Originally Posted by 3db
My bad. The Velodyne was used in the other forum and I mixed the two up. I meant to post sensing coil.
So if the servo does limit frequency response, does it do it on both ends of the bass spectrum? I thought the sensing coil generated feedback based on the amplitude of the signal the driver is producing rather than its frequency. That's where I'm getting stuck. If what you say is true, then the LV12-R that I own is capable of 19Hz just from the driver/cabinet alone and that the servo reduces the amount of distortion of the output?
No need to get chirpy. I'm not criticizing you.
Sorry, general snarkiness on several fora seems high lately and I must have read more into your post than you meant. My hand-waving posts are not usually meant to be 100% rigorous, just trying to provide a little insight without assuming an engineering background (in which case you wouldn't be asking what feedback does).
It is hard to describe feedback theory and application without knowing the technical level of the audience. And even harder in words; I work better with a whiteboard. The following is hand-waving.
The servo can extend the useful linear range of the sub, which makes it possible to modify the intrinsic frequency response curve of the driver/box combination, which allows for a bit more bass extension. The limitation in frequency is at the upper limit; the lower limit is generally unaffected by feedback unless intentionally modified (e.g. analog tone controls were often designed as feedback circuits).
The frequency limitation of a servo, like any feedback circuit be it in your speaker, power amplifier, or car's engine controller, depends upon how much gain and bandwidth you have. See e.g. https://en.wikipedia.org/wiki/Loop_gain
. In words, you sense the output, invert it, and apply it back to the input of the amplifier (or apply it to the inverting/negative input of the amplifier circuit with the signal going to the in-phase/positive input). This is feedback, not specific to a subwoofer. There is some delay through the amplifier and potentially through the feedback circuit itself. The amplifier only has so much gain and bandwidth; the greater the bandwidth, the lower the delay, and the greater the gain, the more it can be reduced by feedback to linearize (improve) the output.
The greatest bandwidth is for an open-loop amplifier with no feedback. But that is also the highest distortion way to do it. Most add some feedback, with the output signal used to compensate the nonlinearity of the amplifier (or speaker), and that trades some of the amplifier's gain and bandwidth for greater linearity (lower distortion) as well as helping other things (like lower output impedance). Look up Bode plots. A catch is that, since it takes a little time for the signal to get through the amplifier (plus sensor, accelerometer, voice coil, or just a resistor circuit from the output), at some point that delay shifts the phase enough so that the feedback ends up adding instead of subtracting and now the signal builds up uncontrollably and you've built an oscillator (the amp goes unstable and squeals). You have to design the feedback circuit with the amplifier so that by the time (frequency) the signal reaches that much delay it is below the unity-gain point of the amp. Then the signal is attenuated instead of amplified, and the amp remains stable.
So you end up trading some of the amp's intrinsic gain-bandwidth to provide a more linear outcome, in general reducing both gain and bandwidth compared to an amplifier without feedback. Same thing applies to subwoofers or steam engines with a servo circuit. But, to make a stable design, the bandwidth must be well above what you actually need, and thus it is with Rythmik subs (or other servo speaker designs). Adding a servo means you may need a little more bandwidth in the amp to achieve the same bandwidth as a non-servo design, but for a sub that is easily achieved. For Ka band radar amp up around 40 GHz, it's a little harder
, and you rarely find the kind of gain-bandwidth you find in audio circuits.
That is why I used (albeit quite erroneous) light analogy; the bandwidth may be a little lower, but still so high it doesn't matter. If the light reaches my eyes when I throw the switch at only half the speed of light in a vacuum that is still quite fast enough to do the job without me noticing, and listening in a vacuum is uncomfortable and has numerous other drawbacks.
HTH - Don