I am not sure that the servo is as relevant a factor as the limited frequency band that a sub is likely to be asked to perform in. I am not at all an expert on servo's but I think that the benefit that is derived from the servo relates to the ability to control the driver such that you are able to move "big" air and still be ready for the next note. If you are going to move a good deal of air, you do not want your driver to still be resonating when it is time to reproduce the next string of notes. This is where I believe the servo comes in "controlling" the resonance. Ever notice that one of the major design trades that sub designers make is two smaller drivers as opposed to the one larger driver. The designer using the two smaller drivers is attempting to offer an appropriate level of ability to reproduce low notes loudly while being faster to recover. The designer using the single large driver has the advantage of being able to move more air but needs to find a way to control the resonating cone so that it is ready to reproduce the next note. That is why dual driver sub designs are generally thought of as faster than the single driver subs. However, there are all sorts of design tools that can be used and there are variations on every theme.
Back to your original question, I think the attractiveness of the class D amp over a class A/B biased amp is efficiency with the designer taking advantage of the fact that the amp is not being asked to perform over a broad frequency band. It takes power to drive speakers at low frequencies so efficiency can be an advantage. Notice however that you will not see a class D amp being asked to drive speakers over a full range of frequencies such as those from your main left and right speakers.
There are some sub designers that frequent this site. Hopefully one of them will step in and rescue me. In fact I would love to hear from one of the designers that frequents the site because I think that this is a really interesting design topic and one that I would love to hear more about.
We've evaluated many Class D designs, and rejected them all...for now. Compared to a beefy, conventional Class AB subwoofer amp, they sound wooly and sluggish. Underdamped. I wish it wasn't so, because they are super cheap! Class AB amplifiers cost more and probably weigh ten times more per watt.
Compare a big Class D amp and a big Crown amp on the same passive subwoofer, and tell me what you think. There ain't no free lunch.
Funny you should mention that Paul. I was just looking at Crown the other day because I am strongly considering doing something with a passive sub (if I can find one that is just right for the room ) and was looking at the Crown amps. Was wondering if they were A/B biased or not and I guess you just answered the question.
I can corner place the sub and if I were to have my way I would probably put one of the Sound Physics SPL-BDEAP32 subs in the room even though it might be overkill. Just don't know if I have room for 42" x 18" x 42".
Class D amps are digital amps that use "pulse-width modulation", which is slightly similar to FM (Frequency Modulation). There is an oscillator that produces a very high frequency square wave (the carrier), which means the output transistors are either always fully on or fully off, which lets them run cool.
By being used as switches (digitally), rather than variably (linearly), the transistors remain much cooler, since they never have high-voltage, high-current across them, and don't waste electricity generating heat, resulting in a highly-efficient (around 90%) amplifier (more power-supply electricity actually moving the speaker).
The input signal modulates (varies), not the amp's output voltage, but the 'pulse-width' (the ratio of time the output is on as compared to off), a.k.a. the 'duty cycle', of the output signal. Finally, a low-pass filter strips the output of the high-frequency component of the signal.
The resulting output is a 'stair-step' representation of the original signal, the number of steps-per-second determined by the carrier frequency, and the faster the carrier, the higher the frequency that can be reproduced, without 'aliasing', just like with CDs and other digital media.
Also, the higher the carrier, the easier it is to filter it from the output. It's the speaker itself that "converts" the stepped signal into a (hopefully) accurate copy of the original, the same way a light bulb 'averages' A.C. into the 'heating equivalent' of D.C., which is what RMS represents.
As for Sunfire amps, the power supply is basically a digital amplifier, and its output is used to power the linear-transistor amplifier. By modulating the power supply, its voltage is kept 6 volts above the output voltage, which, like a digital amp, keeps the voltage dropped across the transistors to a minimum, allowing for cool operation.
Jnug, how about using the infinite-baffle design that's becoming popular (again)? Just do a search here or on the web if you're interested in more. I'd explore it myself if I had a space that would work.
My first sub was passive with a Samson amp (class a/b) which I sold but still have the opportunity to listen to occasionally. I now have a hybrid class a/b & D (Class t) powered sub. And I haven't noticed any difference in sound between the 2. But I know a little of the physics of electricity, and increased distortion (inter-modulation and harmonic are an inescapable fact of non-linear loads, such as the output devices of efficient amps.
nice explanation. I plan on building my first sub soon so this is very interesting. Not knowing squat about amps it raises some questions in my mindâ€¦..
A square wave is composed of only the odd harmonics from a signal. If the output of a class D amp is a square wave, is much signal quality lost due to missing theoretically Â½ of the signal? Maybe this explains the â€œsluggish soundsâ€? I realize we all hear things differently and "sound" is subjective.(personally, my high freq hearing is shot)
Do you know if the sampling rate varies amongst the amps? I would think it is at least twice the rate (Nyquist theory) With the low subwoofer freqs does this mean bigger "steps"? equals not as smooth sound?.....
Do you know if the sampling rate varies amongst the amps?
Yes, the sampling rate varies between the various digital amps. The newest digital amps sample at frequencies well above 40khz (more like 500khz+). Although old, check out this review ... it includes a good explanation of D-class amps.
A forum community dedicated to home theater owners and enthusiasts. Come join the discussion about home audio/video, TVs, projectors, screens, receivers, speakers, projects, DIY’s, product reviews, accessories, classifieds, and more!