Q&A # 171 - April 7, 2000

Staff

Q I have a modified CD player connected to my DSP-A1. It has been modified so that it has a higher output voltage. I've noticed that while playing a CD at high volumes there is noticeable distortion. It even gets to the point that the song stops playing. This only happens when the Yamaha is in a DSP mode. In "Effect Off" play the sound is fine. A friend said that the gain may be too high for the modified CD player and that this may be oversaturating the DSP chip. Does this sound accurate? If so how do I change the gain?

A I suspect the higher voltage is overloading the DSP circuit, as you said, and I assume you are talking about the analog outputs of the CD player. I would suggest using the digital output of the CD player instead, and letting the DSP-A1's DAC handle the decoding.

Q One of my friends had a problem similar to one of your readers: a dented speaker cone dust cap. Instead of using the pin on the shiny silver dust cap, he was told by some local audiophiles to use (CAREFULLY) the vacuum cleaner hose. It worked great, and no pinholes. (We started with the hose's pressure vent all the way open.)

A Thanks for your input. It sounds like a great idea.

Q I just read your opinion on the Yamaha 2095 and was very impressed. I have been looking at this receiver for a couple of weeks now and was hoping with your expertise you can help me. I have had an HK AVR30 for years and never suffered any problems with it. Along with the 2095 I am looking at the HK7000, and the Denon 4800. What would you recommend? I have also looked at the DSP-A1 but was totally lost. I love quality and want something that will last and can grow with the future. I hope you have the time to give me some much needed advice.

A The Yamaha DSP-A1 is one of the best receivers out there. Now that the RX-V1 has become available, I imagine you would be able to find the DSP-A1 at a good discount, and I would suggest getting one (the DSP-A1) if you can buy it at $1,500. The only thing you would be missing is the surround EX channel which goes in the rear - middle. Otherwise, one of the new receivers that has surround EX would be best. Remember, though, EX is not a discrete digital channel. It is extracted from the rear left/right channels.

Q Can you please explain zero, half, and full space acoustic loading?  Advantages and reasons why they use it etc?  The resources I have don't really fully explain it.

A These terms are referring to effects imposed by nearby boundaries which affect the efficiency/directionality of the output of some radiating device. It might be the baffle of the speaker with relatively higher frequencies and larger baffles, or it might be a wall or floor. As you add another theoretical plane, the effect doubles again, and with a third, creating a corner situation, the effect doubles once more. Conventional wisdom is that the gain for each added plane in output is 3 dB, although some sources have quoted 6 dB, pointing out that the increase in efficiency by the greater acoustic impedance is on top of the increase in relative output in the given direction because the output itself is focused. The matter becomes more academic when you put speakers in a real room which introduces factors such as reflection, absorption, and losses due to energy leaking through the boundaries, which we could think of as absorption. So, along these lines, one could think of full space as no boundaries, half space as one boundary, and so forth, with zero space being six boundaries so closely spaced that there is no air volume, an infinitely high acoustic impedance, and an extremely high ratio of radiated energy from the transducer.

In terms of baffles, whether or not the wave "sees" the increase in acoustic impedance of the baffle depends very much on the frequency and the size of the baffle. As the frequency decreases, the wavelength increases, so it requires a larger baffle or boundary to create an effective boundary. If the wavelength is much longer than the boundary, the pressure wave simply rolls around that boundary. An example of such would be a subwoofer, in which case both the woofer (the radiating surface) and the side of the box that the woofer is mounted to are usually far smaller than the associated wavelength (20 feet or so at 50 Hz). As a result, bass in this situation is absolutely omni directional. It's interesting to note that this is not the case with a baffle at higher frequencies. Not only does treble and upper midrange not radiate significantly to the rear on a unidirectional loudspeaker of conventional design, but given a flat total power response, in an anechoic situation, the on-axis response (straight ahead) will have a rising response with frequency due to the narrowing directivity due to the effects of the baffle, and even due to the diameter of the driver diaphragm, which in turn has the same effect, even if the baffle is no larger than the driver itself.

In addition, within a multi driver system, a large midrange, if run high enough, can start to become quite directional. (Even more so than a simple baffle, as the large radiating area itself focuses the output. Roughly speaking, as the diameter approaches or exceeds the wavelength of the frequency, the driver will "beam.") In contrast, at the crossover frequency, because of its much smaller physical size, the tweeter is relatively omni-directional, its dispersion only possibly confined with the baffle's "half space." It is not until the very top of the treble range where the tweeter will "beam." And so, speakers with large mid-range drivers (more so that those with stiff cones which radiate higher frequencies from the whole diaphragm as opposed to mostly from the center) often have more uneven off-axis frequency responses, typically with a dip in the upper mid-range, which may alter the ratio of energy in the total reflected sound, creating a rather laid-back presentation even if the on-axis frequency response is relatively flat.

In terms of really low frequencies that cannot "see" most baffles, walls are much larger, and in the cases of most rooms, they do not simply end, but become the beginning of other boundaries. If the baffles were infinite in size, the effect would be consistent, but since they're usually part of each other, and some of the sound bounces back and forth multiple times, the 3 dB (or 6 dB) gain per nearby boundary becomes more theoretical than exact. What becomes really interesting is when you get VERY low frequencies in a relatively small room. As the wavelength approaches 4 times the  distance between the farthest walls, in a sealed environment, the loading does not act as a wave at all. Instead, ALL the walls begin to become a single, or rather, an infinite number of boundaries, and response BELOW that point will rise at a rate of 12 dB/octave as frequency falls, where the loading is referred to as a pressure mode. Of course, this won't EXACTLY occur in real life situations either, because we don't usually live in air tight canisters.

Q I have some Carver ALIII ribbon speakers, and I need help with room treatments.

A Since you have the ALIIIs, you will need to deal with radiation patterns of the rear wave from the back of the speaker (due to the dipolar nature), which will primarily affect treatment of the wall behind the speakers. The lack of energy above and below the ribbon (due to the line source) alleviates the need to put much in terms of ceiling treatments. You'll probably still want some carpeting for the sake of damping standing waves, but your concerns are the side walls, the wall behind the speakers, and maybe the wall behind the listening area. There are two basic kinds of treatments (not including bass traps). Absorption kills reflections by, you got it, absorbing the energy. Diffusion kills reflections by sending them in all different directions at slightly different times. The general rule adopted by THX and others is that one usually places more absorption to the front of the room, and to the bottom, and let the waves reflect if not simply break up near the rear and top, though it's not a hard and fast rule.

Practically speaking, keep in mind that even well designed absorption products cannot absorb a wide range of frequencies evenly, particularly low frequencies, unless they become very thick. In addition, too much absorption makes the reverberation time unnaturally like an anechoic chamber, with the result being a "dead" room. On the other hand, just enough can bring a soundstage right into focus, and tame the brightness associated with an overly reverberant room. Diffusion products, if designed well, will randomize the wave so that it does not arrive as an organized front. To place the treatments to kill the reflection off the side wall, have a friend place a mirror along each of the walls, and make note of locations where you can see your speakers though it (the mirror). Put treatment at these locations.

For treating the rear wall, it will depend greatly on what you want to accomplish. If you want to use the rear wave as much as possible to enhance the depth and physical presence of the sonic image, you'll probably want to do very little on the rear wall, perhaps adding a small amount of treatment to tailor to taste. If, on the other hand, you'd rather take the studio monitor approach, and hear as little enhancement as possible, and keep the extra reverberant information to a minimum, you'll probably want to treat the wall behind the rear speakers heavily, either with absorption or diffusion. As always, experiment. The wall behind the listening area should be kept in mind for treatment, not only to limit possible confusing reflections from behind, but to dampen slap echoes, particularly if the wall behind the front speakers is left relatively untreated. After addressing the first reflection, look into slap echoes. Walk around the room, stopping to clap your hands. If you hear a long reverberation, you're probably between two untreated parallel surfaces. To fix the echo, you only need treat one of the surfaces, though it may be a good  idea to stagger between sides for the sake of symmetry. Again, choose between absorption and diffusion based on how you want it to affect the reverberant and tonal character of the room.

By now you should be good with the higher frequencies which are the most critical to imaging and much of our most sensitive range. Unfortunately, low bass frequencies are far more difficult to treat. If perchance you do get a great big collection of standing waves that results in an extreme null or peak at some location, your only real options are moving the speakers, moving really big pieces of furniture, moving yourself, or putting in a specifically tuned bass trap, usually placed in a corner. This requires a lot of experimentation, but that's part of the fun. Enjoy!

Q What should be considered in selecting a driver for a vented enclosure? Where does box frequency, driver Fs, driver Vas, and port tuning fit in? Coming from the viewpoint of a retrofit, would you go with higher qts, which should yield  lower Fb/F3? Once Fb is established, what guidelines do you use for port calculations, certainly as to the lower extension of the bass?

A The box itself, or rather the air inside it, does not have a frequency of resonance, but rather, a set of losses determined by the air and stuffing material, as well as a contribution to the suspension of the driver as far as the air provides a "spring" for the driver that combines with the mechanical  suspension of that driver. A vented enclosure can have a tuned port, whose resonance is determined by the area of the ports opening, the length of the port, and the volume of the enclosure. The longer the port, and the smaller the opening, the lower the tuned frequency. Or, the vent can be accomplished with a passive radiator, which uses an extra cone instead of a tube. The passive radiator is tuned by its mass, and how its surface area reacts with the enclosed air volume. The driver's individual Fs (free air resonance) is determined by the moving mass and the stiffness of the suspension. The looser the suspension, and the heavier the mass, the lower the Fs. Generally speaking, drivers for vented enclosures do not need as low an Fs for the same degree of extension in a given box size, and so can use lighter cones, and hence, achieve greater efficiency. Also, in order to combine with the output of the port in a controlled manner, they usually require a lower Qts (greater total damping of the driver) for a relatively flat response. And, because the tuned port loads the driver acoustically to yield the extra extension, minimizing acoustical output from the driver itself, they do not require as much linear excursion.

If you're going with a vented enclosure, you're will want to have a driver that allows a flat summation with the tuned port frequency, which is another variable you're going to have to figure out. A high Qts driver with a given Vas and Fs will yield a lower F3 than a low Qts driver with identical parameters in other regards in a sealed application, although it's unlikely that you'd run across such a case, as one parameter is usually the effect of something that profoundly affects the others. But, for instance, take a driver that has a larger magnet that increases the flux density in the gap. The increased magnetic field (assuming that the field is not wasted due to saturation of the iron) will do two things. One is increased efficiency. Above the roll-off point in a sealed system, you will get more output at a given input. However, it also lowers the Qes (increases electrical damping) by putting a braking force on the coil. It seems counterintuitive, but this decreases output near the area of roll-off, and hence raises the F3 (-3dB) point of the system.

If you really wanted to keep the F3 and utilize the bigger magnet, you could then add mass to the driver, lowering Fs. That, in turn, raises the Qms (lowers mechanical damping) and decreases efficiency. Buy some speaker design software, unless you want to spend days running a calculation that will probably have to be recalculated anyway. If you're set on doing it yourself, look into software that will allow running simulations of frequency response based on box and driver parameters. Take some published specs from available drivers and combine them with your box dimensions, and then play with the tuned frequency to see what you can get. If the tuned frequency is too low, you'll get a dip before the final rise (probably below the main line at it's peak.) If the tuned frequency is too high, you'll get a really nasty peak before the eventual roll off. AND, if the box is too big, you'll get a dip before the port output. OR if the box is too small, you'll have a hump above the ported frequency. Perhaps I should add, you could have a hump, dip, and a peak before the final plummet if the box was too small, and the ported frequency too low!

If you start to consider that the T/S parameters of a driver change under different output levels, it becomes a real headache. To do a truly well-done bass-reflex system requires a lot of effort if you're designing from scratch. It can be done, but starting from a preconceived box size will only make things more difficult. They're a lot trickier than a relatively simple sealed enclosure. Two places I'd start are http://www.madisound.com and http://www.parts-express.com. In comparison, if you took that sealed enclosure with a driver that was designed to have an F3 of 25 Hz and a Qtc of 0.85 in a given box size, and had a box slightly larger, you could maintain the F3 point until the box got large enough to lower the Qtc to 0.71, even though the Fs of the system would decrease. The effects of doing so would be a more gradual roll-off, more even phase response, but lower power handling. Or, if you had a smaller box than the "optimal" design, you could keep the same F3 until you had raised the Qtc to 1.0, even though the smaller box and raised Fs of the system. This, of course, would do the opposite, slightly compromising the phase response, but increasing output right above F3, while keeping the response fairly flat and even, and increasing power handling. It should be noted, I think, that the efficiency would remain the same.

Q I am confused about satellite hook-up options. I'd like to get a big-screen RPTV soon, and I'm trying to decide between 16:9 and 4:3 and between HDTV-ready and not ready. I have a DVD and DD audio system, and will get a satellite dish. My question is, if I get the new satellite receiver with the larger elliptical dish, what exactly do I get in the way of sound and picture? And what do I need to decode and display the signal? For instance, will I get HDTV quality 16:9 pictures with DD sound? If so, will this only be on anamorphic widescreen movies, or will all movies be this way? What about "regular" programs on satellite? We mostly watch sports and "regular" TV, but do enjoy a good DVD-DD movie now and then. Thanks for your advice. I think TVs are the hardest component to buy right now because of the changing technology!

A RCA currently has the DTC-100 which does Standard DSS and HDTV DSS. It also acts as an over the air HDTV set top box. Toshiba and Sony should be coming out with one shortly. There are several issues to deal with. If you use the DTC-100, you must have an HDTV set that is capable of accepting an RGB signal. This means no Toshiba TV because it wants a component signal. The way around this is to get an additional box called a transcoder. There is one coming out that will retail for $129.

Currently there are two HDTV channels on DSS. Channel 199 is a pay-per-view/demo channel, and 509 is HBO HDTV. All HDTV will be broadcast in the 16x9 format, but it is not anamorphic, because it is broadcast as native 16:9. If a show is in 4:3, they automatically place black bars on the of the image so it is not skewed. We have 4 local over the air (rooftop antenna) HDTV channels in my area, but only one really broadcasts true HDTV. And that is only an hour or two a day. The rest is a demo loop. Right now most HDTV broadcasts are in 2-channel Dolby Digital. This is because new equipment is needed from Dolby to make 5.1-channel work. For standard DSS, a couple of pay-per-views are actually in Dolby Digital.

If you want to get the best possible picture from DVD, you need a 16x9 TV. If you go for a DTV-ready set from Toshiba, then progressive DVD will take you to the next level in performance. Toshiba lets you change the aspect ratio when using the DTV/progressive input. Mitsubishi and Pioneer do not let you change the aspect ratio (on the DTV input) and lock you in FULL mode so you will get a distorted (stretched) picture on all Pan-and-Scan and non-anamorphic DVDs. FULL mode is the mode that HDTV uses as does anamorphic DVD. In a way they are tying to make things easier for people going to HDTV, they want to ensure you watch it correctly. At the same time, they are causing problems with DVD playback.

Q (1) Although my AR subwoofer actually plays quite deep, it just does not seem to "keep pace" with my Maggies (Magneplanar speakers). Is this something to do with the transient speed of the planar design? (2) Could you please recommend a few subs at various prices that might be a good fit for the Maggies? I believe I read that HSU Research subs are quite fast in their response. Any others to consider?

A (1) Contrary to the popular myth, planars, ribbons, and electrostatic speakers are not inherently faster than cone drivers. Speed, in terms of reaction time, is more a function of the bandwidth than the mass of the diaphragm. That's not to say that it's easy to blend a dipolar planar speaker with a boxed closed woofer. Dispersion patterns are different, and blending at the crossover point can be difficult to begin with, and only more so with vastly different radiation patterns. One thing to consider is that if a subwoofer is geared towards the "theater style" lower octave, with an in room response that is less natural but more impressive for most movie buffs. This can sound somewhat out of place with speakers that tend to have a leaner presentation.

(2)  HSU subs have a well-deserved reputation of performance for the money. JJ and SS both thought highly of them. M&K's MX-700, though not the all out pounder that many look for, may just be your ticket for matching the musical, controlled character of your Maggies. I'd also suggest looking into Mirage, Velodyne, and PSB as well, or even the Sunfire if you keep the crossover frequency WAY down and the relative output modest compared to the main speakers. Also, try moving the sub around the room a bit and adjusting the phase switch.

� Copyright 2000 Secrets of Home Theater & High Fidelity
Return to Q&A Index.