Sorry Folks, I guess I should make things clearer. In the old days, 35 years ago or so, we used devices with something like 4MHz F(t). Sometimes, people used even slower devices like the 2N3055, or the 2N3773, or even the MJ15003, that had an even lower F(t). With a 4MHz F(t) and a first class driver stage that looks even more up-to-date than the schematic in the Motorola app note mentioned previously, I could get 100V/us with Motorola 4MHz devices. With 2MHz Motorola devices, I could get only 50V/us and twice the distortion at all levels.
In 1980, the Japanese released new devices that had over 20MHz F(t). Then I could achieve about 500V/us with the same circuits. Now I am not claiming that 500V/us is absolutely necessary, but we instead were able to remove the 2UH output coil and do other things that the lower F(t) devices did not allow us. To alternately use these very different output devices in the same exact design, is surprising to say the least.
In 1980, the Japanese released new devices that had over 20MHz F(t). Then I could achieve about 500V/us with the same circuits. Now I am not claiming that 500V/us is absolutely necessary, but we instead were able to remove the 2UH output coil and do other things that the lower F(t) devices did not allow us. To alternately use these very different output devices in the same exact design, is surprising to say the least.
Re: Design is like cooking...
That's the difference between actual cooks and mere combiners. And a true chef is someone who can do it consistently, and do it either way. Nice analogy- my cooking partner and I are constantly saying that the most important step to great cuisine is the shopping part.
dhaen said:
That's the difference between actual cooks and mere combiners. And a true chef is someone who can do it consistently, and do it either way. Nice analogy- my cooking partner and I are constantly saying that the most important step to great cuisine is the shopping part.
I cannot see something subjective here ??? I'm agree with everyone!
We live in an imperfect world, and audio world is even worse. Even a $30,000 system is imperfect. So, how about a $300 system? With a lot of circuits and trade-offs in electronics, magic caps, magic resistors and even magicians
D) do exist.
Okay, (schematic) design is important, but what assumptions have you made when you design the circuit? Components behaviour? Components quality? Have you designed the amp to work with un-ideal input signal? How about the speaker? IMO, capacitors had been valued mostly based on listening, which means that the speaker also determine the result.
Somebody mentioned that tubes of different brands sound different. Forget the difference in the measurement of those tubes but look at this point: simplicity. Tube circuits are simple, a slight difference in a the tube will greatly affect the performance. But try to use a "perfect" amp design using more than 8 stages (and feedback everywhere) and change the capacitors.
But noone will disagree against the power of simple circuits (short audio path). Complex power supply tend to limit the sonic performance, but simple ones tend to bring more noise or ripple (Do you design the power supply when you design your amp?). In many cases I prefer to use simple power supply (and also short audio circuit) with MAGIC CAPS (Of course I have a very complex switching power supply for my DAC).
Design your amp with small budget (less components, component's quality and environtment independent, or whatever) and big performance or big sales, then that is GOOD design and GOOD designer. Sorry designers, I think dhaen, SY and MCP have the potential to be a very good one.
We live in an imperfect world, and audio world is even worse. Even a $30,000 system is imperfect. So, how about a $300 system? With a lot of circuits and trade-offs in electronics, magic caps, magic resistors and even magicians
D) do exist.Okay, (schematic) design is important, but what assumptions have you made when you design the circuit? Components behaviour? Components quality? Have you designed the amp to work with un-ideal input signal? How about the speaker? IMO, capacitors had been valued mostly based on listening, which means that the speaker also determine the result.
Somebody mentioned that tubes of different brands sound different. Forget the difference in the measurement of those tubes but look at this point: simplicity. Tube circuits are simple, a slight difference in a the tube will greatly affect the performance. But try to use a "perfect" amp design using more than 8 stages (and feedback everywhere) and change the capacitors.
But noone will disagree against the power of simple circuits (short audio path). Complex power supply tend to limit the sonic performance, but simple ones tend to bring more noise or ripple (Do you design the power supply when you design your amp?). In many cases I prefer to use simple power supply (and also short audio circuit) with MAGIC CAPS (Of course I have a very complex switching power supply for my DAC).
Design your amp with small budget (less components, component's quality and environtment independent, or whatever) and big performance or big sales, then that is GOOD design and GOOD designer. Sorry designers, I think dhaen, SY and MCP have the potential to be a very good one.
john curl
Thats right! In the schematic on my website I refer to MJL2194 or MJL1302A with complement devices. These are just two of the versions I have running. I also have one with MJL4281A/MJL4302A, one with MJ15022/MJ15024 and the first one build using 2SA1943/2SC5200. And I'm not recommending any of them on my site......
If I should note every component tested you wouldn't be able to see the schematic
However as diy's haves the tendensy to use the components they have "on stock", I do not care so much about which components are listed in the schematic until the final judgement have been done
Thats right! In the schematic on my website I refer to MJL2194 or MJL1302A with complement devices. These are just two of the versions I have running. I also have one with MJL4281A/MJL4302A, one with MJ15022/MJ15024 and the first one build using 2SA1943/2SC5200
. And I'm not recommending any of them on my site......If I should note every component tested you wouldn't be able to see the schematic
However as diy's haves the tendensy to use the components they have "on stock", I do not care so much about which components are listed in the schematic until the final judgement have been done
As this design is so buffered and 'detuned' to accept virtually any part, then I would equate it to adding a fancy air cleaner to a Volkswagon Beetle. You would not get much difference in performance between different devices, with your Beetle. However, if you tried the same thing with a Porsche or Ferrari, I bet some difference would be noted in most cases, especially if you tried to use a cheap knockoff.
Of course, the same can be said for hi fidelity equipment. You can note larger differences on average, with better, more finely tuned components. Cheap caps, transistors (or out of spec) will make a bigger difference with more exotic designs.
For the record, ACD's amp is loosely based on a Marantz 16 power amp, developed about 1968. I used to sell them, and they worked OK then, as they were rather 'polite' sounding.
Of course, the same can be said for hi fidelity equipment. You can note larger differences on average, with better, more finely tuned components. Cheap caps, transistors (or out of spec) will make a bigger difference with more exotic designs.
For the record, ACD's amp is loosely based on a Marantz 16 power amp, developed about 1968. I used to sell them, and they worked OK then, as they were rather 'polite' sounding.
john curl
Sorry to inform you, but if you had read the text (the small black symbols that looks like ants ) you will find, that I have based this circuit on old reference amps like Phase Linear etc. And you would also have notised that THIS particular design never was entended to be a 2003 "New Standard" amplifier, but as a tribute to the lost amps from back in time...
Sometimes it's harder to recreate something, than it is to create a new......... and I'm doing this for fun, and not for setting a new standard!
And by the way.... The Volkswagon Beetle was revolutionary, when it first hit the streets
Sorry to inform you, but if you had read the text (the small black symbols that looks like ants
) you will find, that I have based this circuit on old reference amps like Phase Linear etc. And you would also have notised that THIS particular design never was entended to be a 2003 "New Standard" amplifier, but as a tribute to the lost amps from back in time...Sometimes it's harder to recreate something, than it is to create a new.........
and I'm doing this for fun, and not for setting a new standard!And by the way.... The Volkswagon Beetle was revolutionary, when it first hit the streets
Getting back to the original topic, ACD basically said that in his opinion, exotic semiconductors and 'super' caps were not worthwhile. He is correct, when it comes to his design.
However, other designs both tube and solid state, can use better parts to some advantage.
Again, I want to make a comparison:
Let's say that you have 2 cars, both running OK. One is a Volks Beetle, and the other is a Porsche 944 (early). Now why don't we swap the tires between the two vehicles? I am presuming that both wheels are 15 inch rims and of comparable radius. Heck we can swap the tires with their respective rims.
What would happen? Would the Volks be much better off? How about the Porsche 944? Could you tell the difference if you just went to the store and back, or would it take a mountain road to give you any insight? For the record, I made this comparison in my earlier Porsche 924, so I know the answer from experience.
However, other designs both tube and solid state, can use better parts to some advantage.
Again, I want to make a comparison:
Let's say that you have 2 cars, both running OK. One is a Volks Beetle, and the other is a Porsche 944 (early). Now why don't we swap the tires between the two vehicles? I am presuming that both wheels are 15 inch rims and of comparable radius. Heck we can swap the tires with their respective rims.
What would happen? Would the Volks be much better off? How about the Porsche 944? Could you tell the difference if you just went to the store and back, or would it take a mountain road to give you any insight? For the record, I made this comparison in my earlier Porsche 924, so I know the answer from experience.
The problem with discussing a multitude of tweaks is that someone, somewhere, will just freeze up, and lose confidence in the argument. Of course, some of you don't accept much, others a little, and still others remain open to whatever works for them, or in general.
For example, I spoke for about an hour with Geoff Kait, the guy who sells rocks in a jar, to improve some audio systems. Geoff happens to be a physicist who could teach the vast majority of you, a thing or two about audio, but he doesn't want to bother. I can't blame him.
For example, I spoke for about an hour with Geoff Kait, the guy who sells rocks in a jar, to improve some audio systems. Geoff happens to be a physicist who could teach the vast majority of you, a thing or two about audio, but he doesn't want to bother. I can't blame him.
john curl said:
My recommendation is - stay natural. Make a Ferrari and then change it's response to (1 - exp(-t/Tau)), this one will be the same for small and large signals not affected by the SR, but 5 - 10 times well above audio band. Then the change of a screw holding a battery will not matter that much.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.