+1
But it might be a very cheap and easily implemented sensor, which means it might be used in small popular products such as computer speaker systems, Sonos, Siri....
B.
Yup, they married the sensor to a DSP : klaraT Development Kit | Subvo, thus no solution for analogs.
Last edited:
How do you know they are using a Cirrus Logic, the video mentions dsp selection based on customer requirement ?
The video clearly shows the usage of a feedback loop ?
Bruno P:
BP: We have been looking at the sensor thing but actually it's more expensive and less effective than actually fixing the driver. Because for instance, motional feedback - there are no practical ways that you can make the loop band of a motional feedback system much better than a couple hundred hertz. Which means that if you have intermodulation distortion, that is not being corrected by the motional feedback system. So, if you want to fix that as well, actively, you can, but it means that at that point your DSP will have to contain that particular model, so that it can actually predict what that motor will do. Motional feedback with a simple sensor is kind of a dead-end. The sensor is useful in the case of where we have a model and there is a positional parameter in there and then it's a good idea to know the position within a half a millimeter or so... and to get rid of drift. But at that point that sensor is no longer really in the loop. It's only applied to that function of the model.
I see the point of selling a driver with an amplifier, because that's how we started this project. It's just that it turned out that it was cheaper to build a driver that didn't have all those distortions. I don't think that's necessarily the case that companies will be co-selling drivers and amplifiers. Of course, it just happens that we would be terrifically pleased to do exactly that. But at the moment...
BP: We have been looking at the sensor thing but actually it's more expensive and less effective than actually fixing the driver. Because for instance, motional feedback - there are no practical ways that you can make the loop band of a motional feedback system much better than a couple hundred hertz. Which means that if you have intermodulation distortion, that is not being corrected by the motional feedback system. So, if you want to fix that as well, actively, you can, but it means that at that point your DSP will have to contain that particular model, so that it can actually predict what that motor will do. Motional feedback with a simple sensor is kind of a dead-end. The sensor is useful in the case of where we have a model and there is a positional parameter in there and then it's a good idea to know the position within a half a millimeter or so... and to get rid of drift. But at that point that sensor is no longer really in the loop. It's only applied to that function of the model.
I see the point of selling a driver with an amplifier, because that's how we started this project. It's just that it turned out that it was cheaper to build a driver that didn't have all those distortions. I don't think that's necessarily the case that companies will be co-selling drivers and amplifiers. Of course, it just happens that we would be terrifically pleased to do exactly that. But at the moment...
Ok, but this is where Bruno Putzeys talks about his Purifi driver which is kinda different from the route the sub-vo folks are taking which to me looks like a genuine feedback loop design. While at it, their unique selling points Low Force Factor Modulation, Very Constant Force Factor over Excursion and Low Surround Radiation Distortion remind me of the ones advertised by CSS in their XBL patents.
Purifi and the KlaraT both have the same goal though; reducing distortion produced by the drivers.
The thing is, feedback is not able to correct ALL forms of distortion. It's only able to correct for controllable and observable parameters.
If the non observable and non controllable parameters are the dominant cause of distortion, it only makes sense to solve those first before applying motional feedback.
What spec would you be looking for? High samplerate seems like an obvious one, but there's problably more processing in the DAC/ADC going on that causes delays.
Measuring the signal and reproducing the signal on the DAC would give an indication of the delay, but it's quite expensive if it happens to not be good enough.
You will need a codec with 5-10uS latency including adc and dac. At this moment only the ADAU1772 or 1777 can do this ( at a reasonable price).
You hear that all the time: just a matter of bandwidth. That's the basic 7th grade physics and who can disagree. Can the simplified physics model be too simplified, as Einstein famously joked.
How come I've never believed it? It is clear even on first hearing an MFB system that they are way "faster" than you'd expect. Maybe a blind test is necessary.
B.
@Ben. Such "fastness" claims have also been made for other ways of equalizing. Some time ago the Dutch version of Elektor published an article about a closed box subwoofer system. The Scan Speak woofer (W25or W30 series?) was passively corrected by LCR networks (with the typical 20mH or so coils) to equalize the impedance peak at Fo-box and a active equalizer to extend the low end of system.
Guess what? The author claimed fast, firm and controlled bass. In order to prove this, simulated tone bursts were displayed in th article. It was not at all hard to replicate these tone bursts in LspCad by simply mimicking the transfer function of the woofer.
These claims have been made as long as deep bass is reproduceable.
That being said, I am quite willing to accept that the distortion reduction might be the cause of your perception. Regular equalization/bass boost in the low end often causes a whopping distortion boost to 20-30% at e.g. 30 Hz.
Guess what? The author claimed fast, firm and controlled bass. In order to prove this, simulated tone bursts were displayed in th article. It was not at all hard to replicate these tone bursts in LspCad by simply mimicking the transfer function of the woofer.
These claims have been made as long as deep bass is reproduceable.
That being said, I am quite willing to accept that the distortion reduction might be the cause of your perception. Regular equalization/bass boost in the low end often causes a whopping distortion boost to 20-30% at e.g. 30 Hz.
With MFB you can achieve results like this if you do the engineering.
I linked to the document earlier. I guess that the high distortion level with non feedback is due to the amplitude.. And the rusults speak for itself.
You can always argue that you can use better speakers and more drivers. Yes!
Thats why I use one accelerometer and several drivers. The distortion will be even lower. I will make impulse response plots, to satisfy you
and me ;-)Attachments
Good point.
I would think most of the latency delay comes from the ADC and DAC, especially the needed decimation filters.
One way to go would use external (external to DSP) ADC and DAC with lower delay.
I´m quite surprised how unknown MFB seems to be in this community. Most know theory, few have heard it.
There are quite a few working MFB woofer and subwoofer examples out there, some very old. Some try to make a cheap woofer better, some try to impress with small size, but only a few aim at high quality lower bass reproduction.
You get less distortion or deeper bass, usually a mix of both. In the limits of chassis, cabinet and amplifier, of course.
The quality of reproduction is out of question, anyone who has heard a working MFB system will agree. I do not think there is a single closed box sub in the world that would not improve with some kind of MFB.
As all patents so far have expired, it seems odd it is not a standard option with high quality powered sub woofers. Maybe the industry does not see the need to evolve in this direction, as it might harm sales. If, for example, the MFB sub shows the same result as the 4 times larger "normal" top model, people might go smaller, which means cheaper.
Also, no patents mean any commercial build system can be copied without penalty. This might be a very interesting aspect today, as one of your first customers will usually sent the new article to the Chinese copy centers. Small means less transportation cost.
On the other hand, there have been and still are quite some products using some kind of MFB, but they do not promote it openly. Harman used it, but only to save on chassis cost.
Maybe the customer is to blame, too. If he buys the large vented sub, because he mistakenly takes the boomy sound for bass power and the huge cabinet for high value, smart quality products vanish.
Many business decisions about new products are not made by technicians and about building the best product possible. So real sound quality and low distortion may be secondary in most cases. Home cinema seems to be the ignoring bass quality too. Loud and ground shaking does not care for 50% distortion.
Most new developments and improvements come from the top of the line products and find there way down to consumer class. If the mass consumer does not see, want or need the better product, quality improvement stop.
For me the advantages of the principle are not in question. The question is, why is it not used much more?
There are quite a few working MFB woofer and subwoofer examples out there, some very old. Some try to make a cheap woofer better, some try to impress with small size, but only a few aim at high quality lower bass reproduction.
You get less distortion or deeper bass, usually a mix of both. In the limits of chassis, cabinet and amplifier, of course.
The quality of reproduction is out of question, anyone who has heard a working MFB system will agree. I do not think there is a single closed box sub in the world that would not improve with some kind of MFB.
As all patents so far have expired, it seems odd it is not a standard option with high quality powered sub woofers. Maybe the industry does not see the need to evolve in this direction, as it might harm sales. If, for example, the MFB sub shows the same result as the 4 times larger "normal" top model, people might go smaller, which means cheaper.
Also, no patents mean any commercial build system can be copied without penalty. This might be a very interesting aspect today, as one of your first customers will usually sent the new article to the Chinese copy centers. Small means less transportation cost.
On the other hand, there have been and still are quite some products using some kind of MFB, but they do not promote it openly. Harman used it, but only to save on chassis cost.
Maybe the customer is to blame, too. If he buys the large vented sub, because he mistakenly takes the boomy sound for bass power and the huge cabinet for high value, smart quality products vanish.
Many business decisions about new products are not made by technicians and about building the best product possible. So real sound quality and low distortion may be secondary in most cases. Home cinema seems to be the ignoring bass quality too. Loud and ground shaking does not care for 50% distortion.
Most new developments and improvements come from the top of the line products and find there way down to consumer class. If the mass consumer does not see, want or need the better product, quality improvement stop.
For me the advantages of the principle are not in question. The question is, why is it not used much more?
Yes, MFB with DSP is the Holy Grail goal. I've never seen actual speed tests for introducing digital processing into the MFB loop. From what I know (not much), there remains a shortfall or maybe you'd say a slowfall (but I'm just guessing).
For now, lots that can be done entirely in the baseband analog audio realm.
B.