|
|
Ranting of a crazed Audiophile...Vol. 1Mar 12 '06 Write an essay on this topic.
Popular Products in Music
The Bottom Line There are no good commercially available speakers, the task is to find the least bad ones…or build your own.
Ranting of a crazed audiophile Vol. 1 I’m often asked for suggestions on audio equipment in particular, on loudspeakers. Over the years my thoughts on this have gradually been focused to one simple response. They’re all bad! If I say, “Bose sucks” and you are a Bosephile, don’t get offended because I am not prejudice, I hate them all! Well ok maybe that’s an exaggeration. Based on price, build quality etc, speakers could be categorized into multiple levels ranging from, “junk-fi” at the bottom consisting of typical Walmart level equipment and certain more expensive systems designed for compactness rather than sound quality all the way up to the, “Ultra-hi-fi” consisting of the $125,000 Wilson Audio and similar products. It is not a good assumption to think that the more you pay the better quality you will get. While this is somewhat true in general, there are simply too many exceptions in both directions to rely on it. While several types of speaker systems exist, the most common is the conventional monopole. A monopole is a point source or point sources all with the same phase and orientation. For example, a typical three way speaker with a woofer, midrange, and tweeter all mounted on the front of the speaker is a monopole. Similar speakers with side mounted woofers are actually also monopoles because the radiation of the woofer is omni directional and it makes no significant difference that it is side mounted. In order to create a true to life recreation of a recorded musical event, a speaker must convert the electrical information into acoustical information as accurately as possible and then it must couple this acoustical information to the listening environment correctly. If it fails in either of these, the listening experience will suffer. First let’s examine the monopole concept itself without evaluating any particular loudspeaker product. If the monopole proves to be incapable of properly coupling the acoustic energy to the listening environment, then it should be discarded as a viable approach to loudspeaker design. If this proves to be the case, then we know that the majority of modern loudspeakers used around the world, regardless of brand name, price, or quality are fundamentally flawed. Even the most audio inept person probably has a basic understanding that a properly designed speaker system should have a flat on axis frequency response. What most people probably do not understand however is that this data contains very little information about the loudspeaker’s true frequency response. The true frequency response of a speaker in a room is more commonly called the polar response or power response. This information can not be ascertained by a simple on axis measurement but rather requires a large collection of measurements made off axis up to and including 180 deg off axis. Most speaker manufacturers do not publish polar information about their speakers. One company in the “low-fi” end of the spectrum (certainly at least in terms of price) steps above many higher priced competitors by publishing data to 75 deg off axis. These guys are pointed in the right direction but fall short by not evaluating the polar response of the speaker to 180 deg off axis. Because the data at 75 deg already proves the speaker will not uniformly load the room with acoustic energy and because it is a foregone conclusion that the problem will only get worse especially as you go beyond 90 deg, it would be pointless to continue. Let me see if I can explain this farther. . Think of it like this. Imagine a circle drawn around the speaker with the speaker at the center of the circle. Let’s say the circle has a radius of 10. It doesn’t matter 10 what, we won’t need units here but you could think of it as dB because, along the circle edge, the distance from the speaker represents the level in dB. Imagine that this circle is the polar pattern of the speaker at a particular frequency in the bass region. The cross sectional area of this circle defined by pi*r^2 (pi * 10^2 = 314) represents the acoustic energy radiated by the speaker at this particular bass frequency. Again, the distance from the speaker to the edge of the circle represents the level in dB so at this bass frequency, a sound pressure meter would measure the same level in front of, to the side of, and behind the speaker. The bass frequency is non directional so it radiates around the speaker almost equally 360 deg. Now imagine a second circle sitting in front of the speaker with a diameter of 10. This circle just touches the front of the speaker on one side and just touches the outer edge of the first circle on the other side. It must extend to the same point as the original circle for the levels to be the same. This circle has a diameter of 10 whereas the first circle had a radius of 10. Ok, imagine now that this smaller circle represents the polar pattern of the speaker at a particular high frequency. The cross sectional area of this circle defined by pi*r^2 (pi * 5^2 = 78.5) defines the amount of acoustical energy radiated by the speaker at this particular high frequency. In different words, the speaker is radiating a bass frequency at a particular amplitude in all directions. (+/- 180 degrees.) The sum of all directions at this amplitude represents the total bass energy. The speaker is radiating a high frequency at the same amplitude in a few directions. (for illustration lets say +/- 60 deg.) The sum of these few directions at this amplitude represents the total high frequency energy. Let’s say the amplitude in each case is 10. For the bass the total is 10 * 360 or 3600 while the total for the high frequency is 10 * 120 or 1200. If we increase the high frequency amplitude to 30 then the total of the high frequency energy will match the total of the bass but the on axis frequency response will no longer be flat. So, our conclusion is that there is no way a monopole design can ever truly have a flat frequency response. Spending $125,000 will not fix this. It can only be fixed by finding another approach to speaker design, one which will provide a uniform power response. The only way to fix the problem is for the speaker to be made so that it radiates the same amplitude at all frequencies and so that it radiates all frequencies in the same directions. There are two solutions. The first solution is to make the high frequency equally non directional matching the already non directional low frequency. The second possible solution is to make the low frequency equally directional matching the already directional high frequency. The first approach is called an “Omni directional” speaker and the second approach is called a, “dipole” speaker. Both these type radiation patterns can produce very uniform power response in the listening environment. The primary difference between the two is that the omni directional will maximize room interaction by maximizing the ratio of reflected energy at the listening position whereas the dipole will minimize room interaction by maximizing direct energy at the listening position. Only a few omni directional speakers exist in the commercial market, and there are even fewer dipoles. Most conventional speakers are monopole. While it is not physically possible for a monopole speaker to create a true recreation of a musical event, some are less bad than others. Remember, in addition to properly coupling the acoustic energy to the room, the speaker must first accurately convert the electrical information into acoustic information. The variables in this process are numerous and have a great impact on sound quality. So, while two monopole speakers may suffer the same non uniform power response, one may have significant advantages over the other in other areas. Non linear distortion such as harmonic distortion is the addition of energy at multiples of the stimulus frequency. For example if one component of a particular musical note is 100 Hz, the speaker will also produce energy at 200 Hz, 300 Hz, and so on. The degree to which a speaker does this will have a great impact on sound quality. Another example is linear distortion or time related distortion. Stored energy in the cone can cause the speaker to continue radiating acoustic energy after the stimulus is removed causing the sound to become blurred. The loudspeaker driver itself will have problems with this and it varies based on the design of the driver. The baffle can also create stored energy if incorrectly designed. All these factors can play a major part in the overall sound quality of the speaker. The design of the crossover inside the speaker also plays a major part in sound quality. The crossover is an electrical circuit consisting of resistors, capacitors, and inductors arranged such that low frequencies are directed to woofers and high frequencies are directed to tweeters. The design of a good crossover is quite complex especially in the case of passive crossovers which are the most common type. If your speaker has more than one size drive element such as a woofer and a tweeter and if didn’t come with a box that must be connected between your preamplifier and your power amplifier(s), “otherwise damage will occur”, then your speaker should have some sort of internal passive crossover… or an unreasonable facsimile. So, where is all this nonsense going? Are you going to know what speaker to go buy after reading this? Probably not! The best type speaker in terms of how it couples acoustic energy to the room is the dipole because it can, if properly designed, radiate acoustic energy uniformly as a function of frequency; and because its lack of off axis radiation reduces room interaction. Taking it a step farther, the best type is an infinite line source dipole but that will be for a later ranting. The problem is, if you want a truly good speaker, you will most likely have to build it yourself. A number of people have designed good quality dipole speakers (no infinite lines that I am aware of however) that are available either in kit form or the details have been posted for the taking. I’m not here to recommend any certain one so if interested just do a Google search for dipole open baffle speakers. Short of going dipole which generally requires active crossovers and multiple amplifiers, the next best solution is to find the “least bad” monopole or bipole you can find. Someone recently asked me about ##### speakers but until that point I had never heard of them. The moment the ##### web site popped up, the first impression was not good. In my opinion they looked cheap! After a little research it quickly became clear these are very much an entry level product and while I have never seen them in person or listened to them, I find it highly unlikely they are what I term, “listenable”. The two things that most influenced this conclusion was the listing of similar products used by owner reviewers on another opinion site and by an unlikely specification. A listing of other “similar products” used by the owner reviewers were all low end mass market products so that wasn’t a good sign. The second clue was the weight specification. Yes, the weight! Not the frequency response or any other technical specification but the weight. They’re too light! If I assume they use ¾” MDF then based on cabinet dimensions, the drivers must be weightless or lighter. Even if they are using only ½” MDF, the drivers must have very thin stamped frames and have small magnets. So, combined with the low price of the speaker in question and these clues, I concluded this was probably not a great choice and directed the inquirer toward similar priced kits from Madisound and Parts Express. So, there can be clues as to a speaker’s quality other than detailed measurement data. Knock on the side of the cabinet, if it sounds like a hollow door the cabinet walls are too thin and not properly braced. This will produce muddled sound. Pick it up, does it fee like a ton of bricks? If not, again, the cabinet walls are probably too thin and the drivers are probably too light. It’s difficult for me to write about how to choose a good conventional monopole speaker since I believe there are none. It doesn’t matter how good you make a square wheel, it still doesn’t work very well! A future ranting will go into detail about how to choose a conventional monopole speaker but I can already tell you you’re not going to like the conclusions because these will involve you buying and assembling a kit. Invariably you can always do better this way! And now you know why I call it ranting! The Maniac… Note: A variation of the dipole concept is widely used in side surround speakers. In this application the dipole null is directed toward the listening position thus minimizing direct energy and maximizing reflected energy. In the case of a dipole main speaker, the same concept is used to do the reverse. Additionally, properly designed dipole speakers require a complex specifically contoured transfer function in the crossover to correct for the inherent -6dB/octave slope below a value known as F(peak). This transfer function also requires corrections for F(peak) itself and any wave guide effect inadvertently created by the baffle design. Most dipole side surround speakers will not include these important corrections in the internal crossover as it is too complex. This is most effectively done actively prior to the power amplifier. Opinions expressed here are my own and may differ from yours. You have every right to have a different opinion regardless of how incorrect that opinion might be;-)  |
| Write the first comment on this review! |
![Jersey Boys [Original Broadway Cast Recording] by Original Broadway Cast Reviews](http://di2.shopping.com/images/pi/95/a4/8a/2006752046-100x100-0-0.jpg?p=p10.0cc2a76c1ab2015b91d6&a=2&c=1&l=8071320&t=120604042626&r=2)
