What does a speaker amp output?

[Budgeter]

From what I know, speaker cone will move back and forth to create sound based on the signal it receives.
However, I can't find a definite answer on whether an amplifier output varied DC (or AC with DC offset) or AC?

In case it outputs AC, which means there will be positive and negative voltage, Is the current direction reversed when signal become negative and vice versa? And does that also mean the polarity (+, -) are reversed? If it is, the cone will move both inward and forward from its initial position. This is when it makes me confused because if it produce AC current, why the speaker amp has positive and negative terminal.

In case it outputs DC (AC with DC offset), because voltage is varies so it is still move back and forth, but it will not move inward. But since it is now DC, how can the current direction be reversed? Moreover, many document I looked show that the polarity is changed when speaker is playing so it really confused me, such as this one: https://animagraffs.com/loudspeaker/

What is a definite answer for this question?
Below is an image illustrating what I'm talking about, "x" represent for position of speaker cone, x=0 when the cone is at rest. Sorry for my bad drawing.

 

[andreasmaaan]

In case it outputs AC, which means there will be positive and negative voltage, Is the current direction reversed when signal become negative and vice versa? And does that also mean the polarity (+, -) are reversed? If it is, the cone will move both inward and forward from its initial position. This is when it makes me confused because if it produce AC current, why the speaker amp has positive and negative terminal.

There are much more sophisticated ways to answer this question, but the simple answer is that yes, it's AC. The positive and negative terminals simply ensure that the polarity between the amp and the speaker is maintained.

Reverse the wiring at the terminals, and you still have AC driving the speakers, but it's polarity is now inverted, which will simply cause the speaker's cone to move in the opposite direction to a given input vs when it's wired in the standard way.

FYI do such a thing to a mono speaker (or to both/all speakers in a stereo/multichannel system) and there will be little or no audible effect, but do it to only one speaker and not the other(s) and you will have an auditory disaster on your hands. The + and - terminals ensure such a scenario is easily avoided.

 

[Budgeter]

@andreasmaaan Thanks for your answer, but can you explain me more on these thing?

There are much more sophisticated ways to answer this question, but the simple answer is that yes, it's AC. The positive and negative terminals simply ensure that the polarity between the amp and the speaker is maintained.

So the cone will actually move inward and outward from its original state, is that correct? As in my picture, to complete 1 period at x = 0, it will move from 0 -> 1 ->0 -> -1 -> 0.

Reverse the wiring at the terminals, and you still have AC driving the speakers, but it's polarity is now inverted, which will simply cause the speaker's cone to move in the opposite direction to a given input vs when it's wired in the standard way.

I still don't get it, if it is actually AC, why do polarity and current direction matter?
Besides, when you say that it in the opposite way, does that mean it still move in/out and only what change is order of position?
E.g: 0 -> -1 ->0 -> 1 -> 0.

 

[andreasmaaan]

So the cone will actually move inward and outward from its original state, is that correct? As in my picture, to complete 1 period at x = 0, it will move from 0 -> 1 ->0 -> -1 -> 0.

Exactly.

Besides, when you say that it in the opposite way, does that mean it still move in/out and only what change is order of position?
E.g: 0 -> -1 ->0 -> 1 -> 0.

Exactly. If the wiring is reversed, a +1 from the amp will cause the speaker to do a -1.

I still don't get it, if it is actually AC, why do polarity and current direction matter?

Well, polarity matters little or not at all when it comes to a mono speaker. But if you have two speakers running in stereo for example, for the outputs from each speaker to sum correctly, they have to both move in the same direction at the same time.

If one speaker is producing a positive peak at the same time as the other speaker is producing a negative peak and vice-versa, your ears will interpret this as a problem. The stereo image will collapse and the outcome will sound very unnatural.

If you want to try it, it can be easily done in any DAW (piece of audio software, e.g. freeware Audacity), or it can be done by wiring one of your stereo speakers in inverse polarity.

 

[Blumlein 88]

Hope this isn't confusing instead of helpful.

Method 1:
Black lead is always 0 volts. Red lead oscillates between positive and negative voltage relative to 0 volts. Speaker cone moves in and out.

Method 2:
Black lead and red lead start at 0 volts. Once signal is applied red oscillates between 0 volts and some positive voltage. While the black lead oscillates between 0 volts and some negative voltage so the speaker sees the difference between them and the cone also moves in and out.

Method 3:
Black lead and red lead start at 0 volts. Once signal is applied both red and black oscillate between some positive and negative voltage, but out of phase from each other. Again speaker cone sees the difference between them and moves in and out.

 

[Cosmik]

It might be a nice moment to introduce another little snippet of information: the voltage doesn't control cone position; it controls cone acceleration. But because sound pressure is proportional to cone acceleration, it all works out perfectly.

 

[RayDunzl]

It might be a nice moment to introduce another little snippet of information: the voltage doesn't control cone position; it controls cone acceleration. But because sound pressure is proportional to cone acceleration, it all works out perfectly.

DC will control cone position, why wouldn't AC?

 

[Cosmik]

DC will control cone position, why wouldn't AC?

DC will only control cone position against the restoring force of the cone surround etc. At higher frequencies, the primary force to overcome is the acceleration of the mass of the cone - as I understand it.

Edit: There are quite a few mechanical and electrical 'forces' interacting and transitioning as frequency varies, causing amplitude and phase variations with frequency. The passive speaker has to 'wing it' to some extent, but the DSP speaker can compensate for it all, effectively producing a pure signal->acceleration relationship.

 

[Blumlein 88]

This is a nice explanation that is short and sweet. There is a graphic if I can find it about the three regions of cone response.

Taken from johnk in this thread: http://techtalk.parts-express.com/forum/tech-talk-forum/31210-voltage-to-speaker-cone-velocity


In theory, if the voice coil were purely resistive, with no increase in impedance at high frequency due to Le, voltage would relate to current via Ohm's law and current to force. Thus constant voltage implies constant force. Howe this related to cone motion is a little more complex. The cone morion can be divided in to three regions, well below resonance, around resonance, and above resonance. Below resonance is is called the compliance dominated region. In that region the force from the driver motor is balanced by the driver suspension compliance and the driver displacement is proportional to voltage. Above resonance is referred to as the mass controlled region. In this region the force is balanced by the mass and as a result the acceleration of the cone is proportional to voltage. Around resonance things are more complex and a transition occurs.

 

[Cosmik]

There's an interesting thing that falls out of the voltage->acceleration->SPL relationship: if it is true that SPL is proportional to acceleration, it means that you can get a negative SPL generated by a cone that is still moving positively. It is the differential of the velocity that gives you the SPL, so a cone that begins to decelerate even though its velocity is positive will give you a negative SPL.

There's a particular audio writer who I think may be confused (or maybe I am). He says:

... if, for example, that kick drum on your favorite recording gets a good bash and you try to play it back at high volume, the initial impact will cause your bass driver's diaphragm to surge forward a good distance at a high velocity. This will likely cause a loss (or at least severe diminution) of control and, instead of stopping as it should when the amplifier signal driving it stops and reverses phase, the diaphragm will continue to move for some additional distance before stopping. This is called "overshoot", and it's the beginning of a whole series of things that you don't want to happen...

https://audiophilereview.com/audiophile/why-your-kick-drum-doesnt-kick-part-2.html

I'm happy that real world speakers may not behave perfectly, and there might be another factor in that the cone position may also be responsible for compressing the air in the room, but I think he is wrong on the fundamental point he is making.

 

[Cosmik]

It's all gone very quiet. If anyone does agree about the previous comment, it may go onto my list of 'audio complications that turn out to be myths':

1. Myth: digital audio produces distortion and it is guesswork between samples. Reality: digital audio produces no distortion, merely random noise like a very accurate analogue system; and there is no guesswork: the reconstruction filter lives up to its name.
2. Myth: acoustic crosstalk 'to the wrong ear' means that stereo from speakers is flawed. Reality: it's the crosstalk that gives the beautiful imaging - and it is 'true' imaging (for a purist recording made with cardioid mics and with speakers at 90 degrees to the listener).
3. Myth: speaker and room are a system, and the ear hears the combination as coloured sound. Reality: the ear (& brain) separates the room from the speaker; it hears both, but will still hear a neutral speaker as neutral regardless of the room.
4. Myth: Speaker cone mass/inertia causes overshoot and undershoot that inherently produces distortion and loss of detail. Reality: SPL is independent of cone velocity, so overshoot doesn't exist.

Some of it may seem somewhat off topic, but I think as this thread is about starting at the very beginning, it would be nice if the OP could be reassured that audio is *so* much simpler than people will lead you to believe. You could encounter myths that ruin your ideas of audio simply by asking the question at the top of this thread.

The humble stereo setup (similar to that of 60 years ago) is at many fundamental levels 'perfect', not some terrible compromise that masks all kinds of horrors under the surface. It really is as good as it sounds; it's not your imagination responding to wishful thinking. You shouldn't respond to 'FUD' and ruin your listening pleasure by believing the many myths that derive from half-formed ideas about how things work (like I used to).

You can be happy with a simple setup, and many of the myth-based 'solutions' (e.g. 'room correction') can/will make the system sound worse and lead you down a rabbit hole of misery and penury from which there is little chance of escape.

 

[Theo]

the voltage doesn't control cone position; it controls cone acceleration.

Just to be sure. I thought that the current intensity is the actual EM force generating movement in coil-magnet driven loudspeakers, but it is not strictly speaking the voltage which induces the EM field. The statement is ok though at first order scale, as voltage and current are close to proportional.

 

[andreasmaaan]

Myth: Speaker cone mass/inertia causes overshoot and undershoot that inherently produces distortion and loss of detail. Reality: SPL is independent of cone velocity, so overshoot doesn't exist.

I'm gonna think it through a bit more, but my first thought is that every time you have overshoot, the cone must then return to rest (or do whatever it's supposed to do next). To do this, there needs to be a change in velocity, hence acceleration.

 

[Theo]

I'm not sure I agree with #4. If the driver had no inertia at all, it would follow the exact shape of the signal delivered by the amp. In the example of the kick drum, it will reproduce the inertia of the drum itself perfectly. However, when considering woofers, the mass being different, there will be a small difference in the natural movement (hence acceleration) of the driver compared with a drum skin behavior. There is overshoot, and also ringing, depending on the damping. The question being : how small is the effect? Is it audible?
This is actually true of any physical system response :

 

[Soniclife]

The question being : how small is the effect? Is it audible?

Are there any motion controlled subs where the control can be turned off and on?

Some info on motion control. https://rmsacoustics.nl/papers/whitepaperMFBtheory.pdf

 

[Budgeter]

Thanks for the inputs!! I didn't think I can receive many detailed responses like that. But well..... some explanation have gone beyond my comprehension. I will try to go through everything in here and see how much can I get.

 

[Cosmik]

I'm gonna think it through a bit more, but my first thought is that every time you have overshoot, the cone must then return to rest (or do whatever it's supposed to do next). To do this, there needs to be a change in velocity, hence acceleration.

A thought experiment: suppose the cone has no surround/suspension but is on a frictionless track. After a few seconds/minutes/hours of playing music will it still be in the same place it started, or will it have shifted?

 

[andreasmaaan]

A thought experiment: suppose the cone has no surround/suspension but is on a frictionless track. After a few seconds/minutes/hours of playing music will it still be in the same place it started, or will it have shifted?

I can't see why it wouldn't be What's your view?

 

[Cosmik]

I can't see why it wouldn't be What's your view?

I haven't worked it out yet, but my gut feeling is that it should still be in the same place.

What I am getting at is that the suspension/surround is not actually responsible for the restoration of the cone to its centre position; the signal's AC-only coupling does that anyway. And hopefully it is not responsible for limiting the travel of the cone either. If so, then 'overshoot' doesn't exist because momentum/inertia is simply another side of the result of F=ma.

 

[RayDunzl]

A thought experiment: suppose the cone has no surround/suspension but is on a frictionless track. After a few seconds/minutes/hours of playing music will it still be in the same place it started, or will it have shifted?

Hypothesis:
If the first and last sample values (assuming digits, or the digitization of the wave) are zero, it will have returned to the starting position.

Or not, depending upon what confounding factors you throw into the experiment.

The dust blows forward and the dust blows back - Don VanVliet