Interessantes zum Thema Lautsprecher - Englisch

: What is the preference here for the different types of subs????? Or is this more determined by the type of venue and style of music?????

Hi

There are differences between horns and direct radiators which are not often discussed but profoundly effect the
results. Some semi random thoughts.

Horns are much easier to make "wrong" or mis use and the result is normally a non flat frequency response.
Remember one has to deal with both ends of the spectrum in its design, not just the bass end and horn design has much more "art" in it and "rules of thumb" seem to govern many designs.
A non flat response means that the "peaks" in the response by definition "ring" after excitation which colors the sound.
A vented box is simple, just look up the driver spec.'s and use one of many vented box design programs and
viola, one has a design.
Horns have many more considerations past mating the right driver, frequency range and throat
area, how big does it have to be, how to fold it up and so on.

Horns, particularly bass horns are relatively low in distortion however, what ever distortion the driver produces is added to the air-nonlinearity's contribution.
In probably the majority of bass horns the driver distortion is perhaps 10X - 100X greater than the distortion caused by the horn itself and as a result of distortion and non flat response most bass horns are poor in sonic accuracy.
Also, like so many things about bass horns, much of what is done is a result of folk lore, for example it is insane to put a short Xmax driver (like the old JBL K series) in a low frequency horn as it runs out of excursion at a tiny power level and then is grossly distorted yet countless W bins were made with them because "it is the right driver".
There is nothing fundamentally wrong with horns however and one must think in terms of acoustic size and wave
theory to see all horns do a similar thing.

Direct radiators on the other hand generally spend there lives having much more excursion and while manufacturers do not boast about it, one gets a great deal of "free sound" with direct radiators.
The "free sound" is produced at 2,3,4,5,6,7,etc times the input frequency and while it is not part of the original program material it is an every day phenomena.

Outside of a segment of pro audio, this "Free Sound" (AKA harmonic distortion) is actually considered undesirable as it is not part of the signal but there is no denying that the sound changes the sound produced.
Subjectively, judging subwoofers with say a kick drum, one finds the woofer with the greatest amount of "Free Sound" usually also has the "best" sounding attack (as the higher frequencies give an edge to the sound) where a low distortion woofer only produced the signal it was driven with.
In the role of a music reproducer though, a crossover normally routes the upper frequencies of the kick to the upper speakers. Low distortion woofers NEED a more powerful midbass section to "fill in" the hole in the spectrum that taking the distortion away left.

For a distorted woofer, more of those upper frequencies are produced by the driver and so sounds more "complete" when the sub is auditioned alone but usually worse when the real upper signals are added back in as the distortion competes with them.
I know many don't believe there is much distortion in woofers, they have the luxury of never having measured them.
If there is any doubt, I would urge you to take such measurements, you will see why mfr.'s don't provide this information.

Inverse square law.
Everyone knows that sound pressure from a point source falls off at -6 dB per doubling of distance.
While easy to picture, the first real life example of this so far as horns VS direct radiators really took me back.
We had just introduced the Servodrive TPL-1 at the New York AES convention and Tom Furlong (our salesman at
the time) had arranged with Ken Deloria. to loan Phill Lesch 4 boxes to play through.
The Dead were playing at Madison SQ Garden so when they called and said we should come "hear" the bass system
we jumped into a cab and went that evening.
Phil's normal bass rig (the low bass part) was 4 Meyer 650's and Ken and I went out on stage to "hear" what he was talking about.
At a distance of 10-12 feet or more, the horns were louder and lower no doubt but closer than that and the direct radiators were louder.. I will never forget the feeling of being on stage while they were playing, the whole place was moving up and down as result of the crowd, a weird feeling
Also I will never forget seeing Ken's Sound level meter rapidly climb until he had it right against the grill of the Meyer box and how much slower it climbed as he approached the mouth of the horns.

While the horns produced more acoustic power (evidenced by the higher SPL everywhere past 12 feet), the horn is
both much larger in area (lower acoustic power per area) and the actual origin is located at the small end of the horn.
When people are "up close" the direct radiator can be subjectively more powerful while at the same time the horn system (other than up close) can produce a higher SPL in the room reflecting its greater acoustic power.
One has to ask where do i need the SPL?

Acoustic phase
A direct radiator cannot reproduce the input wave shape (if not a sine wave).
On would think that might be important (being able to reproduce the input signal as closely as possible) but the physics
prevent it with a direct radiator. Instead, the direct radiator has an acoustic phase which is about-90 degrees midband
meaning that the output at any frequency is delayed about 1/4 wavelength (not discussing the ends of the band where more phase changes take place)
Since this delay is a different amount of "time" for each frequency, this reassembles the input waveshape into something which has no resemblance to the original signal.
On can argue that as evidenced by the popularity of direct radiators for bass that the ear must be pretty forgiving about screwing up the waveshape and pretty much makes the concept of hearing "absolute phase" a moot point.
On the other hand everywhere else in the signal chain such butchering of the signal would not be considered
acceptable. Other than in a "perfect world" music system, there are times when reproducing the exact signal, no more, no less is desirable in fact active sound cancellation depends on producing the exact inverse wave shape of the problem source.

Horns can have a big advantage here, a proper horn has "flat" response, it is flat because the acoustic radiation resistance dominates the system (between the upper and lower cutoffs).
In this range, it is possible for the horn to have an acoustic phase which is around Zero degrees.
A flat amplitude response AND Zero degree acoustic phase are the two conditions which have to exist to allow reproduction of the input wave shape.
This property can also exist in a Electrostatic speaker fwiw.
Regardless of what one is used to hearing, a horn "can be" a much more accurate reproducer of the input signal than a direct radiator.
I suspect when the LAB horn is in use many will agree.

Just as there are people with different sensitivity's to colors, textures and so on, personally I believe there are some
people who are more sensitive to phase errors, more sensitive to distortion and then there are those who are less.
Also, there is a matter of what we are used to.
I guess for me the "perfect" reproducer has no distortion, no power compression or other dynamic non linearity, no
phase/time errors, has ruler flat response from DC to Ch-5 (Ok I'll accept subsonic to ultrasonic bandwidth) and frequency independent directivity.
Now how to make one?
Cheers,

Tom Danley

Hi John

Thanks for the words.
With the wonders of dsp, one could "fix" the phase response of a woofer by measuring the acoustic phase over the range of operation (and past) and apply a correction that is the opposite phase curve but flat amplitude.
Perhaps the coolest thing about dsp is you can fool with time easily.
In the woofer, it is the fact that it must have an acceleration or mass controlled response in order to be "flat" that also dictates that through most of its band, its output will lag the input signal by about -90 degrees.
An electrostatic speaker has a near zero degree acoustic phase because the force applied is Voltage (force is the static attraction) and that is the reference.
The electrodynamic speaker's motor produces force in proportion to current, in the equivalent circuit the system consists of a resistance (the DCR of the motor) in series with a current to force converter (coil o' wire interaction with the magnet). In addition to producing X amount of force per Amp, it also
is a generator which produces X Voltage per X velocity. The more force per Amp, the greater the generator voltage (also called Back EMF).

At the instant a voltage is applied, there is no velocity so the generator Voltage is zero and the current and force are a result of the entire input Voltage flowing across the resistance.
AS soon as the coil begins to move, it generates a voltage which subtracts from the Voltage across the resistor (raising the ground side voltage), this reduces the current and the force so the rate of acceleration slows.
In a permanent magnet DC rotary motor (which electromagnetically identical), the motor speed will continue to climb, all the while the current drawn is falling. After some time, the motor speed will have reached a speed dictated by the motor generator Voltage constant and the drag on the moving parts.
Some small current is needed to overcome the air drag on the moving parts as well as bearing friction.
Should one add a load, one see's that the current increases, this increases the voltage drop across the Rdc and this
puts a lower voltage at the generator so the motor speed falls. If one could eliminate the Rdc, the motor would turn the same speed regardless of load.

Anyway, the point is that if you imagine this process with AC, where the motor never reaches full speed, one has a woofer motor. As you imagine the current waveform compared to the input voltage, one see's that the current leads the voltage just like a Capacitor would. The moving mass does in fact reflect through the conversion process as Capacitance (one reason amplifiers have to be able to drive capacitive loads).
The speakers output is tied to the moving part, in effect proportional to the voltage across the capacitance.
With an increasing frequency like any RC filter, this voltage falls 6dB/oct which is compensated by the increasing radiation efficiency (the radiator is acoustically larger with increasing F) which causes a 6dB /oct rise with increasing F.
This radiation resistance slope is purely resistance and does not impose a normal 90 phase shift associated with other filters (such as the mass slope of the driver) , as a result, while the drivers amplitude is compensated and now flat, the acoustic phase is not corrected by the radiation resistance and remains "capacitive" looking.
I have left a pretty simple picture but at the bottom end of the response, woofers systems can have an acoustic phase which goes back and fourth from capacitive to inductive several times.
For a real woofer, one can see what the acoustic phase looks like without a TEF machine as it always has the same sign (but not the magnitude) as the driver current.

There are other types of microphones but the capacitive type is a good example and nearly the only type used in
measuring.
It resembles a speaker in that is has a diaphragm and an air volume and even produces sound if driven.
There are a few differences.
The air volume is really small and the mass of the diaphragm is really low so what we call box resonance on a sealed box is really high (like at the top of the response band).
This means that in band, it is a stiffness controlled system and displacement is proportional to pressure over the entire band, the more pressure, the more displacement.
As the stationary plate right behind the diaphragm is charged to a high voltage through a high resistance, changes in diaphragm position also displace the static charge and since the plate is connected through a high resistance, that charge causes a voltage change on the plate which is the signal.
The microphone conversion is then one of pressure to displacement to voltage independent of frequency, down to some low cutoff set by the barometric leak (a small leak to allow for changes in air pressure)
Cheers,

Tom

Time to fire up the grill, its finally nice outside.

Grüsze,

Drum gehts bei den von mir hier reingestellten Texten auch nicht um sein Horn sondern um allgemeine Theorie! Phasenverschiebung des Outputs eines Treibers zum Eingangssignal oberhalb seiner Resonanz z.B.
Und warum das Horn (auch seins *g*) bzw. noch so eine (der Redaktion bekannte :-)) Konstruktion da prinzipbedingte Vorteile hat. Kommt meiner Meinung nach recht gut raus bei dem Text.

Wollte nur mal wieder einen Denkanstoß geben (für mich wars zumindest einer)

mfg
JF

p.s.: Wenn sein Horn nichts taugt im 4er Stack dann müßte man das ja bald im Live-Audio Forum nachlesen können. Wurde ja bei dem Woofershootout unter Volllast gemessen (hoffentlich richtig, nicht nur Verzerrungen und Pegel ....)