
So, given these points, I still ask the question whether there would be any benefit to use constant current sources to supply (bias) a class A Push-Pull stage?
Nelson? Any insights?
-Erik.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.