|
Home » Sponsored » Pi Speakers » H290C Horn/Waveguide (Upgrade for the obsolete Eminence H290)
|
|
| Re: H290C Horn/Waveguide [message #73549 is a reply to message #71929] |
Thu, 09 August 2012 10:38   |
 |
Wayne Parham
Messages: 18789
Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
Some have asked why it is important to maintain 90° beamwidth in the horizontal, and why 100° or more is less desirable. For example, one of our friends here recently evaluated another similar waveguide/horn with ~100° coverage. The other horn was touted as a 90° waveguide, but all measurements pointed to it having at least 100° beamwidth in the horizontal. It became somewhat of a point of contention, with the other horn/waveguide proponents reverting to an argument of "how much does it matter anyway?"
So what is the difference between 90° beamwidth and a little more, say 100° or 110°? What's wrong with a slightly wider pattern in the horizontal? How much does it really matter?
If the speaker were outdoors or in a very wide open space, I would find nothing wrong with the wider pattern. With no boundaries to worry about, I almost don't care what the coverage pattern is. But indoors, in a home hifi environment, the whole point of controlled directivity is to reduce early reflections from the nearest walls. I mean, why bother with having a waveguide/horn at all if your coverage pattern allows a significant amount of energy to be reflected off the side walls?
The sound radiated by a horn/waveguide doesn't end abruptly at the beamwidth "edge". Beamwidth is defined as the angle between the directions on either side of the forward axis, at which the intensity of the sound pressure level drops to one-half the value it has on the forward axis. That's only -6dB down, so sound definitely radiates outside this coverage pattern.
The goals are to provide uniform coverage within the desired beamwidth and to limit radiation outside this angle, to reduce unwanted reflections. So if I were to have to choose between being a "smidge" inside or outside the goal, I'd definitely lean towards being slightly tighter than wider. I'd rather have constant directivity through an 80° horizontal arc than 100° beamwidth for this application.
One shouldn't go too far with this, because too-narrow beamwidth is unnatural too. The square corners of our living spaces make 90° beamwidth a very natural coverage angle. But again, we certainly don't want coverage wider than 90°, because that increases reflections and reduces the benefits of controlled directivity. So I think a good target for horizontal beamwidth is to stay in the 80° to 90° range.
This is particularly important in a constant directivity cornerhorn. The bass bin and midhorn are within 1/4λ from the adjacent walls, so there is no self-interference. The walls become part of the horn, a source constraint boundary rather than a reflector. But the tweeter horn/waveguide is further than 1/4λ, because at high frequencies, wavelengths are short. So it is important that the waveguide beamwidth be 90°, matching the angle formed by the room's corner. A wider beamwidth would illuminate the walls and cause excessive reflections.
Even in a matched directivity two-way speaker, the extra beamwidth is unwanted. When the forward axes are crossed, toe-in at 45°, the edges of the pattern run parallel to the adjacent walls, just like cornerhorns but usually a little further away. The wall behind the speakers isn't usually all that far though - often just a few feet - and you really don't want a lot of off-axis energy to reflect off that wall. Using a 90° horn/waveguide ensures that sound directed at this wall is at least -6dB, even at a grazing angle, and more attenuated at larger angle of incidence. So this is the best configuration, in my opinion.
|
|
|
|
|
|
| Re: H290C Horn/Waveguide [message #73599 is a reply to message #73597] |
Thu, 16 August 2012 10:12 |
|
Wayne Parham
Messages: 18789
Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
The H290C was designed from the start to be a bolt-on upgrade, that would require a minimum of crossover changes. The acoustic centers are in the same place and the patterns are similar, so the pre-2012 crossover used with the old H290 horn can be used and the primary lobe is forward and centered. But the upgraded H290C waveguide benefits from an upgraded crossover that has different values in a couple locations. Its impedance curve is different, necessitating the removal of capacitor C1 and installation of 16Ω resistor Rs. Check the schematic and board layout to identify and locate these components.
The π crossover uses a specialized π pad for its attenuator and top-octave compensation. It provides three functions, 1. To provide full-band attenuation, 2. To provide mass-rolloff compensation in the top-octave and 3. To provide specific damping for the core-splitter, which sets the SPL level of the crossover region independently of the midband. This network provides maximum flexibility, allowing the low, medium and high frequency ranges of the tweeter's transfer function to be set independently.
This network is created by components R1/R2 and the compression driver itself. In many cases, the output shunt leg of the π network is provided solely by the compression driver, and no extra shunt is required. But in the case of the H290C, a shunt resistor Rs is used.
The input leg R2 sets the load on the core splitter, which sets the damping. High levels of damping (low resistor values) brings down the SPL in the crossover region. Low levels of damping (high resistor values) increase SPL in the crossover region. The attenuator R1 sets the midband SPL level, and interacts with the output leg, providing mass-rolloff compensation. Optional components C1 and Rs are used to further tailor the response.
As for audible differences, remember that I always found the old H290 horn to a pretty good part, and didn't change it even when several new waveguide products began to enter the market. I found it to be better than most of the so-called waveguides out there, for one reason or another. The only thing I really didn't like about it was the edge at the mouth. I would have preferred it to have a rounded lip. But that's probably not audible, really. Constant directivity horns with sharp edges in the throat are audibly harsh and "spitty", and I think the edge at the waistbanding expansion in some CD horns also contributes to that somewhat. But it doesn't seem to be so much the case with horns having their only edge at the lip, like the old H290 had. I suspect distance from the throat matters most, and so an edge right at the mouth was least objectionable. Still, I prefer a gentle radius at the mouth, making the horn smooth everywhere.
Of course, the basic flare profile of the H290 is different too. The new H290C uses an OSEC flare profile, which is the shape that best transforms a plane wave into a spherical wave section without "fracturing" it. The wavefront expansion is at right angles to the wall at all points through the horn. Other shapes cause it to twist and bend as it progresses down the throat, making modes or "pockets" of distortion. The wavefront is slightly distorted as it travels down the throat of any other shape. So the oblate spheroidal profile is the perfect shape because it allows the wavefront to pass through the horn without distortion.
How audible is this? How much distortion of the wavefront is too much?
I don't know the answer to that question. At this point, I don't think anyone does. I know that the OSEC shape is mathematically correct, so it's at the "best" end of the scale. I also know that horns with diffraction slots in their throats are just the opposite, audibly harsh and spitty at the "worst" end of the scale. In between are the horns with other shapes.
In my experience, the biggest thing is to stay away from horns with sharp edges in them, particularly those close to the throat. If the horn is gently radiused from throat to mouth and holds the right patern without sharp edges, it's a 90% good horn, at least. The ultimate horn uses a pure OS/EC flare, and this is what the H290C is.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Goto Forum:
Current Time: Wed Nov 27 20:29:19 CST 2024
|
Posting Rules
Members
Search
Help
Register
Login
Get a T-Shirt!
