Motional Feedback (MFB)?

[solderdude]

Yamaha uses a servo system on their YST Subwoofers.
View attachment 164401


I bought an old one this spring. Have it under the bed. Not a giant subwoofer but suitable for my bedroom.

Edit:
Or wait a minute, Yamaha writes negative outpot impedance amplifier. Maybe it's not MFB?

Hm, negative outpot impedance amplifier... wasn't that the technology Audio Pro used and called ACE-Bass? I think it was the engineer / designer Karl-Erik Ståhls at Audio Pro who received a patent for that technology. If I remember correctly.

Edit 2
Apparently.
Inventor Karl-Erik Ståhl
The engineer Karl-Erik Ståhl is behind many of Audio Pro’s patented ingenious innovations, most notably ACE-Bass, which allows a deep bass from small subwoofers.

Attached image,Karl-Erik Ståhl

Vår historia - Audio Pro - Play Nice

För över fyrtio år sedan började vi bygga högtalare med en enda mål; att skapa det bästa möjliga ljudet till det bästa möjliga priset. Idag hittar du Audio Pro i fler än 45 länder på fyra kontinenter. Och det är fortfarande samma ambition som leder oss. Små högtalare med stort ljud.

www.audiopro.com

Looks like a bassport to extend the bass and an amplifier that senses current (no MFB)

 

[DanielT]

When a pure 50Hz sine has inaudible harmonics (10%, mainly 2nd and 3rd ?) at 100Hz and up how can those (inaudible ?) harmonics then mask harmonics of the instrument itself that can sometimes even exceed the amplitude of the fundamental ?
Or are you talking about equal loudness contour effects combined ?
Maybe resonances are a bigger issue in muddying the sound ?

In addition, in an ideal subwoofer bracing then the resonances should also be pushed up in frequency, outside the audible range. That as so long steep LP -filters are used, 24dB for example. That's how it should work, I think. See attached picture, it seems to be a sensible bracing
(picture taken from the web, not me who created it).


It is clear if the subwoofer element / driver itself has a high distortion, then no design in the world can help to get rid of it. But modern drivers, there are many that perform well. See attached picture (not me who made that compilation).

Testing of drivers and technical data from:

HiFiCompass |

hificompass.com

 

[solderdude]

You can lower distortion by using MFB. It not only equalizes and extends FR but also lowers (but not removes) distortion up to the clipping points.
EQ and porting does not do this.

Of course cabinet resonances/vibrations and breakup of cones won't improve. Only the driver linearity.

 

[motoindo]

Holidays greetings & the best wishes for the new year from Amsterdam, may the bass be with you all, stay healthy & happy !

 

[voodooless]

Holidays greetings & the best wishes for the new year from Amsterdam, may the bass be with you all, stay healthy & happy !

View attachment 175367

Oe, what’s all that? Great of you to join. Welcome

Let’s start with the questions right away then :
- Generally we see these systems with closed boxes. But what about enclosures what are leaky, like we see for directivity control purposes (like D&D)?
- What about driving multiple woofers? If we have two of the same, how well would it work to only use a single accelerometer?

 

[motoindo]

Oe, what’s all that? Great of you to join. Welcome

Let’s start with the questions right away then :
- Generally we see these systems with closed boxes. But what about enclosures what are leaky, like we see for directivity control purposes (like D&D)?
- What about driving multiple woofers? If we have two of the same, how well would it work to only use a single accelerometer?

I'll respond in detail tomorrow, it's already 22:32 over here and it has been a busy day keeping my grandson happy

 

[voodooless]

I'll respond in detail tomorrow, it's already 22:32 over here and it has been a busy day keeping my grandson happy

No worries. It’s the same time here since we’re not very far away relatively speaking . Have fun

 

[motoindo]

Oe, what’s all that? Great of you to join. Welcome

Let’s start with the questions right away then :
- Generally we see these systems with closed boxes. But what about enclosures what are leaky, like we see for directivity control purposes (like D&D)?
- What about driving multiple woofers? If we have two of the same, how well would it work to only use a single accelerometer?

Ok, 07:02, all fresh and shiny

Motional Feedback - often referred to as "servo drive" - attempts to mimic the physical cone movement to the electrical input signal which in a nutshell explains why it sounds so distinctly better then most bass systems as it eats away the damping influence of the drivers suspension and enclosure. See EVE for more info.

The very cleverly designed D&D 8C uses dsp to LT the twin woofers at the back, I did an experiment with replacing the DSP LT with an analog servo loop but found the results to be of less commercial interest. Unless they decide on doing a "mini" 8C with half the metric volume of the 8C I don't foresee much gain. With regards to it's "leaky" front driver, that could work as long as the leak is purely resistive - i.e. independent of frequency - as any port frequency will mess up the control loop. But the biggest issue with adding a servoloop to midbass drivers like the 8C front unit is with the extended polepiece / phaseplug which effectively rules out using centrally placed accelerometers like my StarBass sensors. Using alternative placements like can be done with ClingOn types also won't work because the added weight messes up the lightweight cone symmetry. I would try a negative output impedance (current drive) amplifier first.

With regards to using primary and secondary drivers, unless you are sure all drivers are within tight production tolerances and all "see" the same enclosure load the highend way forward remains individually looped drivers. Having said that, I did witness great results from an impulse compensated master / slave setup done by RmsRob so definitely do give it a try !

As for the boards displayed in my xmas card above, the grown/up is a 3way system I have under development and planned for feb 22 launch depending on whether it's coax mid-high unit is delivered on-time. The Little/Sub is geared towards the bass deprived LS3/5a Lovers community

Both SDBH2D and SDB1D DIY boards are on hold due to global chip shortages but will eventually come combined with an accelerometer allowing the DIY community to build their own servo systems.

 

[voodooless]

Ok, 07:02, all fresh and shiny

Thanks for getting up that early to answer my questions

The very cleverly designed D&D 8C uses dsp to LT the twin woofers at the back, I did an experiment with replacing the DSP LT with an analog servo loop but found the results to be of less commercial interest. Unless they decide on doing a "mini" 8C with half the metric volume of the 8C I don't foresee much gain. With regards to it's "leaky" front driver, that could work as long as the leak is purely resistive - i.e. independent of frequency - as any port frequency will mess up the control loop. But the biggest issue with adding a servoloop to midbass drivers like the 8C front unit is with the extended polepiece / phaseplug which effectively rules out using centrally placed accelerometers like my StarBass sensors. Using alternative placements like can be done with ClingOn types also won't work because the added weight messes up the lightweight cone symmetry. I would try a negative output impedance (current drive) amplifier first.

I was mostly referring to the mid driver on the front. So a mostly resistive enclosure would work? That is definitely interesting! I was thinking about a bit larger setup though, 1” compression driver + horn and one or two 10” woofers in a resistive enclosure. Choice of woofer is therefore still open, although I narrowed it down to either B&C 10NW76 or Beyma 10MC700nd.

In any case, what would the MFB mechanism do for the response of such a leaky system?

With regards to using primary and secondary drivers, unless you are sure all drivers are within tight production tolerances and all "see" the same enclosure load the highend way forward remains individually looped drivers. Having said that, I did witness great results from an impulse compensated master / slave setup done by RmsRob so definitely do give it a try !

I’ll have a look, thanks! What are the chances that one can have the woofers work in a passive .5 way setup (where onenig the two is filtered 6th order above baffle step)?

Both SDBH2D and SDB1D DIY boards are on hold due to global chip shortages but will eventually come combined with an accelerometer allowing the DIY community to build their own servo systems.

I’m not in a hurry, so I’ll eagerly await what’s coming.

 

[motoindo]

The biggest gain with using servodrive in a open enclosure is that it will attempt to compensate for the driver suspension losses at low volume (excursion) levels which normally cause a drop in lf response.. At higher levels it will attempt to suppress harmonic distortion until the drivercone movement becomes non-pistonic.

Main tips :

• if not registered already do so at mfblabs.nl, then install the woofertool. Intl friends, sorry, Dutch only for now.
• lookup your drivers in the loudspeaker database and download the corresponding driver data into the woofertool
• play around with the woofertool box volume, power and loop gain values to get an idea of what's possible.

The tool assumes you will be using a closed cabinet and as such won't be of much use in your case but it does help in building an understanding of what's possible.

Enjoy !

 

[solderdude]

sorry, Dutch only for now.

geen probleem... de beste wensen voor 2022 alvast... nog 2 uurtjes volhouden.

 

[voodooless]

Main tips :

• if not registered already do so at mfblabs.nl, then install the woofertool. Intl friends, sorry, Dutch only for now.

The main issue is actually that it's a Windows tool . Luckily Crossover is to the rescue.

The tool assumes you will be using a closed cabinet and as such won't be of much use in your case but it does help in building an understanding of what's possible.

Thanks, that's quite interesting. Looks like the two woofers I chose are quite matched. The Xmax difference is only slight, and also cabinet wise, it's more or less the same.

The biggest gain with using servodrive in a open enclosure is that it will attempt to compensate for the driver suspension losses at low volume (excursion) levels which normally cause a drop in lf response.. At higher levels it will attempt to suppress harmonic distortion until the drivercone movement becomes non-pistonic.

The main difference between these two woofers is actually the Xmech/Xlim. the Beyma is speced at 24mm (one way), the B&C does not give a specific number, but I suspect it is not that much (Xvar is 7.5mm, but it's not the same). Possibly that is an advantage for the Beyma, especially when MFB is concerned. Both are ultra-low distortion to begin with, with the B&C having a slight edge on the Beyma.

 

[Biblob]

The biggest gain with using servodrive in a open enclosure is that it will attempt to compensate for the driver suspension losses at low volume (excursion) levels which normally cause a drop in lf response.. At higher levels it will attempt to suppress harmonic distortion until the drivercone movement becomes non-pistonic.

Main tips :

• if not registered already do so at mfblabs.nl, then install the woofertool. Intl friends, sorry, Dutch only for now.
• lookup your drivers in the loudspeaker database and download the corresponding driver data into the woofertool
• play around with the woofertool box volume, power and loop gain values to get an idea of what's possible.

The tool assumes you will be using a closed cabinet and as such won't be of much use in your case but it does help in building an understanding of what's possible.

Enjoy !

Hi Chris,

If I understand correctly a simplified result of using MFB is a trade-off in SPL capability for a lower f3. I haven't tried the woofertool, but I had this thought:
I would think using pro-drivers 6-8" that have a lot higher sensitivity and power ratings but generally lower f3, would be great drivers to apply MFB too. Especially when aiming for an f3 of around 80hz and crossing over to subs, I could see one could receive more SPL headroom this way than using a home-audio focused drivers.
Do you have any experience using pro-drivers instead of more home-audio focused drivers?

 

[retro]

@motoindo
I have checked your website for a couple of years now. Great work!
I have had several MFB systems over the last 30(!) years. Infinity IRS Epsilon, destroyed two woofers while I had them..they just started to oscillate at random times..but when they worked, they were fantastic.
Also had Paradigm Servo 15. Very nice subwoofer. Several Velodynes over the years. Liked them a lot.
So yes, I'm a believer in MFB.
And I want to get back in the game. A couple of of MFB subs at a reasonable cost would definately find their place in my current home. DIY is fine with me. Building a pair of small boxes is no problem.

So, will you offer complete kits in the future with 10-12-15 inch woofers..?

Please tell us about the future plans....?

CH from Sweden..

 

[motoindo]

Do you have any experience using pro-drivers instead of more home-audio focused drivers?

Yes, back in 2014 I did some tests with a rcf L8S800 and a early drummer accelerometer prototype powered by a TDA8950 based class D amp which worked pretty well.

Accelerometer closeup, early design with a 11mm piezo beeper & bf545 jfet buffer

2 ch poweramp with TDA8950 in BTL and a LM3886 for the high channel powered by a connex smps

The Faital pro with another accelerometer test

Same FE100 but equipped with an early StarBass prototype in an attempt to minimize pickup of non-axial cone movement.

Also tested this little SICA neo driver with my Little/One, worked very well with just 50watts. Didn't sing like the scanspeak but played very loud. There are no plans for a pro little/one yet but if that changes this one will definitely be on the top of my list, -3dB @ 50hz from just under 3 liters ....

 

[motoindo]

I have had several MFB systems over the last 30(!) years. Infinity IRS Epsilon, destroyed two woofers while I had them..they just started to oscillate at random times..but when they worked, they were fantastic.

Dunno for sure but these used a pickup coil ? if so that's probably the reason why they became instable as their loop is kinda temperature sensitive due to transformation effects between the driving and pickup coil. If they use accelerometers it's probably electrolytics gone bad causing unwanted phase problems below F3box.

Also had Paradigm Servo 15. Very nice subwoofer. Several Velodynes over the years. Liked them a lot.
So yes, I'm a believer in MFB.
And I want to get back in the game. A couple of of MFB subs at a reasonable cost would definately find their place in my current home. DIY is fine with me. Building a pair of small boxes is no problem.

So, will you offer complete kits in the future with 10-12-15 inch woofers..?

No commercial plans for large systems at this time no, mainly because the biggest gain with servo remains with making small sound large. If you are after more SPL just daisychain a few 8 inchers and be prepared for some raised eyebrows. I know Hypex was working on a mfb addon to their fusionamps, you might want to check with them for higher power systems.

Please tell us about the future plans....?

CH from Sweden..

I do have some rcf L18P200ak which I eventually like to try with a pair of ClingOn accelerometers but am way to busy building small systems for now. Stay tuned

 

[racerxnet]

Dunno for sure but these used a pickup coil ? if so that's probably the reason why they became instable as their loop is kinda temperature sensitive due to transformation effects between the driving and pickup coil. If they use accelerometers it's probably electrolytics gone bad causing unwanted phase problems below F3box.

No commercial plans for large systems at this time no, mainly because the biggest gain with servo remains with making small sound large. If you are after more SPL just daisychain a few 8 inchers and be prepared for some raised eyebrows. I know Hypex was working on a mfb addon to their fusionamps, you might want to check with them for higher power systems.

I do have some rcf L18P200ak which I eventually like to try with a pair of ClingOn accelerometers but am way to busy building small systems for now. Stay tuned

Has anyone ever looked at the Genesis woofers with servo feedback. These are getting old and wondering how we might upgrade them with the old enclosure.

 

[motoindo]

Genesis woofers

Pictures ? Which model ?

 

[racerxnet]

Pictures ? Which model ?

They are the Genesis 2 model. This is just a pic of the speakers, and mine are not for sale.


https://www.google.com/url?sa=i&url=https://www.usaudiomart.com/details/649142475-genesis-2-speakers-used/images/800472/&psig=AOvVaw0-2qRGJL-KaP5pVI2Ir7Gl&ust=1642636926913000&source=images&cd=vfe&ved=0CAgQjRxqFwoTCKi68srBvPUCFQAAAAAdAAAAABAD

 

[retro]

Has anyone ever looked at the Genesis woofers with servo feedback. These are getting old and wondering how we might upgrade them with the old enclosure.

If they have foam surrounds, I'm sure they can be replaced.

If you have problems with the servo, I think you have a couple of options:

Contact Paul at PS-Audio. He's the one who designed the servo system. I'm sure he can either help or point you in the right direction.

Or..

Replace the woofers with Rythmik Audio drivers and amps. They sell the 12" woofers and amps as DIY kits.