John,
I watch this thread for enjoyment and laughs. Why do you guys have so much fun talking about stuff 30+ years old? Certainly an op amp with a 30 volt swing and a gain bandwidth product of 10 MHz will work fine at audio frequencies.
Also, why would anyone in their right mind pay $20,000 to $40,000 for an amplifier? Lots of snake oil.
Rick
I watch this thread for enjoyment and laughs. Why do you guys have so much fun talking about stuff 30+ years old? Certainly an op amp with a 30 volt swing and a gain bandwidth product of 10 MHz will work fine at audio frequencies.
Also, why would anyone in their right mind pay $20,000 to $40,000 for an amplifier? Lots of snake oil.
Rick
Well, I wouldn't pay $20,000-$40,000 for a world class preamp, but people do, all around the world. I just design them. AND there is no 'fat' in this design, just quality, and versatility.
Everyone, who can think outside their specific values and considerations, must understand that there is a market out there for Bugatti's and Bentleys. I can't afford them, but some people can, and I wish them well. SOMEONE had to design them, and how much fun, as an engineer, it must have been, to do so.
Sawreyrw, just get a preamp based on the design we just made, and live a happy life. It should cost you around $100, if it is done right.
Everyone, who can think outside their specific values and considerations, must understand that there is a market out there for Bugatti's and Bentleys. I can't afford them, but some people can, and I wish them well. SOMEONE had to design them, and how much fun, as an engineer, it must have been, to do so.
Sawreyrw, just get a preamp based on the design we just made, and live a happy life. It should cost you around $100, if it is done right.
Since John is under heavy moderation, my answer probably hits you first.
John is a very respected audio electronics designer who happens to share some of his knowledge here. His good memory and historic accuracy is valuable to some of us here. He puts things into perspective like no one else.
So your comment about "any opamp will do" just shows your lack of knowledge. Sorry to say that. There is a lot of snake oil in this business. Please don't connect John to to that.
Thanks 2 quad, yes there is a lot of 'snake oil' in this business, but not EVERYTHING is 'snake oil' any more than Asian medicine or holistic (sp?) medicine is completely bogus. In fact, I recommended to my boss, just today, to check out Chinese Medicine for his problem, since the hospital is not working out. Why not?
I just threw out the 10 Mhz comment just to raise some feedback. But this amp would have a bandwidth of 100 KHz at a gain of 100. Isn't that way more than adequate for an audio amp? In addition, the full power bandwidth (slew rate limit) is well above the audio frequency range.
One thing I'm not sure about is how much distortion is acceptable. I doubt anyone can detect .1% or even 1%. I know it's not that simple and depends on which harmonics, etc., but still, why get obsessed with it?
I know that John is well respected, but he is also an arrogant so and so, and that's why he is under moderation. Sorry about the negative comment, but that's how I feel.
One thing I'm not sure about is how much distortion is acceptable. I doubt anyone can detect .1% or even 1%. I know it's not that simple and depends on which harmonics, etc., but still, why get obsessed with it?
I know that John is well respected, but he is also an arrogant so and so, and that's why he is under moderation. Sorry about the negative comment, but that's how I feel.
Last edited:
This is getting a bit off topic (as usual), but you'll find some of your answers here: http://www.meridian-audio.com/ara/coding2.pdf
Good for you. Enjoyment and laughs are healthy.
I, for one, find some of 30+ years old knowledge that John Curl shares quite illuminating.
From engineering view point, yes. From audiophile view point, I haven't heard one yet.
How do you know it's nothing but snake oil?
A pull-up resistor loads the Opamp so a constant current sourse is better.
Build an opamp linestage and try yourself if there is any change in sound. I have tryed the OPA134 with pull-up resistor to the negative rail and to the positive rail. I have played that circuit to visitors and made a simple A-B test. They knew when i made a change but i did not say what i did so in strict terms the outcome of this test is purely anecdotal because the visitors knew that i made a change and the experimenter ( me ) was in the room.
My personal experience is that i will not be able to differentiate the change in a double blind test. It has being discussed here many times that many little flaws that are hard to differentiate in a scientific proove test may nevertheless culminate and that a circuit made with utmost attention to detail may cross the audibility border although all changes made in isolation are not audible enough to pass the double blind test.
Build an opamp linestage and try yourself if there is any change in sound. I have tryed the OPA134 with pull-up resistor to the negative rail and to the positive rail. I have played that circuit to visitors and made a simple A-B test. They knew when i made a change but i did not say what i did so in strict terms the outcome of this test is purely anecdotal because the visitors knew that i made a change and the experimenter ( me ) was in the room.
My personal experience is that i will not be able to differentiate the change in a double blind test. It has being discussed here many times that many little flaws that are hard to differentiate in a scientific proove test may nevertheless culminate and that a circuit made with utmost attention to detail may cross the audibility border although all changes made in isolation are not audible enough to pass the double blind test.
The beauty of offering a $40,000 preamplifier that only costs a few hundred dollars to build is that you only have to sell a few of them a year to make a very handsome living out of it. But how do you get people to buy it? First have a famous name based on a "reputation." Then get someone who publishes a well read magazine that circulates to those who have more money than real engineering knowledge to say it's the best in the world. It doesn't matter if it's no better than anyone else's as long as that is what an important review of it says about it. And it is after all just a matter of opinion, there is no right or wrong answer.
Why would people have paid a lot of extra money to buy an Olds 88 when a much cheaper Chevy Caprice was the identical car? Name brand prestige. About 20 odd years ago it came out that the olds "rocket" engines were made in a Chevy plant and were in fact the same engines. Owners of Oldsmobiles were furious but there wasn't a mechanic in the country who couldn't have told you that. In fact if you owned an Olds, you'd usually get the same replacement parts as Chevys and Pontiacs with a different catalog number stamped on them and a higher pricetag. This happens all the time. There's an old saying that has it that no one ever went broke underestimating the intelligence of the American public.
Please enumerate the parts that contribute to the BOM. At $40,000 typical practice would place the actual parts cost at $8000 or so.
Things don't add up.
I opt for the LME49990 as the input chip. Only one. If we make the feedback resistors 600 Ohm and 20 Ohm the noise contribution of the 20 Ohm is acceptable. Say, if we move up from an affordable MM to a Denon DL103 that has 40 Ohm the increase the noise of the 20 Ohm resistor makes is ok i think. The noise the LME49990 makes is something less then a 50 Ohm resistor anyway. It has a nice low 1/F noise though and even the 1.4nV/qHz my INA produces does not bother me much because it sounds very balanced with no shot noise i could detect. Making the feedback resistor as high as 40 Ohms whould reduce the common mode distortion to zero at the price of some noise added. The 600 Ohm can then go up to 1.2kOhm loading the chip less.
It's a pity schematics aren't available for the better opamps or some other trickery might be possible. E.g. looking at the NE5534 schematic, it seems like a suitable resistor (47R?) connected between pin 5 (compensation) and the output would bias it very nicely into class A.
I suspect something similar might be possible with other opamps having external compensation, but without a schematic it's guessing or trial and error.
Attachments
Which are the expensive parts and about how much do they cost retail?
Of the 20,000 or so parts in an automobile which can retail for anywhere from umteen thousand dollars on up to anything but most cars under $30,000, the part which winds up being the most expensive is the nut behind the wheel.
How many parts in your design John, a few dozen? A hundred? Two hundred? Which one is the most expensive? Do you prefer "customized" parts built OEM to your specifications to off the shelf parts which perform the same function just as well? The advantage to special ordered one of a kind parts is that when the equipment needs repair, you have a monopoly on spare parts and can charge whatever you like for them. Who would not spend $1500 to replace what would otherswise have been a six dollar transformer in a $40,000 preamplifier?
- Status
- Not open for further replies.