John Dunlavy's posts to rec.audio.* during 1999: New messages will be added here as I see them on Usenet; hopefully in less than a week. Thanks to Dunlavy for this information. Since July, Mr. Dunlavy has been posting with much greater frequency and in a more interactive manner than previously; I am archiving primarily the posts where he explains something I believe will be of general, enduring interest. Last revised 12 November 1999. -------------------------------------------------------------------------------- From 102365.2026@CompuServe.COM Mon Apr 05 17:42:38 1999 Newsgroups: rec.audio.opinion Subject: Re: Sweet Spot From: Dunlavy Audio Labs <102365.2026@CompuServe.COM> Date: Mon, 05 Apr 1999 18:42:38 -0400 From recent posts I have read, there seems to be a lot of confusion regarding the subject of a loudspeaker's "sweet-spot" and the properties of a loudspeaker that contribute to the width of the listening area over which an accurate stereo soundstage can be perceived. To begin, the term "sweet-spot" applies to two separate properties of a loudspeaker: 1) the listening area over which an accurate spectral balance is reproduced, and 2) the listening area over which accurate imaging and an accurate stereo soundstage are reproduced. These two kinds of sweet-spots may or may not co-exist, depending upon the following loudspeaker properties: 1) the horizontal width of the listening area at a given distance over which the spectral balance (frequency response) of the loudspeaker is preserved, and 2) the degree to which the loudspeakers are "pair-matched" with respect to their frequency and phase responses, both on and off axis. Thus, a pair of loudspeakers may exhibit a reasonably good spectral balance on-axis and off-axis and yet blur the reproduction of an accurately recorded soundstage, with respect to precise imaging of discrete sound sources. Indeed, a pair of loudspeakers with a wide horizontal dispersion pattern may be exhibit a subjectively smooth spectral balance over a wide horizontal listening window, and yet be incapable of providing "pin-point imaging" of accurately recorded sounds. Likewise, a pair of loudspeakers with nearly perfectly matched on-axis amplitude and phase responses vs. frequency, may exhibit a relatively narrow listening window with respect to off-axis spectral-balance. Seem contradictory? Hardly! Our hearing process determines spectral balance and direction (angle of arrival) of a wide-band sound source (music, for example) by different means. The sensitivity of the human hearing process to the "spectral balance" of a loudspeaker is relatively poor. However, relatively small differences in the phase response between two loudspeakers comprising a stereo pair, can markedly blur the reproduction of an accurately recorded soundstage or shift the perceived direction of a given sound at a given frequency. Thus, truly accurate reproduction of the recorded soundstage is dependent upon accurately matching the amplitude response, phase response and directivity patterns of a pair of stereo loudspeakers and ensuring that the listener is located precisely equidistant between them. This explains why a pair of truly accurate, well-matched loudspeakers require that more attention be given to choosing the correct listening position, equidistant from both speakers, so as to obtain the most precise stereo imaging. A pair of loudspeakers with less "pair matching" with respect to amplitude and phase Vs frequency, will yield a wider listening area over which, however, the stereo image will be less precise and more amorphous. Anyone wish to comment? Best of listening, John Dunlavy From daljhd@cs.com Sat Jul 24 17:22:50 1999 Newsgroups: rec.audio.high-end Subject: Re: Naive question on directional cables From: daljhd@cs.com (DALJHD) Date: 24 Jul 1999 22:22:50 GMT Steve Lampen is a competent engineer and a friend. We have discussed how so many audiophiles came to believe that electrically conducting wire and cables exhibit an audible directional property. I would suggest that it probably began in the fertile imagination of someone within the Adv or Mkt Dept. of an audiophile cable company looking for a "unique claim" to boost sales of cables. As a competent engineer/physicist, I join Steve and others like myself, in stating unequivocally that the electrical wire used in manufacturing loudspeaker cables does not and can not exhibit any audible or measurable directional properties. This lack of directionality has been demonstrated on many occassions during properly conducted, non-intimading, "blind A/B comparisons" of numerous loudspeaker cables. It can also be demonstrated to not exist by a complete set of highly-accurate, "lab-quality" measurements of a cable's electrical properties, e.g., series resistence, series inductance, parallel capacitance, loss vs. frequency, frequency dispersive properties, V.P. factor, etc., PERFORMED IN BOTH CABLE DIRECTIONS. AMEN! So! Why do so many honest audiophiles, who have participated in "blind A/B comparisons performed by cable manufactureres (or their agents)", become firm believers in the "directionality of loudspeaker cables"? Well, ask any medical doctor about what has become known as the "Placebo Effect", a well documented human trait? It is frequently used to treat patients with "imagined illneses" by prescribing pills that contain nothing but sugar. (It "cures" their illness almost every time.) Any comments? Best regards, John Dunlavy From daljhd@cs.com Sat Jul 24 17:22:11 1999 Newsgroups: rec.audio.high-end Subject: Re: Electrostatic Loudspeakers, information requested From: daljhd@cs.com (DALJHD) Date: 24 Jul 1999 22:22:11 GMT ESL's, and membrane type loudspeakers in general, present numerous demanding challenges for anyone undertaking their design. This would especially apply to anyone without reasonable competency in the disciplines of electronics, physics and acoustics Yes, they look relatively simple - but don't be deceived by their appearance. For example, the choice of membrane material (its thickness, weight per unit area, elasticity, dimensional stability (Vs temp., etc.), electrical properties (dielectric constant, ohms/cm, etc.), the properties required of the step-up transformer, etc. etc. Then comes the "stretching of the diaphragm over the rigid frame and permanently securing it in place, etc. This is followed by the design and construction of the grid of charged wires. Thereafter, the design of the high-voltage power supply and the required wideband audio step-up transformer present their own challenge. I could go on and on. How do I know - well, I was enamored with membrane loudspeakers back in the 1950,s and '60's, owned (and loved) a pair of Quad ESL-63's, and decided to design my own. Although I possessed a competent background in both physics and E.E., I soon discovered the project was beyond my capabilities. First, and perhaps most important, I lacked the necessary lab equipment and anechoic chamber for performing the necessary measurements. Set back a bit by daunting aspects of the task confronting me, I decided to examine the properties of membrane loudspeakers in more detail before proceeding. After some considerable "text-book" research, I decided to abandon the project for the deeper I delved into the measurable and audible performance attributes of loudspeakers using a large membrane radiating surface, the more I came to realize that it was not as accurate a reproducer of complex musical transients, etc. as I had envisioned. Don't get me wrong! It was fun learning how something that looks so simple and straight-forward can turn into a near nightmare of challenges. And, without wanting to purchase the necessary measurement equipment, e.g., precision microphones, chart recorder, a large number of expensive acoustic foam wedges for creating a small "anechoic environment (for measurements), etc. , I simply gave up. Perhaps, there are those who have accepted the challenge and succeeded in designing and constructing their own ESL - but how did it sound and measure. Maybe good - but probably not as good as a used pair of ESL-63's! Or as good as a well-designed, conventional loudspeaker using a time/phase aligned array of drivers, fed by a first-order crossover, compensated for driver anomalies. (But here, too, an anechoic measuremnent capability is required to obtain documentably accurate performance.) Anyway - look before you leap. ESL's look simple and their operation is relatively simple - but their design and construction can be a nightmare. Anyone have a different experience? I hope so! Best of luck, John Dunlavy From daljhd@cs.com Mon Jul 26 13:14:29 1999 Newsgroups: rec.audio.high-end Subject: Re: Question on directional speaker cables From: daljhd@cs.com (DALJHD) Date: 26 Jul 1999 18:14:29 GMT Having lectured "transmission theory" and "network theory" before E.E. classes on several occassions, I concur with the comments posted by Dick Pierce, Steve Lampen (of Belden), and others possessing competent E.E. backgrounds. It is too bad that so much distorted information is being disseminated about audible and measurable properties of audiophile cables by persons (and companies) who should know better. The properties of electrical cables (including loudspeaker and interconnect types) and the way electrical currents flow along them is a subject very well known by competent electrical engineers and physicists. There are no secrets nor new discoveries within this field - only the "gobbledygook and flooby-dust" interpretations given the subject by those lacking a proper grasp of the subject. I suspect that many well-known audiophile cable companies rejoice when reading posts that appear to concur with their advertisements regarding such non-existent properties as "cable directivity", etc. But without such advertising claims, I doubt they could sell many of their expensive, hi-tech appearing cables. At least some of the bogus advertising claims may be traced to the fact that few, if any, audiophile cable companies have any competent electrical/electronic engineers on their payroll. But most audiophile cable companies do not manufacture their own cables, the manufacturing being accomplished by large companies such as Belden. (And engineers working for these manufacturers, such as Steve Lampen of Belden, do not agree with the advertising claims of the audiophile cable companies that promote and sell the cables.) Some audiophile magazines also deserve a bit of criticism because some of the "technical" articles they have published appear to support such nonsense as "cable directivity, slow-moving electrons carrying the signal, significant audible differences between cables, and so forth". Likewise, the contention of audiophile cable companies that "blind listening comparisons" are invalid because they "create stress that precludes listeners hearing audible differences between cables" is pure nonsense. The proper "non stressful" protocol to use in determining whether audible differences actually exist is to begin by letting listeners know which cable is being heard - but, at a time unknown to the listeners, the identity of the cables is switched, such that cable "A" becomes cable "B". This protocol, though employing deception, prevents any possible build-up of "stress" among participants. Anyway, the subject of "audible differences between cables" is an important one that needs and deserves informed discussion and comment. Best regards, John Dunlavy From daljhd@cs.com Tue Jul 27 16:32:28 1999 Newsgroups: rec.audio.high-end Subject: Re: Should I change my cables? From: daljhd@cs.com (DALJHD) Date: 27 Jul 1999 21:32:28 GMT Hi Suzy, As a competent engineer with good technical credentials (and the designer of the loudspeakers you are using), my advise is simple: don't waste money on expensive cables - hoping for an audible improvement in the sound sound quality of your system. The audiophile cable industry is rife with false advertising claims that cannot be verified either by competent measurements or blind A/B comparisons. Most comparisons conducted by those who have monetary gain at stake, use cleverly planted suggestions to "help you discern differences" that do not actually exist and cannot be discerned during "truly blind comparisons". Anyone saying otherwise is either not worthy of being believed or is someone who has yet to discover the truth about most "cable differences". Go slow and don't let a fast-talking salesman sell you a load of expensive, high tech appearing cables that have a very very low probability of audibly improving your existing system. If, however, you are seriously concerned about the quality and performance of the cables you are presently using, look for cables whose manufacturer furnishes a complete set of "performance specifications" (based upon measurements of guaranteed accuracy). This may at least provide some peace of mind, compared to the purchase of cables advertised by means an overdose of meaningless superlatives, etc. Caveat Emptor! Best of luck, John Dunlavy From daljhd@cs.com Tue Jul 27 16:32:18 1999 Newsgroups: rec.audio.high-end Subject: Re: abc's of cables From: daljhd@cs.com (DALJHD) Date: 27 Jul 1999 21:32:18 GMT It's a bit of a tough question to answer because it involves some design aspects of the electrical circuitry, etc. of the equipment components being connected. Assuming high-quality audiophile components,with well-designed circuitry, there should be little or no audible difference between the use of shielded balanced interconnect cables and coaxial (low capacitance) cables. Within some systems, however, shielded balanced cables might provide improved rejection of nearby A.C. hum fields (if they are a problem). Also, it should also be noted that with an "all balanced system", using a pre-amp with "balanced (push-pull) stages" and a power amp with a "balanced (push-pull) input circuit", the use of balanced interconnect cables may help preserve the rejection of "even-order harmonic distortion products" inherent in such a system, compared to that of a system with "single-ended circuit topologies" and coaxial interconnect cables, I have used both balanced and un-balanced systems during the past several years and seldom encountered any consistently audible differences between them. Thus, since an "all balanced system" typically is more expensive than one designed with an un-balanced topology, I would personally spend the difference to purchase a more expensive un-balanced system with better performance properties. Best of listening. John Dunlavy From daljhd@cs.com Wed Jul 28 22:18:49 1999 Newsgroups: rec.audio.high-end Subject: Re: Question on directional speaker cables From: daljhd@cs.com (DALJHD) Date: Thu, 29 Jul 1999 03:18:49 GMT Steve Lampham is quite right. It should be pointed out, however, for those lacking a E.E. background, that electrons travel very slowly along a conductor. However, when elctrons are induced to flow by a "potential difference", they create an electromagnetic field that propagates along the wire at nearly the speed of light, slowed down only a relatively small amount by the presence of a dielectric (insulating material) or by the diameter of the conductor if it becomes very large. Those posting on the Cable Group of the Audio Asylm refuse to believe this and many other well known attributes and properties of cables and the travel of currents and fields along them. As a consequence of their simplistic views, many fervently hold to the belief that the propagation of electrical currents along a cable is directional. Oh well, Hmmm! In general, expensive audiophile loudspeaker cables are a disgrace, of sorts, to our hobby. Yes, I designed (and DAL sells) a cable called the Z-6 which exhibits a characteristic impedance at high frequenies (above the audio range) of approximately 6 Ohms. Of course, at audio frequencies characteristic impedance has little meaning and engineers characterise a cable using "network parameters", such a R,L and C. However, both methods essentially "say the same thing" and are, to some degree interchangeable. By that, if one is rigorous, transmission-line calculations should yield the same answers as R, L and C calculations, if resistive losses are ignored. Yes, I know this is a bit of an over-simpification, but I state it to merely convey that their is an intimate connection between the two. To subjectivists, there are significant audible differences between cables that defy measurement. However, I suspect that most competent engineers soudl contest this view on the basis of "theory", accurate measurements, and blind A/B comparisons within a properly operating system. To this, subjectivists reply that blind A/B comparisons induce psychological stress that dulls the listeners ability to discern audible diferences. Hmmm! Any comments - (No flames)? Best of listening, John D. From daljhd@cs.com Fri Jul 30 10:44:51 1999 Newsgroups: rec.audio.high-end Subject: Re: abc's of cables From: daljhd@cs.com (DALJHD) Date: 30 Jul 1999 15:44:51 GMT Perhaps, I am missing something but I fail to understand how any property of an interconnect cable can limit the "dynamic range" of an audiophile system? Maybe, if its shielding is inadequate and permits hum to enter the system, that could be interpreted as limiting dynamic range by virtue of degrading the signal-to-noise ratio of the system. Likewise, I fail to understand how an interconnect cable can "... generate all kinds of different frequency responses." An interconnect cable of any reasonable length can only alter the "high frequency roll-off" of a system by virtue of its parallel capacitance (and, perhaps, by an unlikely high series resistance component). The series inductance exhibited by almost any cable, largely a function of the diameter of the center conductor, is unlikely to be sufficiently high to create a roll-off in high frequency response within a signal path whose impedance is higher than 50-75 Ohms. (Most standard coaxial and balanced shielded cables exhibit nearly flat responses up to hundreds of MHz. I agree with Fred that the use of balanced cables, especially within pro-systems at many recording and mastering studios (typically 200 to 600 Ohm circuits) makes a lot of sense with respect to potentially improving system signal-to-noise ratio and reducing hum levels. Best regards, John D. From daljhd@cs.com Sun Aug 01 11:09:14 1999 Newsgroups: rec.audio.high-end Subject: Re: Should I change my cables? From: daljhd@cs.com (DALJHD) Date: 1 Aug 1999 16:09:14 GMT Don't spend much time worrying about audible differences between loudspeaker cables. Most of the advertising copy used to sell expensive, "hi-tech appearing" cables is written by adv/mkt execs who know very little (if anything) about cable properties and measurable/audible performance. Indeed, during competently-conducted, blind A/B comparisons between high-quality 12 AWG ZIP cord and several very expensive loudspeaker cables (within a preoperly-operating system using accurate loudspeakers with a decent input impedance) it is virtually impossible to identify which cable is being heard - beyond that predicted by "chance". Read the latest issue of the Sensible Sound - which contains a well-writen article covering the subject. And, read some of the recent postings here on the NET by several highly qualified audiophiles possessing competent technical and academic credentials. Don't forget. Above all, refuse to participate in "non-blind", subtly coached, A/B cable comparisons at some dealers who love the high profits they reap from cable sales. When "told what differences to listen for" during cable comparisons, the listener will almost always believe they can hear such differences. John D. From daljhd@cs.com Mon Aug 02 10:30:22 1999 Newsgroups: rec.audio.high-end Subject: Re: abc's of cables From: daljhd@cs.com (DALJHD) Date: 2 Aug 1999 15:30:22 GMT Gary seems to imply that some property of loudspeaker cables might limit their "dynamic range". As an engineer and physicist, with appropriate professional credentials, I fail to understand the basis for his claim. I am unaware of how any known property of a loudspeaker cable can affect its "dynamic range", either measurably or audibly. Perhaps, lousy end-connectors with oxidation, etc.? All of the electrical properties of cables are well known, are measurable with tolerances that significantly transend our hearing/perception capabilities, and the effects of these properties and their interaction with amps and loudspeakers is well understood by competent E.E.'s. The big problem is that many manufacturers and some magazine editors have a lot to gain by promoting highly-profitable cables with a high-tech appearance. Sadly, their promotion often consists of flooby-dust/gobbledygook specs., cleverly-managed (not-very-blind) A/B comparisons (where listeners are coached as to what they should hear), etc. But the teachings of competent electrical engineering and transmission-line theory reveal no audible properties of audiophile cables that cannot be measured, properly quantified and fully explained. (Wanna bet otherwise?) Read some of the recent postings by competent engineers, professors, etc. who have spoken out on the subject here on RAHE. Also read the latest issue of the "Sensible Sound" magazine, which contains an excellent article on the subject of cables by Howard Ferstler. Best of listening, John D. From daljhd@cs.com Thu Aug 05 13:11:15 1999 Newsgroups: rec.audio.high-end Subject: Re: Most Accurate Loudspeaker? From: daljhd@cs.com (DALJHD) Date: 5 Aug 1999 18:11:15 GMT Indeed! Perhaps, we should give up our pursuit of "true accuracy", based upon a competent evaluation of a complete set of accurate measurements, combined with carefully controlled comparisons with live music, and attempt to locate an old non-electronic "Victrola" with a fresh supply of steel needles? (But where would we find a supply of old 78 RPM records?) Hmmm! Those who do not realize what a competent interpretation of a complete set of accurate measurements can reveal about a loudspeaker's ability to accurately reproduce complex musical waveforms, are simply uninformed about what today's true science and tecnology have to offer. And, to give up chasing "perfect reproduction", although knowing it will never truly exist, seems a bit like saying we should no longer attempt to design and manufacture better cars, homes, TV sets, etc. (But while engaging in the "chase", it is simply prudent to recognize that not all loudspeakers, etc. are created equal - with respect to true, documentable accuracy.) Further, it seems to me that giving up the "chase" would make for an awfully boring and unrewarding life. Best regards, John D. From daljhd@cs.com Thu Aug 05 13:11:30 1999 Newsgroups: rec.audio.high-end Subject: Re: Wilson Watt/Puppy From: daljhd@cs.com (DALJHD) Date: 5 Aug 1999 18:11:30 GMT To reproduce squarewaves (which consist of the fundamental plus odd-order harmonics (of gradually diminishing intensity) does not require that amps, loudspeakers, etc. exhibit bandwidths approaching infinity. Indeed, to reproduce squarewaves (or other "complex waveforms) with an audible accuracy equal to that available from a system whose high-frequency response is flat to infinity, requires only that the audio system (amps, loud-speakers, etc.) have a bandwidth extending to the limit of human hearing,.e.g.,, 20 kHz., with a good modulus of amplitude vs. frequency and a good impulse response. Since squarewaves, impulses, etc. are waveforms frequently approximated by the sounds produced by many musical instruments (and their combination within an orchestra), it makes good sense that all components of an audiophile system be able to accurately reproduce such waveforms. Why not? Because limited audibility testing, conducted under questionable conditions, etc., failed to fully confirm that the accurate reproduction of complex waveforms was not essential? Gosh! Gee whiz! Hmmm! John D. From daljhd@cs.com Fri Aug 06 13:06:30 1999 Newsgroups: rec.audio.high-end Subject: Re: Most Accurate Loudspeaker? From: daljhd@cs.com (DALJHD) Date: 6 Aug 1999 18:06:30 GMT Some good questionsare being asked here on RAHE - the answers to which I believe may affect the future vitality and integrity of High-End Audio. There is an old saying: "Measurements don't lie - but measurers often do"! Don't audiophiles understand that a lack of accurate measurements (especially for loudspeakers) is a "cover up" for hiding poor performance or the kind of performance that "can be heard but not measured"? Why are most audiophiles so trusting when it comes to the accuracy of technical specs, performance claims, measurements, etc.? Are companies with little technical integrity "winning the battle" by convincing audiophiles that measurements are useless for determining the potential of a product to deliver accurate reproduction? Gulp! And, where is our expensive government (Consumer Protection Agency, etc.) with respect to providing some assurance that highly advertised performance claims have at least a small measure of legitimacy? Obviously, none of us want our government to pursue a "witch hunt" for bad guys - but how about letting companies know they are "responsible for the accuracy of advertised performance claims, specifications, and measurements"? And, one might ask, where are our cherished audiophile magazines when it comes to truly accurate coverage of equipment and loudspeaker performance? Don't they make enough money from advertising and the sale of mags to occassionally hire a technically competent engineer or physicist as a consultant to verify their opinions regarding product performance, etc.? You would be surprised how many times I have had to answer the question, "If a full set loudspeaker performance measurements are important, why is your company apparently the only one to provide them and guarantee their accuracy?" Hmmm! Perhaps, one answer might be for audiophiles to insist that manufacturers making claims for accuracy, etc. back up their claims with guaranteed accurate measurements, etc. - not merely subjective opinions? And, yes! Some of today's best "digital room correction systems" can work wonders with really bad rooms - but they cannot fully correct for poor time-domain response and impulse smearing exhibited by a poorly designed loudspeaker. Keep tootin' guyes and gals! There is still the old saying, "In the end, the truth will out!" Best of listening, John D. From daljhd@cs.com Wed Aug 11 12:18:09 1999 Newsgroups: rec.audio.high-end Subject: Re: abc's of cables From: daljhd@cs.com (DALJHD) Date: 11 Aug 1999 17:18:09 GMT Rob Gold wrote: >OK, John. What. for us non-technical types for whom R,L and C are >fuzzy letters on the eye chart, constitutes a basic "good enough; >anything more is turd polish" cable? May I assume that you'd need >information on things like a speaker's impedence and a pre-amp's >output impedence to make a reasonable recommendation? How important >is "dressing" (set-up) in terms of cable performance? FWIW, I am >running Kimber PBJ interconnects and Kimber 4TC speaker cable, all >bought on the theory of "minimum good enough." Actually, as I have said many times here on RAHE, within most audiophile systems, No. 12 AWZ Zip Cord is adequate for use with most amps and well-designed loudspeakers with an input impedance that does not drop below about 3 Ohms. With respect to interconnect cables, the best-quality ( but modestly priced) ones sold by Radio Shack are among the very best we have measured and used. You can spend a lot more and, believe it or not, get a lot less with respect to such important attributes as capacitance, shielding, mechanically induced noise (piezo-electric properties), etc. Sure, we have designed and sell interconnect cables that are without peer, with respect to important electrical properties, but they are no better, audibly, than the Rat-Shack types within most systems. So - don't be deceived by a glitzy hi-tech appearance, meaningless hi-tech sounding specs, a high price-tag, etc. Spend you money for components that might at least have a chance of improving the audible accuracy of your system. And don't get caught near "magic disks, green pens, etc."! Best of listening, John D. -------------------------- From daljhd@cs.com Thu Aug 26 11:24:19 1999 Newsgroups: rec.audio.high-end Subject: Re: A question for Mr. John Dunlavy From: daljhd@cs.com (DALJHD) Date: 26 Aug 1999 16:24:19 GMT Hi Michael, Thanks for your reply. It was not my intent to question your opinions - only to make certain that readers understand the amount of technical effort we put into ensuring the "true accuracy" of our products. To begin, properly designed "fabric dome tweeters" perform better than "metal dome tweeters" for several good and important reasons. The softer material of a fabric dome provides much superior "damping" properties than that of a stiff, rigid metal dome. This is because a stiff, rigid metal dome possesses a hi-Q property that leads to the "storage of energy" that is not radiated. Since this energy must eventually be dissipated by some means (it cannot simply vanish), it appears as "ringing in the time domain" and as "large resonant peaks" in the frequency domain. A dome made of a soft material with a "high-loss factor" is less efficient than a metal dome (typically by a few dB) but stores much less energy per unit of time, resulting in the reproduction of a more accurate impulse and step, with far less ringing - and a much smoother frequency response. This superior damping and better accuracy of fabric domes is evident in measurements of "impulse response", step response, and frequency response. Keep asking informed questions. We all have a lot to learn! Best regards, John D. From daljhd@cs.com Sat Aug 28 13:51:11 1999 Newsgroups: rec.audio.high-end Subject: Re: Ported Spkr Bass Question From: daljhd@cs.com (DALJHD) Date: 28 Aug 1999 18:51:11 GMT Just thought a few technical observations about "ported subs" might be interesting to you and other readers. In physics, there is an old saw that says, simply, "there are no free lunches!". If you begin with a woofer driver in a sealed enclosure whose volume more-or-less matches the requirements of the driver's electrical/mechanical properties, the bandwidth, the minus 3 dB frequency, the roll-off slope and the pass-band SPL can be calculated with reasonable accuracy using the well known methods published by Thiel and Small many years ago (not the Thiel who has a loudspeaker company). The low-end "roll-off" of a typical ported bass enclosure exceeds 24 dB per octave below system resonance. Generally, the slope of the low-frequency roll-off of a sealed woofer enclosure, without internal absorbing material, occurs at a rate of about 12 dB per octave (after the 1st octave, or so). If you add open-cell foam, etc. with good acoustical absorption properties, to the interior of the sealed enclosure, the "system Q" will drop, the overall efficiency will be lowered, the system resonance will drop and the overall system bandwidth will be increased (with fewer peaks and nulls in response). Again, these properties are readily calculable and measurable. These "improved properties" can be partly attributed to the simple fact that the velocity of an acoustical wave propagating through a lossy medium (such as open-cell foam) is lower than it is through air. And, the presence of "lossy, open-cell material" greatly diminishes the amplitude of "standing-waves" within the enclosure (which often create peaks and valleys in amplitude vs frequency response). If you take the same enclosure, sans the absorbing material, and create a "port", the efficiency will increase within that portion of the frequency range affected by the port. However, the laws of physics require that the "system bandwidth of the woofer-enclosure" be reduced. This results in the low frequency roll-off (below resonance) becoming much more rapid than 12 dB per octave, e.g., 24-30 dB/octave (or more) being typical. This much more rapid roll-off actually reduces the amplitude level of radiation at very low audio bass frequencies - where bass energy is often "felt more than heard". Thus, while a port can add a "boost in efficiency" to a woofer-enclosure system, it will typically also reduce the amplitude of bass radiated at frequencies below resonance So, the laws of physics always "exact a price" - nothing is ever a "freebie". Thus, you have a choice between a "slightly higher efficiency" with a ported enclosure (but with "a rapid roll-off in low-end bass") or a "slightly lower efficiency" with a sealed enclosure (but with an extended low-end response). And, of course, the impulse response of a properly designed and damped sealed enclosure is virtually free of "over-hang and ringing", while the impulse response of a ported enclosure is hardly one to brag about. Personally, being an "accuracy freak", I much prefer the accurate sound quality of a properly designed, sealed woofer enclosure. Best of listening, John D. From daljhd@cs.com Tue Sep 14 15:38:55 1999 Newsgroups: rec.audio.high-end Subject: Re: A question for Mr. John Dunlavy From: daljhd@cs.com (DALJHD) Date: 14 Sep 1999 20:38:55 GMT Yes, as I thought I had mentioned in previous posts, we take a very large number of anechoic chamber measurements, both on and off axis. As one would expect, the measured properties related to accurate reproduction do not significantly change out to about thirty degrees off-axis. Beyond that, the "path alignment" of the drivers becomes increasingly altered by the differential path differences. This results in an increasing degradation in the "off axis" properties of multi-driver loudspeakers. Even membrane loudspeakers, having a single radiating surface, suffer off-axis performance problems because of directivity problems at frequencies where the width of the radiating surface becomes large with respect to a wavelength. This is the reason that the manufactureres of most accurate loudspeakers recommend "toeing in" their loudspeakers such that they are pointed at and equidistant from the preferred listening location. What we need is a small diameter dome radiator that exhibits flat amplitude/phase response over the entire audio spectrum and exhibits a high efficiency, good input impedance, etc., etc. Hmmm! John D. From daljhd@cs.com Fri Sep 17 11:07:03 1999 Newsgroups: rec.audio.high-end Subject: Re: Another question for Mr. John Dunlavy From: daljhd@cs.com (DALJHD) Date: 17 Sep 1999 16:07:03 GMT Locating a woofer driver on a large ground-plane, such that the outer edge of its cone is essentially flush with the surface of the ground-plane may not be an accurate means for determining the woofer's "anechoic response", etc. When a radiating surface (woofer driver), with dimensions small with respect to a wavelength within the spectrum being measured, is located flush with a ground plane, it is radiating into "2 Pi space", which yields a "uni-directional" radiation pattern with a 3 dB "directivity gain" at frequencies where the driver's "radiation pattern" would be essentially "spherical" when mounted in the box with dimentions small relative to a wavelength.. One way to visualize this is to imagine a "spherical balloon" in free space with a radius equal to "1" (one). When the same balloon is mounted over a ground-plane such that the lower-half of the balloon is flattened, its new radius is increased by 1.414 times (as a hemisphere with the same volume). This difference in "radius" yields a "power gain" of two (2) or 3 dB. This is elementary theory for antenna engineers who deal daily with such problems. Any comments? John D.. From daljhd@cs.com Sat Sep 18 11:20:20 1999 Newsgroups: rec.audio.high-end Subject: Re: Another question for Mr. John Dunlavy From: daljhd@cs.com (DALJHD) Date: 18 Sep 1999 16:20:20 GMT Sorry, Stewart, but burying a loudspeaker with its cabinet face flush with the ground will produce measurements that exclude the rather nasty radiation components attributable to "diffraction" from the edges of the enclosure. Solving this problem led to my using acoustical absorbing material between affected drivers and the edges of the enclosure. Originally, AR claimed they had received the patent for this invention. NOT SO! I was the inventor and received the patent - not AR. (US Pat. 4,167,985) Hmmm! Anyway, burying a loudspeaker with its face flush with the ground does not yield an effective nor accurate means for measuring the SPL, freq. resp., directivity patterns, etc. of a loudspeaker. Nor will laying it with its back side flat on the ground solve the problem, because at mid and low frequencies, radiation from the drivers will refract and diffract from the enclosure edges, with considerable energy headed in the direction of the ground, where it will be reflected in forward directions, combine with the direct driver radiation and create peaks and nulls in freq. resp., exagerate impulse-response ringing, alter the shape of the step-response, etc. etc. Best of listening, John D. From daljhd@cs.com Wed Oct 06 17:38:07 1999 Newsgroups: rec.audio.high-end Subject: Re: Dunlavy vs. The Tube From: daljhd@cs.com (DALJHD) Date: 6 Oct 1999 22:38:07 GMT In comparing the measured square-wave reproduction of a loudspeaker to that of an amplifier, one must consider the different measurement protocols involved. Measuring the square-wave reproduction of an audio amplifier is relatively simple, straightforward and unambiguous. However, measuring the square-wave reproduction of a loudspeaker within an anechoic chamber is hardly a simple or straight-forward task. This is because most anechoic chambers, including the two large chambers at DAL, are imperfect with respect to internal reflections. In particular, they impose various limitations on the accuracy of frequency and time domain measurements at the normal on-axis listening distance of 10 feet, especially below about 200 Hertz. Indeed, even relatively large chambers (like the two at DAL, 24 ft. long X 20 ft. wide X 16 ft. high) with the best, most efficient absorbing material (alternating, two-feet deep wedges of high-density, open-cell acoustical foam completely covering all internal services, exhibit some measurable reflections. As these reflections become increasingly large in amplitude (relative to the direct-path signal), with decreasing frequency, they prevent taking highly-accurate "direct" measurements of some performance attributes, such as amplitude Vs freq. response, low-freq. square-wave reproduction, etc. Indeed, measurements of a loudspeaker with "perfect" square-wave reproduction would reveal some "ripples" and a "tilt" at the top and bottom of square-waves at frequencies below a few hundred Hertz, due to these reflections from internal chamber surfaces. Thus, when viewing "anechoic chamber measurements" of loudspeaker square-wave reproduction with those of audiophile amplifiers, the "chamber anomalies" must be taken into consideration. When this consideration is factored into the "comparison", the reproduction of DAL loudspeakers is found to be comparable to that of any tube-amps with the same frequency response attributes. Thus, in this context, the square-wave measurements of DAL's physically small SM-I loudspeaker, sent to a few requesting them, are not too "shabby". Best of listening, John D. From daljhd@cs.com Fri Nov 12 16:05:11 1999 Newsgroups: rec.audio.high-end Subject: Re: Hypothesis: Cables Do Sound Different From: daljhd@cs.com (DALJHD) Date: 12 Nov 1999 22:05:11 GMT Hi Phil, To begin, we used our SC-IV/A and SC-V loudspeakers, powered by either a Spectron Musician (digital amp) or by two other high-quality, high slew-rate, solid-state amps of different manufacture, both using class-A circuit topologies. Listening material consisted of very high quality DVD, CD and SACD recordings played through the latest Cal Audio transport and a 24-bit 96 kHz D-A converter. The comparisons were made within our main audiophile listening room, an acoustically well-damped room about 16 feet by 24 feet (with the loudspeakers placed along the 24 ft. wall with a separation of about 100 degrees (included angle viewed from the listening position). We used our own ultra-low loss, low capacitance (less than 6 pF per ft), well shielded interconnect cables for connecting the various equipments. Loudspeaker cables were either switched manually or by means of switches possessing very low loss properties (less than 0.05 dB into loads as low as 1 Ohm, measured in our very well equiped lab facility). Manual switching of loudspeaker cables was accomplished by using three separate persons, one behind each loudspeaker and one adjacent to the power amp. Switching between cables was performed within a maximum interval of three to four seconds. I believe that this was sufficiently rapid to permit listeners to store what they had heard and mentally compare the sounds of the cables. To ensure that "personal biases" did not affect the results, comparisons made use of two different regimens: 1) the listeners knew which cable was being heard only by "number" and, 2) at reasonably frequent intervals, the cables appeared to be switched when they were not. The results of these "blind" comparisons clearly revealed that listeners could not identify any audible differences between loudspeaker cables beyond that predicted by "chance". I believe our results confirm those of prior cable comparisons performed by other persons using competent protocols. The results of our listening comparisons of loudspeaker cables also confirm the lack of audible differences predicted by an informed evaluation of a full set of accurate lab measurements, encompassing all known and measurable properties of cables, including series resistance, series inductance, parallel capacitance, VP factor Vs frequency, non-linear properties, etc So! Along with many other competent investigators, we have concluded that no audible improvements are available from the use of any expensive, hi-tech appearing cables compared to an equal length of quality 12 AWG ZIP Cord (such as that made by Belden and sold by many audiophile companies - as well as Home Depot, etc.) Perhaps, the most important attribute of any loudspeaker cable is the quality of the connections at both the amp and loudspeaker ends of the cable. For best long-term results, these should be gold plated types, soldered or competently "crimped" to the wires. Once again, I suspect that a few posters will ask why I designed (and DAL sells) loudspeaker cables? Well, as I have said before, it began as an "engineering exercise" to determine whether a reasonably long, nearly lossless cable with a 6-Ohm "characteristic impedance" (not resistance) and almost imeasurably low losses over the entire audio spectrum could yield an audible improvement within a typical high-quality audiophile system. Our numerous, carefully controlled listening tests have consistently revealed that not even DAL's Z-6 cable can achieve an audible improvement over other cables, ranging from 12 AWG ZIP cord to the most expensive offerings of well known audiophile cable companies. Comparisons of lab measurements of the loss Vs frequency of 20 ft. lengths of different loudspeaker cables reveals why: even a 20 ft1length of 12 AWG Zip Cord added only about 0.2 Ohms (at 20 kHz) to the measured resistance of a precision 2 Ohm "load resistor" - hardly an amount that would be audible within any audiophile system. And the measured phase angle with the 20 ft Zip Cord was less than about plus 10 degrees at 20 kHz. With respect to the question posted regarding "Acoustic Memory", as it applies to the proper protocol for accurate A-B comparisons, I beleive that trained listeners can compare the important properties of two different acoustical events if they are sequentially presented within a time lapse not exceeding about 3 to 5 seconds, Beyond this time lapse, memory probably begins to fail at something like a log rate. Anyway, I hope that the above info satisfies the curiosity and or disbelief concerning our measurements and listening comparisons of loudspeaker cables. There really are no unknowns that cannot be resolved by a combination of a complete set of accurate lab measurements (competently interpreted), the teachings of competent science and engineering, and true, blind comparisons using a proper protocol (and properly operating equipment). Best of listening, John D. End of archive. New posts will be added as I become aware of them. -------------------------------------------------------------------------------- I hope you find this information as interesting and useful as I have. The other contributions to these threads (and many others) can be found by searching at www.dejanews.com (deja.com) and/or in the rec.audio.high-end archives. These and other topics are also discussed in a series of 'white papers' available from Dunlavy Audio Labs, as well as in the August 1996 'Stereophile' interview with John Dunlavy. I have no connection with DAL except customer (and enthusiast...:-).