Richard Vandersteen of Vandersteen Audio: Part Two

by Doug Blackburn (db@soundstage.com)

Continued from Part One

dB: What’s different about time and phase performance when you use a concentric driver (i.e. tweeter mounted in the center of the woofer cone) like you did with your center-channel and surround loudspeakers?

RV: Concentric drivers are interesting -- you can make them time/phase correct everywhere except dead center on the axis line of the driver. So you don’t want our center-channel or surround speakers pointing directly at the listener’s ear(s). You want to be a little bit off-axis. This actually works extremely well for center-channel and surround speakers. With a center-channel, the loudspeaker is above or below the image which puts the center speaker above or below the viewer’s ears. With surround speakers, they are usually above and/or behind the listener. We think a full array of time/phase correct loudspeakers makes the best possible home-theater setup. Something happens that you just can’t get with typical loudspeakers. In fact, we are getting quite a bit of notice lately from the home-theater crowd because or sound is so much different than typical home-theater setups.

dB: How much of a change would you hear if you get outside the time/phase correct window for one of your loudspeakers?

RV: Let’s use a three-way as an example. At most listening positions, there’s a window about one foot high where you are aligned with the loudspeaker. If you are a little low, the top end will sound rolled off and the bass will be a little prominent but the sound will still be pretty linear, if tilted a bit. The worst case is standing so that you get the sound from the tweeter first. In that case you’d hear a 6dB to 8dB trough in the midrange and a 5dB to 6dB peak in the tweeter. That would be a very audible change. It’s also one of the reasons I hate to see people listening to our loudspeakers while they are standing up. You just can’t make any accurate observations about our sound until you are seated in that time/phase correct window. But this happens all the time at shows and at dealers. It’s something we have to live with to make the products we make.

dB: Speaking of auditioning loudspeakers, what can people hear or listen for in a time and phase correct loudspeaker that they won’t hear from typical loudspeakers?

RV: Well that’s a bit of a problem because there are only four or five brands of loudspeakers being made in the world that are truly time and phase correct. One thing you definitely want to be aware of is, if you are listening to a new loudspeaker and it just literally blows your socks off, it just really gets your attention and you come out of that demo saying something like "Wow, I’ve never heard anything like that before!" That should be your first indication that something is wrong. Loudspeaker sound/performance does not change that rapidly, at least not as things are in 1998. For some loudspeaker to be that different to an experienced listener compared to anything else they’ve ever heard -- somebody’s playing games. There are undoubtedly going to be some very attractive peaks and dips in that loudspeaker’s response in order to get it to be that noticeably different. A really good time and phase aligned loudspeaker will leave you very unimpressed with the loudspeaker, but very involved with it, very drawn into the music. The sound should sound less hi-fi (perhaps less audiophile) and more like live music. Even recordings that are not all that well done technically should sound at least entertaining and listenable. You should find yourself transported to the performance or find the performance transported into the listening room depending on the perspective of the recording. That’s the best way to "get" time and phase correct loudspeakers. Listen for the emotional connection with the music and stop trying to enumerate the Top 10 Audiophile Sound Characteristics (i.e. transparency, dynamics, bass, mids, highs, tonality, harmonics, etc.). Not that those things are unimportant, but if you never stop listening in that way, you’ll never notice the other side of loudspeaker performance. A realm of performance that is often completely ignored by dealers, reviewers and audiophiles.

The soundstaging of time and phase correct loudspeakers tends to be more natural also. You’ll get randomly spaced images of performers spread from wall to wall. Other loudspeakers tend to bunch the sound so that everything happens between the loudspeakers. This is often described as the speaker having a strong sense of focus. All this means is that the loudspeaker in question doesn’t have the ability to produce a panorama of sound across the room like a good time/phase correct loudspeaker. To one degree or another, the sound favors the center and doesn’t have the expansiveness that good time/phase correct loudspeakers have.

Q-Sound[TM] is another area where time/phase correct loudspeakers will do things that typical loudspeakers can’t do. Most typical loudspeakers will keep the time and phase manipulated Q-Sound effects in the front half of the listening room. Time/phase correct loudspeakers will put some of the effects in their correctly intended locations behind the listener and well off to the sides of the room behind the listener making a 360-degree sound field. Typical loudspeakers may crudely position some of these sounds behind the listener’s head, but they will not have the lateral specificity that they have when played back on time/phase correct loudspeakers.

dB: Why are all Vandersteen loudspeakers made with a single driver covering each frequency range?

RV: From 150Hz on up, single drivers are the only way I know to make a true time and phase correct loudspeaker. Below 150Hz, the wavelengths are long and the events in the sound are long, so using multiple drivers down that low is not a problem. From around 150Hz and higher you must have only one sound source. This is because it is impossible to get two exactly identical drivers to cover a frequency range and because they will be physically separated in a vertical array, it is impossible to get the distance from each of those drivers to the listener’s ear to be exactly the same. Because of those differences in driver response (even if matched to small levels) and different driver-ear distances, time smear is introduced. If you’re shopping for a time and phase correct loudspeaker, you don’t want to be paying big bucks for a design that undoes much of what is good about time and phase correct loudspeakers.

dB: Can panel loudspeakers be time and phase aligned?

RV: For the most part no. Panels tend to be large and the distances from the panel to the listener’s ear are different for each point you measure. A sound radiating from a panel has many points of origin on the panel and those all have different distances to travel to get to the listener’s ear. So panels introduce quite a bit of time smear for the most part. An exception is the Quad 63, which mimics a point source because all the sound radiates from the center of the panel. You can physically observe phase problems in panel speakers by shining a light on the panel and observing the reflection on it. The panel will shimmer. It’s obvious after seeing this shimmer that the panel is not operating as a true piston. It is a light membrane, but the field acting on the membrane is very weak compared to the field in the voice coil of a good dynamic driver. Because the magnetic field is weak, there isn’t a strong degree of control of the membrane. That shimmering that you can observe is what gives panel speakers the sound that some people like. It’s like having thousands of tiny low-level reverbs added to the music. They give a sense of space like reverb on recordings does. The best thing about panel speakers is that they have nothing that reflects the back wave back onto the membrane. The backs are open so the back wave can get away from the speaker without causing any secondary distortion. We try to mimic that with the enclosures we design for our drivers and by constructing our drivers so there are no surfaces behind the cone for sound to reflect off of and back onto the cone. This approach eliminates a lot of time smearing and amplitude problems.

We’ve been working on reducing the back-wave reflection and reflections off of driver parts for 20 years and have some patents on our methods. Now you are seeing some major players incorporating these features in some of their very expensive loudspeakers. I guess that means that more than one little loudspeaker company thinks it’s an issue.

dB: Why is it so hard to integrate a dynamic woofer with a panel loudspeaker design? I don’t think I’ve heard one yet that sounds right.

RV: It is a very difficult thing. For many years, there seemed to be some kind of "ideal" loudspeaker which had the positive characteristics of a panel speaker with the positive characteristics of the best dynamic woofers. The difficulty is that if the crossover is done anywhere near the midrange frequencies, say anything over 100Hz, your ears are very sensitive and regardless of how good a job you’re doing with your panel and your woofer separately, your ears will always pick-up on the fact that the sounds are being reproduced differently. Our ears are often not real accurate when we look at a specific point. But our ears do a heck of a job when they have to relate one sizeable chunk of the frequency spectrum to another large chunk. Maybe I’m overstating this a little. It’s something that we tune into over time. We might not notice it in a 15-minute demo or even in a whole day with such a loudspeaker. But given days or weeks, it is something our ears will pick out. I believe you can integrate a woofer with a panel, but only if you do it at very low frequencies -- 80 Hz or lower would be a good crossover point for that kind of design. But this means a huge panel with flat response to 40Hz in order to be able to mesh with the dynamic driver. When building to a price point, panel size is usually one of the first things to be reduced and as you do that, the crossover point has to go higher and higher till you get into the trouble zone at 100Hz and above.

dB: What are some of the things about the drivers used in Vandersteen loudspeakers that make them suitable for use in time and phase correct loudspeaker designs?

RV: Our drivers are our single greatest technical achievement. Nothing we can buy off the shelf can come close to what our drivers do for us. They literally make it possible to design and build loudspeakers with first-order crossovers without severe compromises. Dealing with the rear-wave energy is probably the most significant single technical feature of our drivers. There is just as much energy coming off the rear of the cone as comes off the front. What you do with that energy is critical. When you see a typical midrange driver, you see a structure that looks a lot like a cup with a few holes in it. All the energy reflected by the basket and magnet are reflected right back onto the cone giving a source of 100% distortion -- a lot of time smear and amplitude problems. This is a pretty severe distortion. It tends to make midrange notes harsher than they should be. Typical loudspeakers and drivers only attenuate this rear wave reflection 3dB to 6dB. If your midrange driver is cranking along at 85dB, most loudspeakers are hitting the back of the cone with 79dB to 82dB of reflected sound from the driver and/or enclosure. This is very significant and gets to be easy to hear when you are familiar with the sound of loudspeakers that do not have as much internal reflection. In the live versus recorded test it is especially and immediately obvious.

The average audiophile listening to loudspeakers is at somewhat of a disadvantage because they don’t have the original sound in the room for comparison. People tend to get used to the way loudspeakers sound and most loudspeakers have very similar levels of distortion (for like types of loudspeaker). This gives audiophiles the impression that a certain kind of sound is "correct" because the reference point shifted from live music to comparing one loudspeaker to another. When you hear a loudspeaker without all that rear wave distortion, it will sound like the notes are coming out of a blacker background. There’s less hash -- it’s less harsh. Some people will even perceive the lower distortion loudspeaker to be less transparent or a little sweeter sounding than it ought to be. If they can get past that initial opinion which came from comparisons with other loudspeakers with higher levels of distortion, and live with one of our loudspeakers for a while, they will find that the sound is more like what they experience in a live setting and less like the sound of a loudspeaker.

dB: Where did the transmission-line enclosures and separate enclosures for each driver come from?

RV: We use a terminated transmission line behind the tweeters, midranges and woofers. These aren’t completely open like a true transmission line. What we try to do it direct that rear-radiated energy away from the driver in as long a path as we can. When the sound reaches the end of the long chamber, which is not in a straight line, it has to pass back though the maze losing energy at each reflection point and through the damping material. We manage to attenuate the rear wave energy to the point that it is inaudible. That’s probably one of the hallmarks of our speaker designs. Separate enclosures, I think, is pretty obvious. You can optimize the enclosure for one specific driver and you can maximize the absorption of the back wave when you are dealing with a known frequency range within the separate enclosures.

dB: You have an interesting perspective on a long-time audiophile tweak -- removing the stock internal speaker wiring and replacing it with high performance audiophile cable/wire.

RV: A properly designed crossover is an all-encompassing kind of thing. Every detail matters when designing the crossover. This means that the enclosure and driver itself affect the crossover design. The interaction between drivers affects the crossover. The length and gauge and other electrical properties of the wire connecting the drivers to the crossover also affects the design of the crossover (if the designer is doing a thorough and detailed crossover design job -- a lot of them do not). To get all of these elements to do what you want to have happen at the listening position is an incredibly complex and time consuming task. Change the wire and you literally change the loudspeaker into something else very different -- at least this is true for Vandersteen loudspeakers. So an audiophile could go into a Vandersteen loudspeaker and install what we would all agree is a superior wire and end up with a loudspeaker that sounds worse. Not because the replacement wire was bad, but because we’d have to rework the crossover to incorporate that specific wire to overcome the other problems that were induced by changing the original wiring. And this wouldn’t be the kind of crossover reworking you could do with a computer program. This is the kind where you spend tens or hundreds of hours testing and experimenting to get the fine tuning of the crossover correct.

This wiring issue is one reason no Vandersteen loudspeaker has ever had one inch of "full-range wire" in it. "Full range wire" means wire that carries the audio signal from the amplifier -- the entire audio signal from lowest bass to highest highs. We avoid this by putting our loudspeaker terminals right on the back of the crossover. When you tighten down the terminals or insert the banana plugs, you are plugging your speaker cable directly into the crossover. Our speakers all have the terminals up about mid-height so that the wires to each driver can be of similar lengths and therefore, similar electrical characteristics. All of the wire in every Vandersteen loudspeaker is after the crossover to connect the driver to the crossover. This (enclosure, driver, wire, and crossover) becomes a network that has to be tuned. The crossover is where you do the tuning.

dB: Over the years there has been a list of audiophile complaints about Vandersteen loudspeakers. [slight chuckle] Let’s talk about some of these and put them in perspective.

RV: [more obvious chuckle] OKaaaayyyyyyyy.

dB: Let’s talk about appearance first. A Vandersteen loudspeaker made in 1998 looks an awful lot like a Vandersteen loudspeaker made in 1980. People like things that look new and different. There must be some pretty good reasons why the speakers look the way they look and why the appearance doesn’t change much over the years.

RV: The structure of our loudspeakers is an important part of why the speakers sound the way they do. The average loudspeaker sold today has 70% of the parts and labor cost in the cabinet and finish. Unfortunately, the prettier and shinier that cabinet is, the more it is going to interfere with the sound quality. You can’t get away from this -- it is pure and simple physics. Our cabinet is very functional. In effect, our cabinet is designed not to be there as far as the sound of the loudspeaker is concerned. We do everything we can to make the cabinet acoustically disappear. That’s what our boxless design is all about. Our cabinet and finishing represents about 17% of the parts and labor cost for our loudspeakers. The typical loudspeaker has 30% of the parts and labor cost in drivers and crossover. Vandersteen loudspeakers have 83% of the parts and labor cost in the drivers and crossovers. This is why we have achieved a reputation for building products that are incredible values for the money spent, even among critical audiophiles. Our enclosure isn’t very pretty to look at without the grille cloth in place. But that construction method allows us to create amazingly high-tech cabinets without having to worry about making them pretty which increases the cost of the loudspeaker beyond what I think is reasonable.

There is a growing number of narrow loudspeakers. Some people wish we’d do a narrow design. We just can’t figure out how to do that without doubling the cost of the speaker or compromising the sound quality and I won’t do either of those. One of my goals for Vandersteen products is to give people the most performance I can for the money. I’m just very uncomfortable putting too much money into the enclosure. The Model 5s are an exception. From 1977 when we started we have always promised Vandersteen dealers and customers that we would not spend one penny on the looks of our loudspeakers if it didn’t enhance the performance of the loudspeakers. The Model 5 is different because it is so expensive to manufacture that it sells for a price high enough to leave some room to do some things appearance-wise. So when we did the design of the 5s, we put in everything we knew to do in the design of a loudspeaker. We optimized its performance in every way that we could think of. Once we had the technology package, then we put some money into the appearance of the loudspeaker. But even with the Model 5, appearance is the lowest-priority element of the product.

dB: Another perennial comment about Vandersteen loudspeakers you hear repeated over and over again from certain quarters is that they sound dark and slow.

RV: [chuckles at the "dark and slow" phrase] Let’s touch on the "dark" aspect first. Our top-end response is flat all the way out to 20kHz or 30kHz depending on the speaker model. We have single drivers for each frequency range. They are time and phase correct. We work very hard to make the bass and midrange response as flat as possible. We do not put intentional peaks in our tweeter response. Yeah, I understand why that [dark and slow] comment gets made a lot. In a dealer showroom, there are going to be other speakers with tweeters playing 3dB to 6dB louder than ours. So, when you put us up against those kinds of loudspeakers, yes, we sound a little dark, a little laid back. But when you compare us to real music, we’re right there. Those peaked tweeters are completely intentional and are designed to cause the customer to spend money on the speaker with the peaked tweeter rather than on the more natural sounding loudspeaker. I could fight back and put 3dB to 5dB more into our tweeters, but I think it is the wrong thing to do and I’m going to keep building loudspeakers as honestly as I can. Our loudspeaker with flat response does not sound very flamboyant next to an intentionally manipulated loudspeaker.

There are all kinds of other potential problems in dealer environments -- other speakers in the room, inappropriate cables (perhaps better suited to other loudspeakers with peaked tweeters), inappropriate electronics, poor setup, incorrect connections. We try to train our dealers well, but things happen. Even more to the point, the dealer’s show room is often not an ideal listening environment due to the realities of it being a show room and not a listening room. Most of our dealers do a heck of a good job with their demos given what they have to work with which is more limiting than a home listening room. It isn’t that hard to hear a Vandersteen demo where our loudspeakers sound a bit boring next to other loudspeakers.

But if you optimize a system for Vandersteen loudspeakers and put them in a room by themselves you get a listening experience that is very different. People who know Vandersteens have always felt that way and have disagreed with the dark/slow comment that keeps being repeated. People who make the dark/slow comment, I think, have never heard a pair properly set up in the home environment where they were designed to function correctly. We never designed a loudspeaker to sound good at a dealer show room. I don’t want loudspeakers staying in dealer show rooms, I want them in peoples’ listening rooms -- and I want them to get there without playing tricks on people.

What you don’t hear much comment about from the same group that keeps repeating that dark/slow criticism is the number of $1200 Vandersteens out there being driven by $5000 to $10,000 worth of electronics. This isn’t accidental. It would never work if the loudspeaker wasn’t giving performance well beyond its price. This is something the dark/slow skeptics can’t understand. This is a direct result of our strategy for our products.

• Continue to Part Three

• Current Issue Audio Online