Published On: May 16, 2016

First-Order Crossovers: Panacea or Problem?

Published On: May 16, 2016
Last Updated on: October 31, 2020
We May Earn From Purchases Via Links

First-Order Crossovers: Panacea or Problem?

Does the use of a first-order crossover really make a better speaker? Brent Butterworth explores the topic in this week's featured news story.

First-Order Crossovers: Panacea or Problem?

By Author: Brent Butterworth

Brent has been a professional audio journalist since 1989, and has reviewed thousands of audio products over the years. He has served as editor-in-chief of Home Theater and Home Entertainment magazines, contributing technical editor for Sound & Vision magazine, senior editor of Video magazine, and reviews editor of Windows Sources magazine, and he also worked as marketing director for Dolby Laboratories. He's now on staff at Wirecutter.

Thiel-CS17-thumb.pngI've always encouraged critical e-mails from readers because it's good to have the feedback. But a couple of the e-mails in response to my review of the THIEL TT1 got downright nasty. "You are not an audiophile!" one raged in response to my comment that the TT1 is "...a more versatile speaker than any Jim Thiel [the late company cofounder] designed and probably a better value than anything Jim designed..."

I was reminded of this when I recently Googled some information for my upcoming review of THIEL's new SmartSub 1.12 subwoofer. I encountered impassioned complaints about the company's decision to move away from the phase-coherent designs and first-order crossovers that were a hallmark of Jim Thiel's work. In a YouTube video, an audio enthusiast praised Jim Thiel's designs as "light years ahead of what everyone else had to offer" and took the company to task for offering new product that "appears to be the same old crap that we've seen in the loudspeaker world for the past 30, 40, or 50 years."

From a business standpoint, whether or not it was wise for THIEL Audio to abandon Jim Thiel's core design concepts is debatable, but I expect most speaker designers would have made the same decision as the new people at THIEL. That's not because they're dumb. It's not because what they make is "crap" (I don't give out five-star performance ratings to crap speakers). It's not because they're "not audiophiles." It's because they understand that first-order crossovers introduce a surprising number of trade-offs, a point that none of the commenters I've seen lambasting THIEL Audio seems to have bothered to investigate.

I'm fortunate to have spent many hours discussing these issues with Jim Thiel himself. I visited him twice at his product development lab and factory, and he dropped by my workplace a couple of times to help me set up speakers or to chat about what he had in the works. It may be surprising to some that his approach was deeply rooted in science and no-nonsense engineering. To my knowledge, he never indulged in the mystical verbiage and hard-to-support performance claims that are common in high-end audio, and in our discussions he was often dismissive of faddish audio practices that weren't firmly supported by scientific evidence.

First, some basics for those who need them. The crossover is an electrical network that divides the sound into bass for the woofer and treble for the tweeter (and often midrange for a midrange driver). A two-way crossover has two filters: one that filters the treble out of the woofer and one that filters the bass out of the tweeter. (A three-way speaker adds filters that remove the deep bass and upper treble signals from the midrange driver.) These filters are characterized by the frequency at which they begin to attenuate a signal, and by the slope of that attenuation. A first-order filter attenuates at -6 dB per octave, a second-order filter at -12 dB per octave, and so on. These filters affect the phase of an audio signal, slightly delaying some frequencies relative to others.

The benefit of a first-order crossover is that it maintains the phase of the original signal, as long as the speaker that it's used in has the acoustic centers of the drivers aligned in the vertical plane, often through the use of a sloping front baffle. As long as your head is positioned so that your ears are equidistant from all the drivers, the phase of the direct sound that first reaches your ears shouldn't vary by more than a few degrees. That's why THIEL and other manufacturers refer to this as a phase-coherent design.

How does this affect the sound you hear? I've reviewed perhaps 10 THIEL speakers since the early 1990s, as well as several other phase-coherent designs. To my ears, the benefit of phase-coherent design is that you get a more enveloping soundstage, more like what you'd expect to hear from a panel speaker such as a MartinLogan or Magnepan--but with more precise imaging than a dipolar panel speaker can produce. It's a great sound, one I strongly recommend every audio enthusiast check out for themselves in a well-designed dealer showroom or at an audio show.

As nice as the phase-coherent sound can be, it's not "light years ahead" of everything else. If the benefits of phase-coherent design were so great, more speaker companies would use it, because in its basic form it's also the least-expensive crossover you can build. For a two-way speaker, a first-order crossover requires just one capacitor, one inductor, and one resistor. For every added order in a passive filter, you need an additional capacitor or inductor. That's why the most common use of first-order crossovers isn't in audiophile speakers, but in soundbars, wireless speakers, and home-theater-in-a-box systems. (For the record, Jim Thiel's crossovers were amazingly complex, with numerous networks added to correct impedance and frequency response anomalies, but the core functionality was still provided by those three basic components.)

So what's the downside of first-order crossovers? There are three: distortion, driver longevity, and dispersion.

As Jim Thiel told me, because a first-order filter doesn't attenuate as abruptly as high-order filters, a driver used with a first-order crossover has to exhibit good performance two octaves beyond the specified crossover point--i.e., in a two-way speaker with a 2.3-kHz crossover point, a tweeter must be able to handle signals as low as 575 Hz, and a woofer must be able to handle signals as high as 9.2 kHz. This is especially hard on a tweeter, producing distortion at high volume and sometimes leading to driver failure (early Thiel speakers were notorious for blowing tweeters). It may also excite the woofer's "breakup modes," or distortion-producing high-frequency resonances.

Jim Thiel fully acknowledged these problems, and he largely solved them by developing tweeters with an unusually long excursion of about a quarter inch, and through the use of a unique corrugated-diaphragm concentric driver that handled the mids and highs in his more recent models. These problems can also be lessened by going to three- or four-way designs.

The other problem with first-order crossovers is dispersion. Because the filters are relatively shallow, both drivers are simultaneously audible when sounds are within about an octave of the crossover frequency. This is no problem if your ears are at the same distance from the drivers. However, if your ears are closer to one driver--if you stand up, which puts your ears closer to the tweeter, or slouch down, putting your ears closer to the woofer--the sounds from the two drivers will no longer be in phase at all frequencies. They'll be in phase at some frequencies and out of phase at others, which means some frequencies will be boosted and others attenuated, and you will no longer be getting even frequency response. (Speakers with higher-order crossovers can also have this problem, but to a much lesser degree.)

Again, Jim Thiel took great pains to fix this problem; it's why so many of his designs used a coaxial arrangement, with the tweeter inside a midrange driver or woofer. But the exotic driver designs he relied on took years to develop, cost a lot to manufacture, and sometimes prevented him from getting new models out as quickly as his business partners and dealers might have liked.

Thiel-TT1-thumb.jpgI was fortunate to review both the very last speaker Jim Thiel designed (the CS1.7, shown above) and the very first speaker the company designed after Jim passed away (the TT1, shown right). The difference was dramatic. The CS1.7, like past THIEL tower speakers I'd reviewed, was somewhat fussy to position and couldn't touch the bass response and dynamics that many similarly priced speakers offer; it was spectacularly great for purist audiophile recordings of acoustic instruments, but a questionable choice for heavier rock and pop music or for home theater. Meanwhile, the TT1 delivered loads of dynamics, wasn't fussy about placement, and ultimately delivered a more neutral sound that worked with any kind of music.

Of course, fans of the old THIEL speakers may insist there's nothing special about the new speakers, that they're not too different from what B&W, PSB, and Revel produce. That's true. And of course, many audiophiles prefer a product with charming idiosyncrasies to one that delivers objectively superior performance. That's fine. And of course, some may now gravitate away from THIEL and toward Vandersteen, another company known for phase-coherent designs. That's fine, too.

What's not fine is claiming that the phase-coherent speaker (or, for that matter, almost any other audio technology) is "light years ahead of what everyone else has to offer." Just like every other audio product ever made, phase-coherent speakers represent a series of trade-offs. By understanding those trade-offs, we can make intelligent buying decisions. We cannot make intelligent buying decisions if we base those decisions on a quasi-religious reverence for specific technologies, brands, or personalities.

Additional Resources
The Pros and Cons of Multiple Subwoofers at HomeTheaterReview.com.
How to Choose a Subwoofer for Surround Sound or Stereo at HomeTheaterReview.com.
THIEL TT1 Tower Speaker Reviewed at HomeTheaterReview.com.

  • Linked Devices
    2023-08-24 14:30:43

    Phase coherence is what separates stereo that sounds like 2D from stereo that sounds like it ought to sound I. E. 3D spatial when you turn them up to their volume sweet spot and the reflections from the rear and all around you are about equal just like irl when you listen to say, a man singing in a tunnel. Phqse coherence is the difference between the sweet spot where his voice sounds full with the resonance of the tunnel adding a natural reverb to his voice. Wouldn't you agree that there are optimal places to set up a recording vs going back too far and hearing what sounds more like an echo vs too close where you don't get as much of the reverb in the mix. You don't even need electronics to create a very clean crossover. The KEF htb/2/w is still sold to this day like 11 years after it was first produced now only relying on what they'd perfected with their clacker ball physics where the sealed sub has a single active driver in the center sandwiched against a flat diaphragm. The cone behind the diaphragm is vented and tuned to prevent ring distortion and led have been making the same style LF drivers ever since and the htb/2/w was a successful prototype to now the Blade would hand such a shallow space where the diaphragm is larger than the subwoofer is thick and at about 30 lbs the natural helmholtz resonance of the sealed htb/2/w is 54hz where output from both sides is equal. The driver used the same concept as the XQ / IQ series and is what made the iq7se better than the iq7 going from a 3 way to a 2.5 way. This was already tried and worked excellent with the IQ90 and the XQ40 where the bottom two chambers are roughly proportioned in the ratio of phi from each sealed chamber from the top to the middle to the bottom. The middle and the lower woofer are electronically parallel but the box design allowed the bottom woofer to fall out of sync and lag about 300ms matching the front port above it while capable of playing lower about 8 extra HZ and sounding as if it were a much deeper box. The LF woofer also had a port so created a fractal like sound which measured as if the bass were bumped up too high but in practice once you turned them up the reflections from the port allowed for the middle woofer to sound more like a mid delivering cleaner punchier bass while the bottom woofer was more spacial due to the delay which imitated now bass actually sounds in a live venue once the ports flooded your walls with what sounded like what you'd hear all around you in a venue while the direct firing radiators from the Uni Q to the woofer to the woofer that was mechanically set to imitate the depth of the 207/2s and you can clearly hear the layers separated rather than sum as a sweep would misinterpret. The fractal sound from the front facing ports each being phi from the top to the middle as the middle is to the bottom made the layering very obvious and gave the speaker this 3D bloom from the bottom up while the bass rather than being merely bumped up by its port was allowed to play much broader a range due to the elliptical ports which carried forward to the LS50s. The speakers are 200 watts are are meant to be played loud while being toed straight and had a very interesting build with a crecrent top hood made of what appears to be copper (it patinas green) and is a pretty massive hunk of metal meant to dampen and keep the Uni Qs from causing the top to wobble as of it were suspended from the air by the top of the Uni Q. The speaker matches the loudness curve naturally and are perfect for rock and especially grunge, metal., dream pop, jazz, or anything that relied on a very tuneful bass section. The way the mid woofer and the top port played as if they were one have rise to this bloom that doesn't sound like just more bass but instead more space while the bass was given extra definition since you could hear it begin with the middle woofer direct, it's resonance in the room reinforced by the bottom woofer which made kick drums sound like they were literally coming from behind the walls up on stage. The IQ7ses played the mid woofer and the LF woofer in parallel but due to the smaller chamber the mid woofer naturally slopes off seamlessly where the bottom (the only LF woofer on the IQ7ses) worked with the mids to give you pinch tight fully saturated bass while the single bowed front and back and elliptical port allowed for this really interesting layered sound that seems almost like the speakers aren't just lute shaped but we're designed for guitar or any genera and you could turn up disintegration or mogwai and always hear more than you ever thought you had. They also work really well for movies since the atmospheric bloom also lent itself to jump scares etc and explosions etc sounded 3D. Going back to the htb/2/ws they're much better than the slimline T2 sub which is exactly the same thing but minus the toroidal design, and single point 2 way LF section. The thing is sold as a subwoofer but is not like any subwoofer one ever seen and I'm pretty sure they were designed to test and perfect what was needed for the Blade and work on tandem with the 3001SEs and the LS50s that came about from the 3001SEs (the waveguide with its sealed suspension tech, its AMT like rear fins that soaked up any residual resonance from the woofer if any could be left since like the Blade the 3001SEs also feature the oversized 2 inch or 2.5 inch? I forget that value exactly, they only dropped info on the 300X and not the 300x SEs so if have to measure it directly but the voice coil nonetheless straddles the compound multi function woofer suspending the woofer by the about 2/5s in from the inner rim where the voice coil is usually situated. There's aor of trickery involved in their speakers and since the XQs they basically rebooted their tech by reverse engineering it and basing it now around the tangerine waveguide which works so well you can cut a gold ball on half or any lid or anything that'd fit and cover the tweeter from the front and you won't even notice any difference. You never hear the sound from the tweeter directly, it's either sprawled then converged with the woofer or Z flex, or from an angle you hear the tweeter and the woofer already converged from the flare. I believe they kept the sealed suspension off both because it would be something they can add on later and then attribute the obvious gains you'd get from no longer having a gasping gap between the tweeter and the woofer which caused the speaker to spine both coagulatied and shouty and grating or "too bright", when not loud enough they don't come to life. The LS50s are an attempt to reboot the ls3 5as but with better newer tech just like 3001SEs which all share the same exact geometry down to the parobola of the woofer / waveguide to the horn that doubles as the inside gasket. Both have a crossover at 2.2khz and a port tuned to 54hz, both use internal moving parts where the same fins that clear the resonance from the woofer alone work as a full range AMT when played backward. If I were to cover the front and remove the tweeter then flip the woofer you'd think the speaker had its tweeter and was fine. The rear fins are so tuned to the tweeter they ghost it when missing. The 3001SEs are the last to play all the way up to 55khz and the 3001SEs removed a tweeter that effectively bypassed 2/3s of the crossover the 3001SEs used and the only reason why they're not noted even though KEF lists them literally twice as gen 7 and gen 8 in their tech museum implying the 3001SEs was immaculately conceived apparently. Lol. But the 54 HZ crossover of the HTB 2 where the high frequency side goes from 54hz to 300hz direct while the other side plays from 54hz down to 30hz before rolling off as a sealed sub does. KEF's total system design seems to be all about reducing the need for electronic components everywhere they could overlooking their active speakers which have built on limitations (the LSx snuck in the sealed suspension system and never documented it, but also has a brick wall filter that drops like a rock at like under 22khz and the immersion studies repeated across the world demonstrate that we can and are more immersed in music that plays up to 38khz and subjects who didn't even know it showed drops in immersion as measured in an immediate drop in alpha and then theta waves. So complex crossovers break down over time. Reach capacitor is literally a battery and the more complex the crossover the farther back you are from the man singing on the tunnel. KEF's use of their elliptical ports to sprawl sound as if from a perfect chamber masking the imperfections of your room I'm pretty positive caused the LS50s to leave the gap open between the woofer and the tweeter so people would play it at the sweet spot where the elliptical flexiport would do the same as the array of voigt pipes essentially basically forcing you to play them within the optimal volume sweet spot where it'd guide you to layer the sound from the port in proportion with the direct firing drivers. The 3001SEs with their sealed suspension tech and not clogged with active clutter and above all that capable of playing 1 to 1 with content that goes up to 50khz with an extra 5khz headroom is such a sealed vessel you can carry water in its crescent contoured front. It's air tight. And they managed to sound more holographic and larger than the LS50s, while the LS50s do have more bass they don't delivery the holophony 3d sound scape due to how the port on the LS50 can't keep up and blast enough air while the metal and rubber pressure vessel with its AMT rear mid woofer plays at any volume, the louder the better just like the 200 watt iq90s/xq40s right before it. 200 watts is a lot and that is a huge suggestion to play them loud and you'll be rewarded if you do. Any amp will do at that point. Otherwise people usually claimed a warmer amp is preferred for more bass control since the layers of bass have to be perfectly phase preserved.

Subscribe To Home Theater Review

Get the latest weekly home theater news, sweepstakes and special offers delivered right to your inbox
Email Subscribe
© JRW Publishing Company, 2023
As an Amazon Associate we may earn from qualifying purchases.

magnifiercross
linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram
Share to...