|
High Efficiency Speaker Asylum Need speakers that can rock with just one watt? You found da place. |
|
|
|
High Efficiency Speaker Asylum Need speakers that can rock with just one watt? You found da place. |
|
In Reply to: Re: Horn stuff and a post back for Tom posted by Mark Seaton on July 9, 2002 at 12:53:43:
Hi Mark!You wrote:
> > Upper referrs to the loudspeaker(s) covering the frequency range
> > above the lower drivers. This could be the compression driver
> > (upper) and midranges(lower), but could just as well be the
> > midranges and the bass drivers, or any two sets of drivers
> > covering adjacent bandwidths.I assumed that was what Tom meant, but wanted to be sure since you have an array of midrange devices.
This was in the context of the delay between devices, and the physical offset. So what is the smallest physical offset between the compression tweeter's diaphragm and nearest midrange ofifice? What is the largest? This would be a measurement taken between edges.
[about subsystem delays at different frequencies]
> > Maybe Tom should have been clearer with his statement in that it
> > is the acoustic response which we are interested in. In the
> > acoustic response, he means the acoustic output from the system of
> > crossover, driver and horn.Yes, which is different at different frequencies.
> > Remember the driver/horn combo is a mostly minimum phase system,
> > and the combination of the horn, and the power response of the
> > driver form a bandpass output.Minimum phase does not mean perfect phase. This defines a type of system, that is all. Also, complex systems made of individual "minimum phase" systems do not perform the same as a single one would. Third order filters are essentially three first order filters in series, but they have different behaviour in the time domain as well as the frequency domain.
> > The electrical filter is then adjusted to compensate for any
> > response variations, which by definition corrects the phase of
> > this minimum phase system.It is not "corrected." That is my contention here. I'm sure that Tom has done excellent work, and that his solution optimizes performance in the time domain. But it is not "corrected" and more accurately stated as "improved."
> > The "crossover" is not so much matching a textbook filter response
> > as it is just seeking to make for a combined response which has
> > linear phase.I understand that. But my point is that it isn't possible; The whole system forms a complex filter. That filter is either optimized in the frequency domain (meaning that it is flat) or it is optimized in the time domain (meaning that it is time aligned). Your claims are that it is both, simultaneously and I take issue with that.
> > If the acoustic phase is near zero, there is no change in delay
> > with frequency/wavelenght. This is a known benefit of a horn
> > system over a direct radiator which operates as a constant
> > acceloration device having a general midband 90 deg. phase lag.
> > On axis, the same holds true of an electrostat.Again, you have some components in your system that have fixed delays, and other components that have moving delays. So an offset isn't sufficient, nor is a simple reactive circuit.
[about my comment that the bypassed attenuator provides a "strange (relatively) linear phase shift."]
> > With higher resolution measurements and accurate measurement of
> > the real impedance of the compression driver it requires a bit
> > more complicated circuit than a bypassed attenuator to work as
> > described, but the concept and end goal are the same. The resulting
> > mostly linear phase response is exactly what is expected from this
> > minimum phase system.Key points here:
I said relatively linear phase shift. I did not say "time aligned" or zero phase shift. The filter I described generates a relatively constant 45 degree phase shift over the span of an octave. But even if it were a perfectly constant phase shift of 45 degrees, that still couldn't be corrected by a physical offset. The distance that is equivalent to this 45 degrees would be twice that at the lowest frequency of the octave span as it would be at the highest frequency.
You have also used words that indicate ambiguity, saying "the resulting mostly linear phase response is exactly what is expected from this minimum phase system."
> > Since we could examine the midrange section as a separate conical
> > horn with its apex equal in area to where the entry holes are
> > located. Please see the attached picture of the inside of the horn
> > corresponding to the other rendering you are inquiring about with
> > the entry holes. Since the distances are acoustically small with
> > in the 60x60 pattern of the horn, all four midranges acoustically
> > couple.And what is the distance to the compression driver? What is the width of the overlap region? It would appear that this distance is larger, and from earlier descriptions of the system, I gather that the overlap region is fairly wide.
Minimum phase systems are, by definition, equivalent to filters having low order slopes and very wide overlap.
It is easy to understand the principle behind the Unity horn. Basically, you want to place the drivers as close as possible, and synchronize phase as well as possible. Then, you limit dispersion so that the parallax which would cause diffraction is outside your dispersion boundary. That much is easy to understand, and this is has merit.
But the claims of the device go much further. It is claimed to act as a point source and to acoustically load the drivers from 200Hz to 20Khz. Why then would you need compensation circuits for frequencies above the vocal range? And what makes the fixed delays of the offsets and the moving delays of the electronic components become even approximately equal in the wide overlap of a low order filter?
That's what this discussion is all about.
Wayne
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: Horn stuff and a post back for Tom - Wayne Parham 07/9/0216:47:53 07/9/02 (0)
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
