Interview with Aperion Audio on Loudspeaker Design Philosophy
In preparation for our up and coming article that discusses what's in a truly high end speaker and the common cost cutting approaches some companies often make to deliver a high value product, we sat down with the well established and revered brand Aperion Audio to get their views. Ken Humphreys (Loudspeaker Design Engineer) was kind enough to provide his answers into a short and informative editorial below.
I’ve got to confess – I had to wrestle a little with “What makes a High End Speaker?” Would a modest looking, low cost speaker with impressively convincing true-to-life sound be a “high-end speaker”? How about a beautiful and expensive speaker sold in high-end salons, even if it has rather pedestrian sonic performance? Or the speaker that allows an audiophile to check-off trendy features – maybe it even looks odd in order to tell some interesting performance story – but it sounds, well, “spectacular”?
So, what is a high-end speaker? Clearly you first need to know what you’re trying to accomplish. Ferrari and John Deere, needless to say, will have different ideas of what a high performance vehicle is. I design speakers intended for 2-channel to surround-sound use in a home environment so, for me, a high-end speaker, first and foremost, is about honest, “true-to life” sound. Since this is what I do (and have been doing for more years than I would have thought possible) I’ll speak to that. At least to two of the performance criteria that I believe translate to “high performance” and to what can be done to help get there.
By far, the most important thing to get right is the tonal balance. This is now almost universally accepted – but not so much so regarding how best to achieve it. Here’s how I try:
- Measure, measure, measure. I move the mic around a lot in order to know what’s happening throughout the front hemisphere – lots of measurements including some to learn the power response. For frequency response measurements I prefer those made with the “ground plane” technique. Speaker and mic are placed on a large, reflective surface like, say, a warehouse floor after the fork lift is out of commission. These measurements are better and cheaper than measurements made in an anechoic chamber. If the nearest wall is 5 meters away measurements down to <40Hz are reliable – lower than any anechoic chamber I know of.
- Listen, listen, listen. Find a few willing souls whose hearing has proven consistent over time and agrees with the objective measurements. I find that a listening room that allows me to conveniently, and in real time, switch between listening, objective measurements and adjusting crossover parts is a must. And I’ve become appreciative of “blind listening” where pre-conceived notions won’t color judgment.
- It sure is easier if you start with well-behaved drivers. It’s great if you can just pick drives off the shelf but often, for one reason or another, custom drivers might be needed. In which case I look for a driver with a non-resonant (or a pistonic) cone through its operating range and a surround that absorbs energy rather than reflects it back to the voice coil. A decent voice coil overhang will cost a little in efficiency but helps lower distortion by keeping things linear through the expected excursion range. For drivers with dust caps, vented pole pieces help with power handling and can relieve pressures behind the spider if the frame isn’t vented there. Extended pole pieces help keep a uniform magnetic distribution around the voice coil. Needless to say there’s much more but the important thing is that many of the engineering choices that translate to good driver performance needn’t cost much, if anything. It’s just that someone who knew what they were doing had to make some right choices.
- Crossovers, I think, get a lot of attention because of their “black box” mysteriousness. Being hidden from view, some audiophiles want to be assured that high quality parts are used. Parts like air-core inductors and polypropylene capacitors. Personally, I’ll use air core inductors if their values are small but have no compunction about using good, hard-to-saturate laminated-iron core inductors. They have lower DC resistances and cost less – I’ll use them wherever they make sense. I like poly caps for the smaller values where they make more of a difference. But with larger values, poly caps show very little difference over good electrolytics in parallel with a small poly caps and they cost so much they steal dollars from where they could be put to better use.
- I want to minimize delayed arrivals from the speaker. This family of culprits includes undamped driver & cabinet resonances, reflections off the grill frame, and more. Although this topic is probably a little to large to cover here, cabinet panel resonances often receive special attention. Cabinet bracing (or using smaller cabinet panels) can successfully move panel resonances up to a less objectionable range or even above the crossover frequency, freeing them from being excited by the woofer altogether. Care must be taken to ensure the cabinet is properly damped from unwanted resonances which can color the sound quality and reduce the designer's efforts of making a truly accurate speaker system.
The left pic above shows the brace directly below the mid chamber in our Verus Grand Tower. This mid chamber, of course, serves to further brace the upper half of the cabinet. You can also see the two long top-to-bottom blocks that reinforce the back-to side panel joint. And lots of glue. The right pic above shows the brace below the woofers. Also notice the pictured woofer has a vented pole piece to increase power handling.
Power Response Balance
Natural sound sources have directional properties of their own. One obvious example is that it’s easier to understand a human voice when you’re in front of someone. In a room this results in the sum of the reflections – the delayed arrivals which mainly come the from “off-axis” energy – having a slow, smooth roll-off as the frequencies get higher. Speakers similarly lose high frequency energy off axis ……..until they transition over to the tweeter where they turn back on before rolling off again at the highest frequencies. This colors the ambient sound even if the axial frequency response is good. For me, getting a smooth off-axis roll-off is an important objective. This is mainly accomplished simply by using lower crossover frequencies so that the transition to the tweeter happens before the midrange gets too directional and where the longer wavelengths through the crossover region reduce phase interference. Using dome (or ring) tweeters at these lower frequencies, however, can be tricky.
Pictured above is the Aperion Audio Verus ASR tweeter rear view showing both pole piece vent and behind-the-voice-coil vents. Also shown is the dampening at vent opening to suppress turbulence. There is a lot of engineering involved in really well designed tweeters that you just can't see looking at the faceplate view or spec sheet boasting exotic dome material usage.
Two problems need to be addressed:
- The first order of business is to lower the tweeter’s resonant frequency to well below the crossover frequency. Increasing the mass is probably not a good option but fortunately, increasing the suspension compliance will also lower the resonant frequency. Making the annular (ring) suspension large & loose and enlarging the air chamber behind the dome (by venting into a rear chamber) effectively does this.
- The second, more vexing problem is that dome (and ring) drivers tend to “rock” when energized around their resonant frequency because the suspension is all in a single plane. We’ve developed a new tweeter design (patent pending) that restricts this rocking, allowing it to operate much closer to its resonant frequency than a conventional tweeter.
I hope that touching on how one speaker designer approaches just a couple aspects of “high-end” performance proves interesting to some. But I’ve come to respect that it’s just one way. Many speaker engineers, with other priorities, methods and objectives have designed plenty of great speakers. One point, though, that I hope comes through is that certain features and design elements often associated with “high-end” speakers can help – if they serve a purpose well – or may be of no use at all. As always, it still comes down to the sound.
My two cents worth,
Many thanks to Ken Humphreys of Aperion Audio for taking time to discuss the topic of loudspeaker design.