|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
213.226.71.67
HiJohn in one of your responses in our discussions here about distortion you mentioned a rather extreme weighting factor:
“The most extreme that I have seen is: Multiplier = (n-1)! / 2”
Do you have the references at hand, because I would like to include this one in my history part of my thesis (Witch I really have to finish soon, I have people bugging me and asking when it’s going to be ready). But I need a reference so that I can check it out myself and I can’t put it in without a proper reference. (Or otherwise I just have to write “Personal conversation with John Curl”) :-)
The other one is the brain-child of D E L Shorter from “The Influence of High-Order Products in Non-Linear Distortion” Electronic Engineering April 1950
“Multiplier = n (squared)/4”
And with Multiplier = n / 2 being to “gentle”
And Soundmind we surely have to be able to use and refer to articles in publications like Wireless World, Electronic Engineering and others, beside AES publications and IEEE?
They sure are in a other class than Stereophile Mag… Or? Because if you are a skilled engineer you should be able to make your own judgement of the content of an article in these mags.
Follow Ups:
For my own estimates, I think that I will use a rough mean between the two weighting factors. This turns out to be:
20dB or ten times for 5th harmonic
40dB or 100 times for 7th harmonic
perhaps the same for 9th harmonic as 7th harmonic.
I am going on experience here, and my understanding of the easy audibility of higher odd order non-linearities.
I usually use harmonic distortion, because I can measure down to about -120dB, but as you know, IM would give us more distortion, especially triple beat, and certainly distortion products in portions of the audio spectrum that would be easily picked up where our ears are most sensitive. Of course, a narrow view of harmonic distortion, makes it almost useless, unless IM is also considered.
Rolf, I got the other weighting factor from a 1973 article in 'Wireless World' by Bob Stewart (Meridian). You don't have to reference me, since I got it from the article, and Bob probably references it himself.
You might contact Meridian to get a copy of the article.
A few years ago, I tried to get another copy of this article, but the university library didn't have this particular year on its shelf. You might be luckier. As I recall, it was a 3 part article.
Rolf, that should have been: 'Bob Stuart' of Meridian. I just checked their website, but the paper doesn't seem to be there.
Hi JohnDo you have any memories of the actual name of the article? Because my university library doesn’t have Wireless World. And I have to order it and they will kill me if I come in with a request with only:
Mag: Wireless World 1973
Author: Bob Stuart
Maybe three partsI’m harassing them enough already with my requests of articles and papers. They know me by now and just asks: What difficult things do you want us to get now?
They are doing a splendid job actually. And you can never do to good research.
Sorry, you should contact Bob Stuart at Meridian. They have a big website, it should be easy.
Rolf, at CES I talked with Robert Stuart. If you e-mail him at :
jrs@meridian.co.uk http://www.meridian-audio.com/ and ask him for the 1973 articles and anything else that he might have, he could e-mail them to you.
Use my name in your e-mail, and please send what you get to me as well, as my e-mail send is still out. This should help you.
Mad Man & John,a '73 paper from Bob Stuart rang a bell and I searched through my 'archives' and found this :
'An approach to audio amplifier design' by J.R. Stuart. Wireless World, Aug 1973, pp 387 - 391. I have only the first part.In covering subjective weighting of harmonics he references Shorter and also mentions Wigan. The reference for Wigan is :
E.R.Wigan "New distortion Criteria", Electronic Technology, April 1961, p.126.
Thanks for the input, 13!
Rolf, got your e-mail. If you get anything from R. Stuart, please e-mail it to me .
Here, we would use: J. Curl, "private conversation". Many times, the info from the conversations doesn't exist in print yet.MMH: ""And Soundmind we surely have to be able to use and refer to articles in publications like Wireless World, Electronic Engineering and others, beside AES publications and IEEE?
They sure are in a other class than Stereophile Mag… Or? Because if you are a skilled engineer you should be able to make your own judgement of the content of an article in these mags.""
The problem I have is that many times, even skilled engineers are led astray by articles in non peer reviewed mags...And then, bandwidth is wasted explaining to all exactly why the article was floobydust. Sometimes they are led astray because the article is just barely outside the zone of understanding...sometimes they are led astray by the credentials of the author. I am reminded of the magazine article on faster than light signal propagation..Yes, sometimes peer review can be responsible for delaying an understanding that is real, but the instances where most are led down the wrong path for decades are far worse than months or years of delays..right now, the physics community is wrestling with co-authors,referees, and on line publication issues, so this is probably gonna get worse. There's a lot of web sites out there that are spouting garbage that is being believed by many.
Three very good examples...
1. The standard argument against cables is the 20 to 20K bandwidth, amplitude variation spiel..this is what was taught, and has served well..unfortunately, it does not address lateralization aspects. While those who state "wires don't make a difference" are correct for the vast majority of applications, their baseline does not include the timing criteria needed for left-right imaging. When that criteria is fully established, the engineering community will certainly be able to correct their design techniques...and then, what is done now will be considered "slipshod" with respect to lateralization. I've posted on this here, at AH, and over at that ghost town AR.
2. The second is the old article on skin effect, which has it's own errors of assumptions at the very start of the paper, and diverges very quickly away from reality, including an error in testing. I posted lots on that here.
3. Third: The Belden site has an analysis on transmission lines, and how, at lower frequencies, the characteristic impedance goes up..Unfortunately, it relies on an initial assumption that uses an RC approximation.... that is incorrect..The analysis is quite useable for all T line use, but it is actually only an approximation to reality. The actual model is better defined using a fifth order polynomial, but that is mathematical overkill..the RC model works adequately. I posted that analysis over at AH, if you would like a link..
BTW, those sites support jpegs, so I can post graphs and pics there, though I've tried not to use their sites for picture hosting without contributing there.
Overall, I'd consider articles in non peer review magazines as a basis for conversation and thought, but have seen many cases where the articles are terribly incorrect but yet take on a life of their own.
Hi John.Anyone guess my position on this? It's always the same. As with the whole cable arguement, the level of harmonic distortion which is audible over an entire recording/reproduction system is something that should and can be identified with DBTs up front. This is the design criteria for the minimum performance of the system. And reliance on that research should be sought in professional journals, not consumer hobbyist magazines or related third party conversations. This is the starting point for determining what is required of each component and of the overall system. Then you can make an intelligent decision about whether a better design is warranted or of value. When no such standard exists, the makers and sellers can have a field day perpetually talking in generalities and perpetually introducing new models long after their older models have far surpassed any benefit improved performance in that criteria can offer in real usable terms. And hobbyists with more money than knowledge will continue to devour it.
“Overall, I'd consider articles in non peer review magazines as a basis for conversation and thought, but have seen many cases where the articles are terribly incorrect but yet take on a life of their own.”I fully agree with you here, I have myself encountered articles witch have been close to the truth and first been fooled. But sometimes after a second reading (or third or forth) I’ve got a sinking feeling that says “Hey just wait a minute, this doesn’t sound quite right”. And consulted books or other engineers with more knowledge in the specific subject.
And a couple of times my fellow engineers have pointed out that on the subject we discussed earlier I was wrong. Then you just have to sit there and be ashamed. ;-)
“There's a lot of web sites out there that are spouting garbage that is being believed by many.”
And just not web sites. I know because I had hard times to explain to non engineers why they are wrong on some specific topic, and their argument is “It said so in this mag, and what it says there MUST be true” OhhhmyyyGoood What to do??? Just kill them??? (Not the writers but the stupid readers) ;-)
I’ve studied transmission line theory but didn’t find it interesting enough to learn more than I needed to pass the course. So I’m not qualified to make claims or defend any regarding transmission line theory. I have to stick to what I know about non-linear components and my EMI knowledge.
Part of my design theory is that “things” have to behave well up to aprox 500 kHz AND be quiet above that. WERY QUIET!
This post is made possible by the generous support of people like you and our sponsors: