Interview with NHT On Loudspeaker Design Philosophy
As long time fans of NHT (Now Hear This) Speakers, I wanted to get a behind the scenes glimpse of the reasons for their success for building no-nonsense speakers for over 25 years. We sat down with Chris Byrne, Co Founder/Owner, and Jay Doherty, Chief Engineer of NHT to get the scoop on the design philosophy behind their products.
AH: What are the primary design goals for NHT loudspeakers?
Chris: Our designs are guided by the study of psychoacoustics, how we perceive sound provides clues to making better speakers. This philosophy began with Ken Kantor the other Co-Founder of NHT, an MIT grad and the inventor of the AR Magic Speaker. NHT designs target applications, not price points or good, better, best and attempt to maximize performance for the intended location. Further, we do not try to make a speaker do more than it is capable of acoustically. The Super Zero is a perfect example. Rather than suck out midrange for the appearance of bass, we freely admit the Zero has no bass output below 100Hz. But above 100Hz, we would put it up against any loudspeaker regardless of cost.
Like many of the early East Coast speaker companies, NHT firmly believes in acoustic suspension designs, as they are, in our opinion more musical and offer tighter response. The bass modules built into our towers and subwoofers originally were vented, however over the last 10 years we have converted every model to acoustic suspension, with only one model left to convert (woofer section of the Classic Four).
We like to think we take an honest approach in design. For instance one of our rules is to not include parts because they are expensive or offer very tight tolerances, or because the audiophile community raves about them. Unless we can measure an improvement in performance, we don’t spend money where it is not needed and keep the price fair to the consumer. While a small company, our people come from large companies, we understand logistics and having parts made in mass scale with consistency, this is a primary reason our products are more affordable than many others.
AH: What drivers and crossover components do you use for NHT speakers?
Chris: Virtually all our woofers are designed by and for NHT. In particular we feel the midrange is the most critical frequency area as it is where the human ear is most sensitive. You can compare speakers designed in the early 90’s and those designed today and you will find almost identical sonic signatures. We’ve gotten better at some things, but fundamentally we’ve not varied from the philosophy we began with. We have modified some tweeters for our purposes and used some off shelf if they happened to work for our design intention. An important element in our process is to design as much of a driver’s performance parameters into the driver itself not the network. Our network designs are kept as simple as possible believing that the less that is put in the signal path, the better the sonic result. We use neo magnets at times when it solves problems; and have done inverted magnets on large woofers for the same reason. Find the most elegant, most affordable solution for the intended application.
Jay: NHT uses a fair amount of driver design software in the early stages of development – some commercial, some proprietary. We also invested in a Klippel system for non-linear analysis. We have found no correlation between expense of crossover parts and sound quality of the final design. That’s not to say that we don’t care about part quality, but spending a tremendous sum on supposedly better parts does not yield better sound quality.
NHT Classic Two Cut-Away - notice the thick and multiple braces and aluminum bar heatsink/brace combo
AH: Please discuss cabinet
construction for your speakers.
Chris: We have almost always used MDF, because of its integrity and its propensity for precise cutting. We use braces or internal laminates to be sure there are not spurious resonances caused by weak surfaces. We sometimes use drivers and other metal parts to help strengthen cabinets, while assisting with other sonic attributes.
Examples: The Absolute Zero, Classic Two, A. Tower and Wall speakers which all employ Neodymium magnets on their tweeter to keep it as close as possible to the mid-woofer for sonic coherence. The tweeter also contributes to rigidity. The problem with Neo is that its properties change with heat. Our solution was to bolt a 3/4” diameter aluminum bar to the back of the tweeter that runs all the way to the back of the cabinet and is bolted there. This solved both issues. The tweeter bar acts as a heat sink keeping response consistent during use and a cabinet brace. Obviously the body of the tweeter has to be designed appropriately.
AH: What measurements do you perform on your speakers and why?
Jay: This could lead to a very long discussion. Here’s a very brief version.
- Impedance measurements - using a constant current source developed by NHT. Used to check driver function and driver matching to the cabinet.
- Frequency response measurements – many different types, generally using some variety of synthetic noise (MLS or pseudo pink) as stimulus, though we also use sine sweeps when appropriate. Used to check all sorts of things – on- and off-axis frequency response, driver compression at a high levels, room interaction, grille effects, crossover output when loaded by the drivers, etc.
- Distortion measurements - There are a couple of different approaches using a slowly swept stimulus (sine waves) or a quickly applied one (multi tone) that give different sorts of information. Useful for making bandwidth evaluation for drivers (especially tweeters) and verifying overall system function.
- Panel vibration measurements for cabinets (with an accelerometer). Useful for bracing evaluation and checking production quality.
As mentioned above, we have a Klippel system for non-linear analysis of drivers. We also do some electronic development. We use many of the measurements above to analyze the behavior of electronics, and some that are specific to the electronics (FFT for hum level, safety checks, long term power testing).
AH: Do you feel an anechoic chamber necessary to design and measure speakers?
Jay: Using semi-anechoic measurement techniques (as long as you understand their limitations) means that anechoic chambers really are not necessary for system design (though, as above, they are very handy for driver QC). Measurements made in a chamber may provide a starting point, but final evaluation of a speaker has to be done in the environment in which it will be used. No one listens to speakers in an anechoic chamber (not for long, anyway). The loudspeaker will always interact with its environment, so the design must reflect that interaction. Factoring in the interaction isn’t always easy to do, but that’s why experience matters (as well as the availability of good measurement equipment, good ears and good judgment on the designer’s part). At some point a purely analytical system for loudspeaker design may exist, but for now there are lots of blanks to be filled in by the designer; virtually every aspect of system design involves choices and compromise.
AH: Why is the sensitivity rating on your speakers often lower than many of your competitors?
Chris: We have never found sensitivity to be an issue. Power is cheap. Acoustic suspension design has many benefits however the price for those benefits is often a little bit of sensitivity. Remember a 3dB difference is basically a click or two of the volume control. It only becomes a problem when a customer buys an inferior amplifier design with no headroom, or wants to play the system at near live performance levels. Most of our speakers will not play as loud as speakers with sensitivity ratings well into the 90dB range, although we have made several systems that could reach 120dB SPL. We do not perceive efficiency as a market need.
Jay: Remember that there is no standardized way to measure sensitivity. Certainly, sensitivity measurements are not comparable between different manufacturers. If we say 86dB, another company might list 90dB for a comparable design. Since you don’t know how the measurements were made, there is no way to compare the two numbers. At best, the sensitivity figures are useful only for comparison of the models within a brand.
For most people, it’s not an important measurement. Adequate power isn’t hard to come by, and most companies will help you to match a system to your room size and output level requirements.
NHT Classic Three (left pic) and Absolute Tower (right pic)
AH: What do you feel makes your products stand out from your competition?
Chris: We make products that we want to own ourselves.
We design for accurate reproduction, it is not our place to editorialize on the decisions made in the recording studio, only to reproduce it as the artist intended, good or bad. We offer very fair pricing for what we think is outstanding performance.
We build what we believe the market needs, not what we think we can sell.
- Example 1: Do you see any sound bars in our line? The answer is no, we don’t think the world needs another; it’s a convenience product, not performance.
- Example 2: Look at our in ceiling technology, a clear example of solving a dispersion problem with a 3 tweeter array.
We spend time making sure that you can listen to our speakers for long periods without fatigue.