 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
In Reply to: Re: What are we here for? posted by john curl on October 21, 2003 at 10:25:55:
John...good point, so here is some fodder.There are a couple of factors to be included in your discussion, source devices, and the role of specifications as a marketing tool.
The frequency response of microphones changed, keeping pace with the technologies mentioned, but dictated intially the requirements for the remainder of the electronics chain from recording to reproduction.
Ribbon type microphones were quite an advancement, and my favorites, the Neumann and Telefunken U47 condenser mics were a huge improvement, requiring much wider bandwith I believe. Low cost dynamic types then became popular, but the bandwith desired for the condenser microphones may have previously set the standard...Also, consumer tube amplifiers designed in the 50's/60's were engineered for stability in general, requiring limited bandwidth...remember the first time someone told you that if you dis-connect a speaker lead when a tube amp is ON, the amp will "blow"? Try it with any Mcintosh or Marantz or Eico; of course they do not experience a failure. The term "phase margin" was fully understood by all but the stupidest of designers at the time. It is no wonder feedback amplifers are out of favour, how many current designers have read Bode's "Feedback Amplifier Design"? You could count them on one hand.
Could one design a stable tube amplifier with response to 300 KHz? Of course it can be done, but given the target build costs for the consumer market at the time, it was not even a possibility.While I like to see full power to at least 100 KHz in a tube amplifier design, and do believe it increases the "realism" of the music,I also consider the Marantz 9 to be a wonderful sounding amp generally. It is steeply bandwith limited at both low and high frequencies; in fact I think it utilizes more reactive components than any other consumer amplifier. Then consider the following:
When solid state designs became popular, frequency response could be extended dramatically in feedback amplifiers, due to a large reduction in reactive coupling components, and elimination of the output transformer phase shifts. Direct-coupled, semiconductor bloated designs could result in some impressive specifications, but sometimes horrible sound. Was the high frequency extension actually sonically beneficial?
Comments?
Follow Ups:
Good input, but I would also like to note that the reason that we were told to always load a vacuum tube amp was that the output transformer would act up without a load to send the power to. Solid state power amps with output transformers would have the same problem I presume.
Yes, I agree that it is difficult to make completely stable solid state power amps, but for a number of subtle reasons, as well as not well understanding Bode plots and stability criteria very well. For example, the Marantz 14, the first solid state power amp made by the Marantz, and presumably by the same staff who developed the Marantz 8,9, etc, could oscillate with a 0.1uF cap across the load. This company had real engineers, but still had a potential problem with stability with this amp.
That's the trouble with much of audio progress -- there are too many variables to really pinpoint cause and effect.In many high-feedback SS amps, you will see "lots" of dist products of 5-10th order. Nasty stuff! And of course, bandwidth-limiting is a good way to avoid slewing problems.
So it's hard to say whether extended bandwidth by itself is a good (or bad) thing.....
I suspect that most modern moving coils extend well past 20kHz. The Ortofon I use (Kontrapunkt B) is spec'ed at 50kHz (-3dB). I believe (but certainly cannot prove) that this contributes to an airier/more transparent top end, since it implies that the mechanical resonances are further from the music.
Peter
I think it was Ottala et.al. who conducted some research into supersonic detectability yeaars ago, and the conclusion was that, yes, the so called "supersonic signals" were noticeably present or absent to the test subjects...However, I have also personally noted that a significant number of "audiophiles" do not immediately detect a failed tweeter in their own system.SA
I am the individual who looked for ultrasonic (supersonic) components to the greatest extent, 25 years ago. Dr. Otala even recommended that I use the term 'ultrasonic' rather than 'supersonic' in my IEEE paper published in 1978, but it had already been submitted. This was in connection with TIM.
I have measured components of up to 500KHz coming from phono cartridges. Typically 200KHz could be seen in several popular MC cartridges; also analog magnetic tape and microphones could have flat response to 40KHz and perhaps beyond, in special cases. The best results shown on the internet are from a fellow at CalTech, but I don't have his name at the moment. Perhaps someone can contribute.
The most important factor is probably transient response changes with severely curtailed bandwidth. Above 20KHz, this is most probably the significant change. Even an uneven and gently rolling response above 20KHz, is MUCH BETTER in transient response than a brickwall filter, especially practical ones, but even with 'perfect' examples.
Not to minimize your research John, I used Otala as an example mainly because there are more older text references to his work than yours.Back to the subject of "ultrasonic" source material content, what are the origins of the ultrasonics generated by a group of musical instruments? Is it simply instrument (violin, for example) harmonics or is it also naturally ocurring intermodulation products generated between instruments? Has anyone done spectral analysis of a symphony orchestra?
SA
SA, it would interest me if you could refer me to earlier articles where Otala used 'supersonic' rather than 'ultrasonic'. You see, he was criticizing me for using 'supersonic' but I got the word from somewhere. ;-) I think that guy from CalTech was named Boyk or something like that. I still haven't found my copy of his work.
It was James Boyk, and I believe some of his more recent work in this area may be online at CalTech. Re: M. Otala, I will keep that in mind, and see what I happen across.
Of course, the guy from CalTech, who I alluded to, has done major work in this area. However, practically, most microphones will start rolling off after 20KHz. It is not a steep rate of rolloff, and it is fairly benign as far as group delay and risetime, compared to CD.
Here is the article:
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
