Hi jzl,
Current drive has been tried several times. Varying damping factor was even popular in the 1950's, so we know a lot about it.
Current drive did not succeed in the open market because overall it did not perform as well as voltage drive. The same for variable damping factor. Finally, the designers of a speaker system must know something about the source of power. So they assume close to a perfect voltage source with zero impedance (although some assume a low impedance, say 0.1 ohm to account for wire and connection resistance). Knowing this they can test and design a system with a good idea how it will actually perform in the real world. They need a target.
Current drive has been tried several times. Varying damping factor was even popular in the 1950's, so we know a lot about it.
Current drive did not succeed in the open market because overall it did not perform as well as voltage drive. The same for variable damping factor. Finally, the designers of a speaker system must know something about the source of power. So they assume close to a perfect voltage source with zero impedance (although some assume a low impedance, say 0.1 ohm to account for wire and connection resistance). Knowing this they can test and design a system with a good idea how it will actually perform in the real world. They need a target.
The surround shape is intended to address mechanical nonlinearity. It is not reflected in the impedance, it is an independent variable.
It is usually (but not always) actively implemented, put simply where a device with a relatively small drop varies to emulate the effect of a larger impedance.
There is no compromise, primarily it is only resonance. (Voltage drive doesn't change this, it is still resonant.. only the level varies.)
We could (to make a point) EQ, and hence fix, resonance like this using any method, at any stage of the chain.
I agree, and that can be a different story. Speakers tend only to produce 2nd and 3rd order distortion.
D
Deleted member 375592
May I add 2c on the loudspeaker directivity?
Once upon a time, a customer of mine asked me to help them design a steerable near-field focusable 12-driver beamformer array. A simple DSP app. The application was commercial, like in the isles of a store like Home Depot, the array had to improve intelligibility for customers' ears 2-5 meters below (configurable) and may be off-center.
During development, we tested it also on music. The effect was quite dramatic. BTW, AFAIK, JBL has been producing analog implementations of such arrays for quite a long time.
Does anyone have hands-on experience using multiple drivers for high-end music reproduction in living rooms?
Once upon a time, a customer of mine asked me to help them design a steerable near-field focusable 12-driver beamformer array. A simple DSP app. The application was commercial, like in the isles of a store like Home Depot, the array had to improve intelligibility for customers' ears 2-5 meters below (configurable) and may be off-center.
During development, we tested it also on music. The effect was quite dramatic. BTW, AFAIK, JBL has been producing analog implementations of such arrays for quite a long time.
Does anyone have hands-on experience using multiple drivers for high-end music reproduction in living rooms?
Hi Mike,
That's called a line array, and it is (or was) very common for speech applications. That's if I understand you correctly. For bass frequencies, they are less directional, and for high frequency drivers we generally go for wide dispersion. The standard tower speaker using multiple low-mid frequency drivers and a tweeter fits your description, or a three way modern speaker vertically aligned. Speakers in the 70's tended to an offset arrangement to make the box smaller, it wasn't optimal at all.
One popular speaker line that uses two drivers covering low frequencies and mids, then a high frequency horn driver is Klipsch. A standard configuration might be something like a PSB Stratus Gold which uses three drivers, woofer, mid and tweeter.
A line array that used the same driver type, 6" or 8" was common, wouldn't sound very good for music. We have done that. Their response is optimized for speech, so highs and bass were attenuated. You might see 4 vertically aligned drivers in a skinny box.
That's called a line array, and it is (or was) very common for speech applications. That's if I understand you correctly. For bass frequencies, they are less directional, and for high frequency drivers we generally go for wide dispersion. The standard tower speaker using multiple low-mid frequency drivers and a tweeter fits your description, or a three way modern speaker vertically aligned. Speakers in the 70's tended to an offset arrangement to make the box smaller, it wasn't optimal at all.
One popular speaker line that uses two drivers covering low frequencies and mids, then a high frequency horn driver is Klipsch. A standard configuration might be something like a PSB Stratus Gold which uses three drivers, woofer, mid and tweeter.
A line array that used the same driver type, 6" or 8" was common, wouldn't sound very good for music. We have done that. Their response is optimized for speech, so highs and bass were attenuated. You might see 4 vertically aligned drivers in a skinny box.
Hi AllenB,
Well, direct voltage drive (active crossover) does damp resonance better if needed. Stick a crossover in there and you may have enough impedance to allow the driver to peak at that point. Hopefully the enclosure design damps the driver at that point, a Qts of approx 0.71 does the job well.
Well, direct voltage drive (active crossover) does damp resonance better if needed. Stick a crossover in there and you may have enough impedance to allow the driver to peak at that point. Hopefully the enclosure design damps the driver at that point, a Qts of approx 0.71 does the job well.
Accounting for it is what we do 🙂
The driver doesn't fundamentally behave differently with differing source impedance per se, which can be seen in the fact that the impedance divides itself a drive Votage which itself accounts for the difference in response, in proportion to the impedance peak which itself does not change. Therefore damping can be seen in the relative sense on a frequency plot in place of looking in the time domain, in accordance with minimum phase criteria.
The driver doesn't fundamentally behave differently with differing source impedance per se, which can be seen in the fact that the impedance divides itself a drive Votage which itself accounts for the difference in response, in proportion to the impedance peak which itself does not change. Therefore damping can be seen in the relative sense on a frequency plot in place of looking in the time domain, in accordance with minimum phase criteria.
Yeah, I'm quite familiar with induction motors.
The point being....induction motors are not exactly the same as dynamic loudspeakers.
They are similar, but quite different in some important ways.
Dave.
No, I don't believe that's what he meant. But, you may be right.
My point being.....that we'd already been down this road a few weeks ago in another thread and he was confused then.
All of this appears to be off-topic for this thread anyways....I guess. 🙂 A lot of discussion in this thread has been deemed "off topic" by anatech for reasons I don't understand.
Oh well.
Dave.
For the general concept of braking, they are close enough.
As for being off-topic, I have been clear. Maybe you're confused.
As for being off-topic, I have been clear. Maybe you're confused.
Hi Dave,
Moderation is done by individuals, we do not have one mind but do our best to be fair and keep threads in order.
The debate between current drive and voltage (normal) drive is heated and there are folks who will not leave it alone. This will destroy any thread. So if this comes up, any thread has a high chance of devolving into morass of bickering. Try to respect the thread topics, that is all anyone can ask.
Normally I do not entertain comments on moderation and this is not permitted if you read the rules. Given that you're an old member, and approached it politely I answered.
You should probably have PM'ed me on this.
Moderation is done by individuals, we do not have one mind but do our best to be fair and keep threads in order.
The debate between current drive and voltage (normal) drive is heated and there are folks who will not leave it alone. This will destroy any thread. So if this comes up, any thread has a high chance of devolving into morass of bickering. Try to respect the thread topics, that is all anyone can ask.
Normally I do not entertain comments on moderation and this is not permitted if you read the rules. Given that you're an old member, and approached it politely I answered.
You should probably have PM'ed me on this.
The point being is that the application is different, as the LS has the task to reproduce music.
Now, music is sound. Note the definition which is an assertion. Don't confuse it with the sound in the thread title .
Is sound ancillary to music?
And regarding music reproduction, is our brain ( animals, attention, cavern, etc.) prepared to the revolution happened 100 years ago?!
I don't think the whole package is well understood
It would take a forum for it
Davey, clearly you have a different point of view. I've heard Jim cover the standard view quite well but I'm not sure I've heard your scenario. I apologise if you've explained it before and I missed it.
hifijim said:
Very interesting subject, and nice if one could listen to the distortion only to determine which driver is best.
Listening to the results in some of the first videos in the thread, I'm wondering:
The driver (DUT) will have a more or less straight frequency response, depending on the driver. If you compare a non linear frequency response to the incoming music signal, I would think you will see and hear the non linearity in the frequency response as distortion, which it is not .... it's just difference in level at different frequencies.
The more non-linear freq response will produce the highest difference to the reference (music).
So to do this measurement you first have to ensure 100% linearity (flat) response, or shape the input with the frequency response of the DUT.
Is that no so??
And maybe tis is also taken into account already.
In my humble experience the biggest differences in sound from any system is the overall frequency response, in-room that is. And this is determined by all drivers together, not just one driver.
On the other hand I also for sure believe that if the driver measures better it will also sound better, if everything else is equal (which is very rarely is).
Listening to the results in some of the first videos in the thread, I'm wondering:
The driver (DUT) will have a more or less straight frequency response, depending on the driver. If you compare a non linear frequency response to the incoming music signal, I would think you will see and hear the non linearity in the frequency response as distortion, which it is not .... it's just difference in level at different frequencies.
The more non-linear freq response will produce the highest difference to the reference (music).
So to do this measurement you first have to ensure 100% linearity (flat) response, or shape the input with the frequency response of the DUT.
Is that no so??
And maybe tis is also taken into account already.
In my humble experience the biggest differences in sound from any system is the overall frequency response, in-room that is. And this is determined by all drivers together, not just one driver.
On the other hand I also for sure believe that if the driver measures better it will also sound better, if everything else is equal (which is very rarely is).
No, it is just not commonly identified in the coil impedance, but it is there. It is after all a power loss like any other.
What aspect is that.. are you taking about a secondary effect such as a change in, or nonlinearity of mass? ..or a change in or nonlinearity of compliance due to the different surround?
Peak SPL levels remain a significant factor in why sound reproduction fails to recreate an impression of "being there". In place of genuine peak SPLs, distortion products can provide an illusion of dynamic range too. Under-biased transistor phono pre-amps enjoyed a period of commercial success possibly for that reason.
I would also not say people "refuse" to use the tools available, just there exists a status quo in the industry that (as pointed out already in this thread) does not prioritize the issues often discussed here. Your previous comments re the development of "PMPO" and the like were correct. There are plenty of useful measures that could be done with level dependent distortion, but these would not serve well in certain marker sectors.
Likewise, channel separation is another barmy one to anyone who cares to know how stereo reproduction actually works. Purposeful comprises in channel separation can afford more accurate sound field reconstruction at the listener's ears, yet we seldom find that in any commerical "hi-fi" product.