Wavebourn, my Jurassic tests were done with B&K 1902 & 2010 in a compressor loop so spls (and excursion) were the same with & without the simple feedback. If you read Stahl's paper carefully, you'll see he gets his distortion reduction tested the same way ie excursion the same in both cases.
No one is disputing your point that bigger cones & smaller excursions give less distortion.
What ACE & Marshy's simple resistor do is reduce distortion for the same spl and excursion. {You can apply EQ to Marshy's system to get the same response as without the resistor. That's essentially what Yamaha AST does.}
Did you measure distortion in your tests? How did you do it?
I've done what you suggest and I don't get similar results with negative resistance. I get less distortion. Another technique where you see distortion reduction is with current drive.
Stiffness & BL profile are NOT the only factors in LF distortion of moving coil speakers. Some of the other factors are just as important. This is a dynamic phenomena and static tests only show part of the story. Try Mills & Hawkesford from AES and look up their references.
Last edited:
And it is exactly what I do in my home theater (I like to use myself as a guinea pig when experimenting) - I drive a pair of 12" drivers in the concrete horn under the floor by an amp with negative output resistance.
Bur we started this off-topic discusion from the point when you insisted that your small speakers will beat Robert's big speakers in a listening test because you once damaged the church.
Sorry if I misunderstood you.
Bur we started this off-topic discusion from the point when you insisted that your small speakers will beat Robert's big speakers in a listening test because you once damaged the church.
Sorry if I misunderstood you.
Of course it is not the only mechanism that shows that higher excursion needed from smaller driver causes more distortions. If to follow your style of discussion I should now refer you to works of Newton, Maxwell, Grasgoff, Prantle, and so on, and suggest you to read and try to understand.
But the main point was, I agree with Richard RNMarsh, and I myself want to use bigger drivers the next time I have a chance to open the floor. Not because I don't have enough of SPL, but because sometimes I hear sounds from subwoofer instead of feeling them. I want lower distortions. Or to use a rotary sub, if I can afford it. I don't need expensive microphones to measure what I hear. I just don't want to hear what I hear. Very simple. If you hear 10 Hz sinewave it is not a sinewave anymore.
Last edited:
This can't actually be true.
But much as I would love to discuss the causes of nonlinear distortion in loudspeakers, I am always reminded of the fact that no valid scientific tests have ever shown a correlation between THD or IMD and perception. In fact all attempts at this have shown that there is no correlation. Hence a detailed discussion of what causes something that is inaudible is rather pointless, don't you think?
That is a completely separate issue and it is clear that we can probably never test my outrageous claim.
Just cos I believe one special small speaker with little bass will be chosen by Mr. Marsh (its Richard BTW) over his big un's dun mean I don't like big un's or can't do them properly. Indeed Powered Integrated Super Sub tech is all about big performance from big OR small boxes.
But if I may be so bold to ask, "What is your point?".
Is it that you doubt any statement on bass response of speakers from a beach bum? This is a perfectly valid view and if such obvious delusions from opinionated people who have only small experience in subjects of which you are expert, offend you, please tell us.
Or are you simply correcting my naive errors? If so, please be gentle.
But back to something useful. How do you measure LF distortion of speakers? What did you find with and without the -ve resistance? I'm eager to learn.
What can't be true, and what is inaudible? Either I have audio hallucinations, or I indeed hear that distortions are higher with higher excursion.
Quite recently (the last week) after making new book shelf I connected signal generator to the subwoofer amp and went through frequencies from 10 to 40 Hz to eliminate rattlings on the shelf. And of course I heard increase of distortions on 10 Hz with increase of signal level. What I heard, were obviously distortions, because if I said that heard 10 Hz clean sinewave I should go directly to psychiatrist.
Here is the picture of the shelf taken yesterday:
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Well said Mr. Lee. I'll draw your attention to Peter Fryer's work on the subject; both in AES as well as the HiFi News article.
However, subs are slightly different in that an overloading sub is usually pretty obvious and give some of Wavebourn's symptoms.
Wavebourn, Powered Integrated Super Stuff is meant to address precisely what you describe. It can be applied to small speakers but also has good results on huge speakers .. reducing measured but also audible distortions.
If I'm ever in Pleasant Hill, may I visit and suggest some simple improvements? Is the system mainly for music or HT? What programme has those 10Hz signals? Do you have a dedicated processor for the LF EQ?
Powered Integrated Super Stuff has many dirty tricks but AFAIK, it obeys Newton, Maxwell & Prandtl (dunno about Grasgrof & Prantle)
I appreciate what you're saying, but subjective hearing is telling me the magnitudes are of the same order as the normal factors, for me. I have absolutely no interest in whether a particular frequency band is being reproduced "correctly" or not as far as amplitude or phase is concerned, what I worry about is what many people call the "noise floor": this is the ability of the system to reproduce convincingly all the low level detail in the recording, that which makes makes playback sound "real" or not. Most systems are pretty hopeless at this, and it is what I particularly concentrate on getting right.
And, these weirdo things that I mention are exactly the factors that have a dramatic effect on the low level rendition, so they are chalk and cheese for me.
The real benefit of doing this, is that I can then put on really "bad" recordings, ones that people constantly point the finger at as examples of lousy recording techniques, and they make musical, acoustic sense; I can wind them up to deafening level and they still make sense, no problems with fatigue. Only problem, these days, my ears tell me, okay, that's enough of that level of SPLs for a while ...
Frank
I'm just impressed that you got your visitors to do a blind test of whether they could detect that 10Hz signal ...
, Frank
I'm just impressed that you got your visitors to do a blind test of whether they could detect that 10Hz signal ...
Yes, sofas were shaking pretty well during the 3-D blind test. They were watching this movie:
An externally hosted image should be here but it was not working when we last tested it.
I thought this was just a theoretical foray on your part but now I see you have a problem with your sub.I connected signal generator to the subwoofer amp and went through frequencies from 10 to 40 Hz to eliminate rattlings on the shelf. And of course I heard increase of distortions on 10 Hz with increase of signal level. What I heard, were obviously distortions, because if I said that heard 10 Hz clean sinewave I should go directly to psychiatrist.
If you don't mind accepting advice from a beach bum, I may be able to help .. but only if it is OK.
If you feel you have the problem and solution in hand, just tell me to go forth & multiply.
It's not about Wavebourn's symptoms, but about Wavebourn's proof that speakers generate audible distortions that increase with excursion.
Even completely blind person can't deny the fact when hears a sub, while in case of woofers can stubbornly cite AES and similar gurus.
You are welcome to be my guest, of course. I am always eager to learn from people who really know well what they are talking about, so can explain in simple terms, without cloudy references to names and papers.
The last 2 gentlemen defined the criterion of laminar/turbulent movement of an air that you will inevitably hit reducing size of the cone. It was a joke, as you may understand. Sorry for misspelling their names, I learned the theory about 35 years ago, in Cyrillic letters.
Come on... You can experiment with any sub and hear the same "problem".
And you are welcome to be my guest, I already invited you, and hear the difference. Until then your speculations sound offensive.
Recycled Knowledge -
Notice that the spl was not the usual for measuring thd or response from a bass speaker.... it is often measured very very close to the driver... the level tested was out 3 feet and at that distance was 90db. I dont know - blanket statement coming - but 90 db is easy to get when you measure right up close to a driver with very little input power.
[Or, a simple analogy is if I talk at 70 db spl from 3 feet away and then at same level right AT you ear and directly into it, it will seem too loud but not at 3 feet. Whispering power levels to a close up microphone hardly is suitable for thd measurments which will be listened to far out in a large room at 90db spl. Most speakers are much worse in practice/listening room power levels than the standard measurments indicate]
The ONLY reason I cut this nice walnut cabinet up, closed off the horn opening and made it a reflex design was: 1- did not go deep as the speaker could do in a reflex design and 2- the thd was audibly unsatisying. Bass was blurry and not clear. All tests on this 15 inch high quality JBL driver were done in same location and conditions. SPL was same each time, as shown, and only the frequency response and thd were measured. The article this came from was about a power amp design and not speakers but I just wanted to add the 'bonus' feature and show the reduction in thd and the Q was also reduced.. less boomy... as you can see from the shape of the thd curve. The end result was thd at a loud level was reduced to 4% at 30Hz.... a figure the original speaker system could not even reproduce. And, the reduction was almost by half for the bass-reflex.
Its interesting ... so here it is back for another round of people to try it. Recycled knowledge, I call it. THx- Richy Marshy
Notice that the spl was not the usual for measuring thd or response from a bass speaker.... it is often measured very very close to the driver... the level tested was out 3 feet and at that distance was 90db. I dont know - blanket statement coming - but 90 db is easy to get when you measure right up close to a driver with very little input power.
[Or, a simple analogy is if I talk at 70 db spl from 3 feet away and then at same level right AT you ear and directly into it, it will seem too loud but not at 3 feet. Whispering power levels to a close up microphone hardly is suitable for thd measurments which will be listened to far out in a large room at 90db spl. Most speakers are much worse in practice/listening room power levels than the standard measurments indicate]
The ONLY reason I cut this nice walnut cabinet up, closed off the horn opening and made it a reflex design was: 1- did not go deep as the speaker could do in a reflex design and 2- the thd was audibly unsatisying. Bass was blurry and not clear. All tests on this 15 inch high quality JBL driver were done in same location and conditions. SPL was same each time, as shown, and only the frequency response and thd were measured. The article this came from was about a power amp design and not speakers but I just wanted to add the 'bonus' feature and show the reduction in thd and the Q was also reduced.. less boomy... as you can see from the shape of the thd curve. The end result was thd at a loud level was reduced to 4% at 30Hz.... a figure the original speaker system could not even reproduce. And, the reduction was almost by half for the bass-reflex.
Its interesting ... so here it is back for another round of people to try it. Recycled knowledge, I call it. THx- Richy Marshy
Last edited:
Just some links, about the tragic of some of our discussions about 'high end'. It is about the quite expensive Wilson audio speakers, but much speakers at no price will not give us much better .
Listen to all those blablas and then, the sound recorded out from those enclosures. Did it makes illusion half a second ? NO ! It is just pathetic.
Building a New Loudspeaker: The Creation of Sophia Series 3 - YouTube
WILSON AUDIO &AR&VTL&DCS&TRANSPARENT - YouTube
Wilson Audio Watt Puppy 5.1 - Mastersound 845 Compact (part 1) - YouTube
Wilson Audio Maxx2, Mastersound 845 Evolution, Audionet ART V2 part 1 - YouTube
Listen to all those blablas and then, the sound recorded out from those enclosures. Did it makes illusion half a second ? NO ! It is just pathetic.
Building a New Loudspeaker: The Creation of Sophia Series 3 - YouTube
WILSON AUDIO &AR&VTL&DCS&TRANSPARENT - YouTube
Wilson Audio Watt Puppy 5.1 - Mastersound 845 Compact (part 1) - YouTube
Wilson Audio Maxx2, Mastersound 845 Evolution, Audionet ART V2 part 1 - YouTube
As far as I'm aware, that 6dB variation of SPL, per doubling, or halving of distance from a driver holds true, forgetting about room effects for the moment.
So, say, 1W gives 90db at 1m, then 96 at 0.5m; 102, 0.25m; 108, 12cm; 114, 6cm. The last is a touch over 2" away, this is probably as close as one can go for this relationship to have any chance of still being true; still a reasonable volume that the hearing system can handle for a short period of time.
In fact, this is a technique I use all the time to "measure distortion"; sticking your ear this close makes distortion artifacts extremely obvious, especially for the treble driver ...
Frank
1W/1M vs actual use distance/SPL/THD -
Now do the same 90db at 9-12 feet away -out in the room at listening position.... The speaker is obviously going to be working harder. The thd at realistic listening levels/distances is pretty high compared to 1W at 1Meter. Simple point. Just that we forget the numbers measured or in a spec sheet (rarely) are almost always lower than what you will be hearing/recieving. Which Maybe why JC thought the THD was higher than he thought it might have been.
However, I wasnt concerned about the absolute level of THD - only the comparitive reduction in THD with motional-feedback. -RNM
Now do the same 90db at 9-12 feet away -out in the room at listening position.... The speaker is obviously going to be working harder. The thd at realistic listening levels/distances is pretty high compared to 1W at 1Meter. Simple point. Just that we forget the numbers measured or in a spec sheet (rarely) are almost always lower than what you will be hearing/recieving. Which Maybe why JC thought the THD was higher than he thought it might have been.
However, I wasnt concerned about the absolute level of THD - only the comparitive reduction in THD with motional-feedback. -RNM
Last edited:
- Status
- Not open for further replies.