What is a 'servo' subwoofer

J

josko

Audioholic
What is a 'servo' subwoofer and how is it different from the non-servo kind? I've heard they are more accurate and therefore more suitable for music reproduction. If that's true, what's the downside of the tradeoff? Could anybody point to a link discussing the differences? Thanks in advance
 
jinjuku

jinjuku

Moderator
A servo based sub will have a driver with multiple voice coils. Each hooked up to a circuit on an amplifier. One voice coil is hooked up to a 'correction' circuit on the amp. This circuit can see the driver doing something it shouldn't and send a 2nd signal to the other voice coil correcting the abnormality.
 
darien87

darien87

Audioholic Spartan
Jinjuku's got it right. I have a servo sub by Rythmik and I think it sounds great. Whether that's a facet of the servo system or just a well designed sub and cabinet I couldn't tell you. But if you want a little more info on how servo works, check out the Rythmik site.

http://www.rythmikaudio.com/DirectServo.html
 
mtrycrafts

mtrycrafts

Seriously, I have no life.
What is a 'servo' subwoofer and how is it different from the non-servo kind? I've heard they are more accurate and therefore more suitable for music reproduction. If that's true, what's the downside of the tradeoff? Could anybody point to a link discussing the differences? Thanks in advance
Here is a quote from Wiki:D
Servo subwoofers
Some active subwoofers use a servo feedback mechanism based on cone movement which modifies the signal sent to the voice coil. The servo feedback signal is derived from a comparison of the input signal to the amplifier versus the actual motion of the cone. The usual source of the feedback signal is a few turns of voice coil attached to the cone or a microchip-based accelerometer placed on the cone itself.[28][29] An advantage of a well-implemented servo subwoofer design is reduced distortion making smaller enclosure sizes possible.[30] The primary disadvantages are cost and complexity.[31]



No mention of two voice coils. Velos are servo controlled with one voice coil. Primarily it is to reduce distortion.
But, one must be aware of the psychoacoustics at low frequency;) Distortion is not easy to detect even at 10% distortion, hence all the ratings at 10% or up to it.

May want to read this
http://www.axiomaudio.com/distortion.html#
 
C

cschang

Audioholic Chief
The Rythmik solution uses two coils instead of accelerometer, a sensing coil in addition to the traditional driver coil. It is a unique and patented system.
 
TLS Guy

TLS Guy

Seriously, I have no life.
Yes, a servo, means there is feedback control. The sensor is a coil or more usually some type of accelerometer.

The net effect is variable Eq. The option of servo control is only possible in a sealed design.

Now the downside, is very large power requirement from the amplifier. This can stress the driver, so the driver must be very robust so as not to exceed xmax and generally not be susceptible to damage from the large powers that have to be applied to keep the cone under control, and above all extend bass response.
 
GranteedEV

GranteedEV

Audioholic Ninja
The net effect is variable Eq. The option of servo control is only possible in a sealed design.
Eh? I know Rythmik's implementation is a bit different, but here's what it says on the DIY section of their website

Alternative subwoofer applications

Direct Servo can be used for any subwoofer, and is not limited to sealed and vented configurations.

In addition to the standard sealed, vented and passive radiator configurations, Direct Servo can work with the following types of subwoofers:

Infinite baffle
Band pass
Isobaric
Open baffle
Horn loaded
Transmission line

If you wish to use any of these alternatives, contact us regarding the purchase of a custom designed Direct Servo board.
 
TLS Guy

TLS Guy

Seriously, I have no life.
Eh? I know Rythmik's implementation is a bit different, but here's what it says on the DIY section of their website
Servo does not work well with applications other than sealed, as you have to have a steep high pass filter above the point were the cone decouples from box tuning, otherwise the excessive cone travel at the pint of decoupling is severely exacerbated. There has never been shown and advantage of servo in tuned alignments and fact if anything the contrary.
 
highfigh

highfigh

Seriously, I have no life.
Jamo had a speaker called the MFB200 (Motional Feed Back) in the late '70s and it was pretty unique. The woofer's baffle was a sandwich of MDF/lead/MDF, it has a 12 band EQ on the front, dome midrange and ribbon tweeter (possibly Bohlender-Graebner) and was actively tri-amped. I'd like to listen to them again and I may ask the owner of the store where I worked if I can- he has 2 pairs. They sounded really good, although my critical listening skills weren't as developed at the time. Still, I knew what I liked and what sounded good.
 
R

Rythmik

Enthusiast
Servo does not work well with applications other than sealed, as you have to have a steep high pass filter above the point were the cone decouples from box tuning, otherwise the excessive cone travel at the pint of decoupling is severely exacerbated. There has never been shown and advantage of servo in tuned alignments and fact if anything the contrary.
That is not quite true. You can read my patent. It has a complete theoretical foundation of servo operation in speakers that no one even discover before I did. Before my patent, all servo operation in speakers is more like tweaking. It is like let us modify the circuit here and there and see what happen. As a result, they can only operate on the simplest configuration -- sealed subs. They are completely at a lost with other configurations. There is no synthesis/prediction per se. We all should learn from past history/experience. For instance, what is the biggest contribution from papers like the original Thiele and Small papers discussing the now de facto T/S parameters? It is the ability to allow us to have a complete flow from modelling (understand the interaction between all variables) -> simulation (without having an actual driver) -> synthesis (come up with physical parameter for the manufacturers so that we can get the desired T/S parameters and the desired frequency response). That is the complete engineering framework. Similar is the contribution of my servo theory. Once you completely understand what servo do, it is a piece of cake to extend it to any speaker operation. I even had a prototype that works up to 1khz that neither Velodyne nor Paradigm can do.

Every time that we come up with a brand new design, we need to understand if we can relate it to something we already understand. For instance, Thilo at TCS had been experimenting with drivers with induction motor. He was excited about it and claimed it would be the biggest break through. I only have one question for him when he told me the idea -- "What is the T/S parameters?" He said he cannot find it because it is non conventional. I told him that he needs to find a way to get those parameters before committing big money because if you cannot even borrow what we already know, he needs to come up with a brand new design methodology/framework and he has no control of what the outcome would be and he will end up with countless experiments. For the minimal, he cannot even compare it to conventional drivers to see how the efficiency compares and how he can improve the efficiency or response. It is every-one's worst nightmare to go back to the old days of "let us tweak this and that value and see what happens".

Come to servo and think about it. There is no myth about it, the servo operation changes something meaningful that we already know from circuit theory. It is as fundamental as feedback in amplifiers. So what does feedback changes in amplifiers? "Equivalent" output impedance, distortion, power supply rejection... Shouldn't servo be similar to that? It changes the so-called "equivalent" physical resistor in the speaker-amplifier interface (resistance, capacitance, and inductance), and it changes distortion,.... Anyway, you get the idea. I have been selling vented servo subs for 4-5 years. I don't sell as many. But it does work.
 
Last edited:
GranteedEV

GranteedEV

Audioholic Ninja
I have been selling vented servo subs for 4-5 years. I don't sell as many. But it does work.
That's just because you don't have a ~3-400 dollar 10 incher with an ugly veneer to get total market penetration/domination :p

Seriously. Put one out. I'll buy it.
 
darien87

darien87

Audioholic Spartan
I have been selling vented servo subs for 4-5 years. I don't sell as many. But it does work.
Thanks for chiming in here Brian. I still love my 12" sub. I have my system set up so that the 12" is always on, but I have to manually turn on the other 15" sub I have when I want to use it for movies and stuff. I keep getting surprised at how deep the 12" plays on it's own. When I turn on the 15" I can get a little deeper bass and more tactile feel, but not much.

My Rythmik 12" keeps up just fine.
 
VonMagnum

VonMagnum

Audioholic Chief
OK, so I'm looking at subwoofers now as I need a second one in my system to even out room modes for the new extra two rows of seating since upgrading from 6.1 to 11.1.6 and I've been thinking why not extend my bass lower while I'm at it. My current sub is fairly old and while some have told me it's "junk" I'm measuring a flat response in my room at least 22Hz at the MLP that started to fall off steeply below that, but with room correction on, it's now down only 4dB at 20Hz (I don't have any test tones below that currently). I don't play much pipe organ music and shaking the walls with notes I can't hear, well I don't know if it's worth it.

But assuming it is, I've been looking at my sub options and for all the rage about SVS subs, their frequency response graphs don't look that great to me. For example, with the SVS SB-2000 If it's showing 93dB in the response curve at 30Hz, it's like 87 dB at 20Hz (apparently -5dB is "fit in" with the +/-3dB by averaging in the middle instead of where it sits even at 30Hz and above, which is misleading in the numbers, but obvious in the graph) and 83dB at 15-16Hz or so (graph isn't easy to read precision). OK, so in other words, that sub is WORTHLESS at 16Hz. It can't put out the pressure wave to "feel" 16Hz and you can't hear it, so I'm not going to get anything out of that. Where's the improvement of this "modern" sub compared to my "ancient junk" from the mid to late '90s? It's down 4dB at 20Hz and only 1dB at 21Hz compared to 30Hz in my room.

How is that "junk" by "modern standards" ? I'm not seeing anything to make me believe that 12" "modern high excursion" sub with almost 3x the power amp rating is doing anything "significantly" different other than being slightly smaller and weighing slightly less. Is it "tighter" sounding? I have no idea without hearing it, but I do know "speed" is typically associated with higher frequencies and thus if you keep your sub out of the >80Hz range, it's not playing the notes that make it sound "slow". It sure as heck sounds good with my PSB speakers (similar to my Carver ribbons with 10" drivers). My PSBs are flat to 35Hz and with music in that range, they sound less loud but then I run the sub +6dB for more bass (hard to do that on the PSBs without EQ controls).

Now I'm using that example since that's supposedly the equivalent "modern" sub spec wise to my old 15" sub (bigger driver, less power that still shakes the walls at a fraction of the power it can run at with the dial at 1/3 in that room).

So I look at the SVS SB-Ultra 16 graph next ($2000 sub rated to 16Hz +/- 3db) that's supposed to be 'all that and a bag of chips' as they used to say. The 20Hz numbers look WAY better on the graph (down only about 2dB @20Hz compared to 30Hz, say 92 instead of 94dB at that voltage input). BUT look at the 16Hz number (what I'd be interested in after paying $2000 for a subwoofer) and I'm seeing about 87dB. Well, that's down 5dB from 20Hz and 7dB from 30Hz. With a bit of fudging (half dB off in reading the graph which is not marked to 1dB), I guess you could call that +/- 3dB, but again it's misleading since it's really more like +0/-6dB based on the long term average from 30Hz-150Hz where it's pretty darn flat.

Is that sub going to rock my world with 32' pipe organ stops at 16Hz? It might do OK, certainly if I bought two of them, but at a cool $4000). That's a lot to get accurate pipe organ music (again, I don't know how much better explosions sound with big 16Hz support as opposed to 'just' 20Hz or even 24Hz.

So that brings me to SERVO subs. I'm looking at Rhythmic Audio's graphs and damn, I like what I see at 16Hz! The problem is that I know that the flat output only extends to a certain volume level and dies above that. What's hard to discern is what that volume level might be, exactly. In other words, I don't think I'm going to hit Dolby's 115dB peak level at 16Hz, probably nowhere near that level and that's the kind of data I need to make an informed decision. 85dB flat at 16Hz isn't going to rock my proposed pipe organ world. Is there some data somewhere on Rhythmic's absolute output capability (anechoic and perhaps a typical medium room sized average) based on frequency? Basically, I need to know what I'm dealing with so I know if I want to order one and how big I'd need to go for various performance parameters at "useful" levels. Given how difficult it is these days to preview multiple subs, I might have to go on data alone.

Overall, I just don't know if it's worth the expense to pursue 16Hz or if I should just be happy with 20Hz and work on evening out the room modes with a second sub like I originally planned. I suppose if the new sub really did sound better by itself than the old sub, I could give that one to my mother and get a second one, but then looking at SVS, two SB-3000s could be one SB-Ultra16 instead (and let's face it, I sit at the MLP most of the time so am I that worried about the occasional person the second and/or third row? I could throw the old Def Tech sub back there for room mode evening out for those two rows in most of the range and have the Ultra16 up front where it's going to do its thing for the MLP just fine (and even turn off the back sub if on one is back there if it's somehow muddying the sound of the front). I didn't really want to spend $2000 (I could get a new powered stretched screen and a 4K projector for not much more), but if 16Hz were going to really rock my world.... Would it? I don't want to shake the Conan replica sword off the wall either.... ;)
 
ryanosaur

ryanosaur

Audioholic Overlord
LFE channel for movie soundtracks, electronic music, hip-hop, some percussion.
 
VonMagnum

VonMagnum

Audioholic Chief
If you really want 16Hz capability, you need one of these:

https://www.deepseasound.com/products/mariana-24sc-24-subwoofer

I need one like a hole in the head, but it still tempts me.
Wow. 210 pounds. I'd have to get some kind of lifting dolly to even move it. There's only two places it could even fit in the room. 4000W RMS amp... I think I'd need a new 50A breaker and outlet just for it (cha-ching). At what point in the frequency spectrum would a 24" driver be too slow to respond quickly enough, I wonder.

A friend of mine I've known online since the days of Fidonet in 1990 on a BBS linked to the Internet/Fidonet system) has a system with as he puts it, "six 18" industrial woofers and there are four special design 18" subwoofers with a peak to peak excursion of about 4" and power handling of 3600W each. They were designed to power outdoor rave concerts and to provide sub bass output for football stadiums. I've only made measurements at 16Hz, using the opening to Also Sprach Zarathustra as the source (16 Hz organ pedal tone) and measured 133dB from 30 feet away, but part of the ceiling fell down during testing and I had to stop. The subs were designed with an F3 of 8Hz." He calls himself THE BASS PIG. Nothing I could fit in my room would hold a candle to his current setup. Oddly, he's got a lot of Carver equipment including Sonic Holography that I use on my Carver AL-III system.
 
S

shadyJ

Speaker of the House
Staff member
Is there some data somewhere on Rhythmic's absolute output capability (anechoic and perhaps a typical medium room sized average) based on frequency? Basically, I need to know what I'm dealing with so I know if I want to order one and how big I'd need to go for various performance parameters at "useful" levels. Given how difficult it is these days to preview multiple subs, I might have to go on data alone.
Here is some measured performance data on Rythmik subs. Click on the model number to see the full measurement suite.

It looks to me like Rythmik's servo system can deepen the system Fs, but the problem is, with sealed subs, they will run out of excursion very fast at 16 Hz if you are interested in that frequency. One of the very few sealed subs that has real output at 16 Hz is the aforementioned Deep Sea Sound sub. If you are interested in 16 Hz, you pretty much have to get a ported sub if you have any kind of budget or power restrictions. Most of the major manufacture direct sub makers have ported subs that are flat to 16 Hz, SVS, Hsu, Rythmik, etc. By the way, I wouldn't say that extension down to 16 Hz is enormously different than 20 Hz. It is certainly a perceptible difference but not a huge one.
 
Irvrobinson

Irvrobinson

Audioholic Spartan
Wow. 210 pounds. I'd have to get some kind of lifting dolly to even move it. There's only two places it could even fit in the room. 4000W RMS amp... I think I'd need a new 50A breaker and outlet just for it (cha-ching). At what point in the frequency spectrum would a 24" driver be too slow to respond quickly enough, I wonder.
For the real world playing real sound tracks and music I would bet a 15 amp circuit would work just fine. Size and weight?... you can't cheat physics. As for driver "speed", the damned thing has an 80lb magnet; that cone is probably going to have no problem at all with a typical 80Hz crossover. See the measurements tab.

Go big or go home, or forget 16Hz. The ported subs James mentions are pretty big too.
 
VonMagnum

VonMagnum

Audioholic Chief
Yeah, I think I need to hear a powerful 16Hz demonstration before committing to that kind of dough. I should go visit my Internet friend and hear his system (about a 10-11 hour drive, but I have a vacation coming up in a couple of months).
 
newsletter

  • RBHsound.com
  • BlueJeansCable.com
  • SVS Sound Subwoofers
  • Experience the Martin Logan Montis
Top