Horizontal Menus in CSS by Css3Menu.com



Q. Fast bass or slow bass ? Is it real or imaginary ?

Q. I have heard that bass distortion is not important since the ear shows less sensitivity to bass frequencies than it does for mid band frequencies !

Q. Instead of using an accelerometer why not use a second voice coil to monitor the cone motion ?

Q. Are Servobass subwoofer designs suitable for Home Theatre use ?

Q. Is there such a thing as a servo controlled vented box ?

Q. Can the DSP Servo be applied to an open baffle woofer ?



Fast bass or slow bass ? Is it real or imaginary ?


Well yes and no ! This adjective describing sound from subwoofers gets thrown around anytime there is an attempt to integrate a subwoofer with a set of main speakers. Whilst the effect can be real its cause is controversial. People usually associate small and light woofers with faster bass, and bigger woofers that have heavier diaphragms with slower bass. However, providing that a woofer can maintain the structural rigidity of its diaphragm under its operating frequency range there should be no difference in the transient response between a small woofer and a large woofer simply because sound pressure is proportional to the diaphragm acceleration and NOT to the diaphragm displacement or velocity as most people believe.

Given two systems with identical system response functions then both should have the same transient response only differing in the sensitivity and maximum output capability. For example a small woofer with a light diaphragm mounted in a small sealed enclosure may have a cutoff frequency(fc) and quality factor (Qtc), whilst a large woofer and a heavy diaphragm mounted in a larger box may have identical fc and Qtc. Because both systems have identical system response functions both will exhibit identical transient responses. Of course the large woofer will have a higher output capability than the smaller woofer and probably exhibit lower distortion compared to the smaller woofer for a given sound pressure level.

The slow and fast description of bass has more to do with the integrity of the loudspeaker diaphragm to behave like a piston at higher frequencies than it has to do with how fast it can travel and this is where the smaller speaker usually wins out because it's vibration modes are located at much higher frequencies. There is also another aspect of different types of enclosure arrangements such as vented boxes which have a poorer transient response compared to its sealed box counterpart simply because it takes longer time to energize the air in the vent. A not so obvious issue is the way the main speakers integrate with the subwoofer around the crossover frequency. For example trying to integrate a set of vented main speakers which have a 4th order (24 db/octave) low frequency response with a subwoofer which only has a 2nd order low pass filter for its crossover will cause phase problems and likely end up with a suckout in the response. Some subwoofers offer a phase adjustment to correct for this but along with changing the crossover frequency this is mainly a band-aid solution !


I have heard that bass distortion is not important since the ear shows less sensitivity to bass frequencies than it does for mid band frequencies !


This argument gets bandied around a lot when there is talk of reducing distortion in woofers and subwoofers and usually the people who parrot this line don't have a low distortion speaker system to compare with anyway. It seems to be more of an excuse to do nothing rather than to fix up an age old problem that loudspeaker driver manufacturers have been trying to tackle for years.

For a start if distortion in subwoofers is not important then intuitively one could just use an el-cheapo driver with limited Xmax, 3 ounce magnet, no inductance shorting rings, cheap flexible cone material and suspension etc and for all intents and purposes it will sound as good as a well-engineered exotic driver that sells for many times the price. But from experience we all know that an el-cheapo woofer usually sounds bad compared to its well-engineered and more expensive counter-part. There is no free lunch here.

While it's true that the ear is less sensitive to lower frequencies the converse is also true that the ear is more sensitive to the higher order distortion artifacts so you can assume that distortion is even more important. World renowned electro-acoustics engineer, Siegfried Linkwitz provides a compelling argument as to why reducing bass distortion is essential.

Lastly, even on the test bench the difference between low distortion (servo activated) and high distortion (non-servo) is remarkable ! In fact once you remove all of the distortion artifacts using high-gain servo control then the speaker becomes a lot quieter and the mechanical noise of the suspension and air pumping around the suspension now dominates. In fact a good test of distortion is to drive a speaker at 15Hz and providing that there are no resonances or rattles in the room, you should hear very little since your ears are very insensitive to a pure 15Hz tone. If you can hear the sound then you are most likely hearing the distortion artifacts and not the fundamental 15Hz tone and this is usually what people mistake as impressive sounding bass ! Most speakers will fail this test and produce prodigious amounts of distortion especially when pushed hard.

This leads to another interesting observation about bass reproduction and that most listeners confuse impressive sounding distorted bass with clean bass and low distortion bass reproduction as weak bass, especially in home theater environments where non-musical sounds dominate. Eventually people tire of this type of bass performance and look for something that is a lot more refined and a lot less dominant which is exactly what low distortion bass reproduction is all about.


Instead of using an accelerometer why not use a second voice coil to monitor the cone motion ?


If you are talking about using a secondary voice coil wound around the main drive voice coil as a motion sensing device then this is plagued with problems. In an ideal world the sensing coil would respond to the velocity of the voice coil only. A simple differentiation of the signal would yield the acceleration of the voice coil. However in the real world, the sensing-coil not only responds to the velocity of the voice coil but because it is tightly coupled to the main drive coil it also responds to current in the main drive coil thus creating an error in the velocity feedback signal.

To add insult to injury, because the sensing coil and drive coil sit in the same gap and share the same magnetic circuit, the sensing coil experiences the same flux modulation as does the main drive coil. The modulation in flux and inductance as the voice coil moves in and out of the gap also causes distortion of the very signal you are trying to measure. Even if you use constant current drive to avoid the inductance it doesn't help because the inductance rears its ugly head because of mutual inductance between the two coils ! A long sensing coil will help mitigate the effects of stray flux but can't correct for flux modulation. You don't have this problem with an accelerometer because it only responds to motion and not electro-magnetic coupling !

When I was debugging the accelerometer in the original prototype I needed something to compare it with. I used one of the voice coils from a dual voice coil speaker and drove the other coil with an amplifier. Noting that the sensing coil should output a signal proportional to velocity so it was just a matter of integrating the accelerometer signal to equate the two. Things checked out right at low frequencies and low currents but at higher frequencies above 100Hz the amplitude of the signal from the second voice coil was constant all of the time and then it struck me that I was essentially measuring a transformer and not a velocity sensing device. Because the sense coil and drive coil are tightly coupled the velocity signal is swamped by the drive signals coupling across at higher frequencies which is a very undesirable scenario.

Since a feedback system is only as good as the accuracy of its feedback the sensing coil arrangement fails dismally. It may work ok for low frequency small signal low amplitude signals but then the driver is probably producing low amounts of distortion anyway so its effectiveness is mitigated to a large extent. At high cone excursions or high drive current is where distortion becomes dominant but at the same time so does the error components induced into the sensing coil thus mitigating any benefit of a servo system.

Are Servobass subwoofer designs suitable for Home Theatre use ?


Yes they sure are ! In designing our subwoofer technology we have placed a lot of emphasis on accurate music reproduction but this has not been done at the expense of high output home theatre applications. In most home theatre applications the emphasis is on non-musical sounds where distortion is not critical. In actual fact, added low frequency distortion can be a bonus and add more excitement and visceral impact to the movie. It’s a case of larger than life rather than real-life here.

Because Servobass subwoofer designs are fully DSP controlled it is quite easy to turn off the servo and run the speaker directly with some EQ to flatten the response. In essence this essentially makes the sub behave like other non-servo subs with lots of low frequency grunt which essentially is distortion masquerading as impressive sounding bass. With Servobass you can have the best of both worlds with just a flick of the switch. We can’t say the same of our competitor’s subwoofers which can’t be switched to produce low distortion bass. In actual fact most owners of audio visual gear have probably never heard or experienced low distortion bass reproduction so it is very hard for them to appreciate what they are missing out on.

Is there such a thing as a servo controlled vented box ?


In short there is no such thing as a servo controlled vented box where it is understood that a servo feedback loop attempts to reduce the error between the desired signal and the controlled signal which in the case of a speaker is the input signal and output acoustic sound pressure respectively. Any other definition of the word servo is just a play on words or some marketing speil.

To understand why this is so you have to understand how a vented box works. Whilst the mathematics behind it can be quite complex in layman’s terms a vented box radiates sound from both a driver and a vent in the box. Above a certain frequency the vent is essentially ineffective and most of the sound energy comes from the main drive unit. At a certain frequency called the box resonant frequency the mass of air in the vent resonates with the air in the box and this damps the cone motion so that nearly all of the sound energy emanates from the vent. Below the box resonant frequency the output from the vent is out of phase with that from the main drive unit and the total sound energy decreases rapidly with decreasing frequency.

Applying a servo to a vented speaker system requires that it correct for errors in the speaker movement and air in the vent. But at the box resonant frequency the speaker no longer moves because all of the output comes from the vent. Because of this a servo system that exclusively relies on feedback from the driver only would be unstable in this application. Also what corrects for the distortion in the vent if it has no kind of feedback sensor ? Obviously it can't be an error reduction servo system.

There have been attempts to apply feedback control to a passive radiator system where the main drive unit and passive radiator both have an accelerometer attached to them but the net phase shift of the system below the box resonant frequency limits the amount of loop gain that can be applied so the effectiveness of servo control is marginal at best.

A proponent of ‘servo controlled’ subwoofer swears that he has devised a patented method to produce a servo controlled vented box but one look at his system prompts the question as to what controls the air in the vent since there is no sensor or pressure transducer whatsoever.  It seems to be more of a play on words rather than effective error reduction servo system, which is why he produces little or no distortion tests and downplays the importance of low distortion. A further patent attempts to improve on the design by incorporating a passive radiator with an accelerometer attached to it. However it seems that it was never successful in practice.

Also a vented box tends to mask distortion performance at around the box resonant frequency because the driver is made to operate in its linear region, but below the box resonant frequency the diaphragm is unloaded and the driver moves excessively and produces large amounts of intermodulation distortion which gives a vented box its typical characteristic muddy sounding bass.

Can the DSP Servo be applied to an open baffle woofer ?


Yes it certainly can ! An open baffle speaker is essentially a speaker operating in free air. From a mechanical perspective an open baffle speaker behaves in a similar way to a sealed box except that it doesn't have the box springiness to raise the system resonance and change the system Q factor so therefore the DSP Servo will treat an open baffle speaker the same as a sealed box. However that is where the similarities end because the difference in the acoustic reponse between a sealed box and open baffle speaker is completely difference.

An open baffle speaker essentially behaves as a dipole where both sides of the speaker radiate out of phase to each other. The only reason why an open baffle speaker makes any sound at all is because of the finite path length between the front and rear of the speaker which is essentially determined by the baffle length. As the frequency is lowered a point is approached where the difference in length between the front and rear radiation is equal to half the wavelength of the sound which is where cancellation occurs ! But before that the signal first rolls of at 6 dB per octave above the resonance frequency and then at 18 dB per octave below the resonance frequency which corresponds to the natural 12 dB per octave rolloff of the diaphragm acceleration.

However in a room environment, room gain plays a vital roll in boosting the acoustic output but because the driver is essentially unloaded it is easy to overdrive the speaker in order to achieve a suitable output level at low frequencies. Attention should be payed to the mechanical displacement limits of the driver when expecting it to deliver low frequencies in this arrangement.



Last Updated November 2, 2014