Directivity of FR drivers

I always drool a bit when Dave shows off the Castle pics, as I find I'm liking the sound of an upward-firing driver. 4 drivers though, so I've been experimenting with 2 and a way to reflect part of the sound forward, vs all of it going up.

If you're after a proven design, versus DIYing "a small FR crossed low to a helper woofer", I'd bet those Castles would do what you want.
That's probably because you like a lot of reflections, like open baffle.
 
That's the object behind all these designs: to increase the ratio of reflected to direct radiation. Possibly more so than a dipole actually, since one of the salient advantages of those is (or can be) a reduction in aspects of room influence due to the figure 8 radiation pattern & lateral cancellation. Depends on usage & positioning of course. Same applies to any design with lateral firing drivers that aren't BW limited to the LF and closely boundary-loaded.
 
FWIW, my most favourite of the half dozen or so “Castle” builds were with the forward sloping top panel. With a placement of between 4-6” from back wall, no near proximity to side boundaries, toe-in of between 5-10dg, and approx 60” to ceilings in my standard height room, the blend of direct to reflected sound adds to “soundstage / ambience” is subtle enough for my needs.
And FWIW, my current pair is actually an MTM with a pair of discontinued CSS wideband mid woofers and Mark Audio soft dome tweeter; so yeah, there’s a crossover involved.
 
Good point, I haven't. It's a large and unusually shaped room. It's about 5x5m, and the roof slopes from about 2.5 m on one side to 5 on the other. The speaker fronts are able to sit somewhere between 1.2 and 1.5m from the rear wall. I'll only be in this house for another 8 months or so though, and will likely have a more average room after that.
 
To date, of the many, many, many FR boxes we have built, one, just one, used a filter. They were $10 each 8” whizzer cone drivers.

(just a small sample)

dave

You don't understand what I mean.
There's more kind of filters than just electrical ones!
You have to take the mechanical and acoustical filters, that every speaker has, intro account as well.
Read my post.

The fact that every speaker is bandpassed, they can't reproduce very high frequencies or very low frequencies, means they have a build in lowpass filter and a build in highpass filter. And if the frequency curve between these cutoff points is not ruler flat in a reflection free enviroment, it has extra bandpass filters build in as well. This includes every speaker ever made.
 
That's what Chris is referring to as far as LF loading is concerned. Above which you will note he made specific reference to direct / reflective configurations; the speaker in question significantly increases (by design intent) the ratio of reflected to direct radiation, as do related bipole types; as a semi-omni the room thus tends to have a rather greater influence upon the end result to higher frequencies than a typical monopole.

Re filtering, as I noted in post 20, while a drive unit is a bandpass device, the term 'filter', though technically correct, is not the usual nomenclature most system designers employ, and the use of TL action in the emitting surfaces is not typically described in terms of 'filters in the signal path', which the majority tend to apply to additional components, physical or mechanical, specifically included in a given design to modify the response away from an initial set of behaviour characteristics.
 
Last edited:
To a point, but you are talking linguistic semantics. We all know exactly what Dave means, and while technically correct, what you are describing is the innate BP characteristic of a drive unit, and the majority of people, when speaking of filters, are referring to additional components used to reshape the frequency, impedance, phase responses away from a base characteristic. That does not imply they are ignorant of the fact that loudspeaker drive units are BP devices.
 
I'm not just talking about the bandpass characteristics of speakers. Every deviation from a ruler flat frequency response, in a reflection free enviroment, is caused by some kind of a filter.
How else can it happen? Magic does not exist, its just how nature works.

You can think this is a horrible thing, but its not.
You can add all these kind of filters together, to make a speaker with no resonanses. Example: diffraction products, acousical filter, can be compensated for with electrical filters. Its standard procedure for every capabele speaker designer.
 
The second sentence in #34 above certainly demonstrates the power of personal truth over verifiable fact.

Of course “filter” is one of those delightful English language words that can be either a verb or noun.

But I think we’re getting too entangled in our pedantry here
 
Last edited:
I'm not just talking about the bandpass characteristics of speakers. Every deviation from a ruler flat frequency response, in a reflection free enviroment, is caused by some kind of a filter.
How else can it happen? Magic does not exist, its just how nature works.

You can think this is a horrible thing, but its not.
You can add all these kind of filters together, to make a speaker with no resonanses. Example: diffraction products, acousical filter, can be compensated for with electrical filters. Its standard procedure for every capabele speaker designer.

Hi, electronic filters can't undo diffraction. Diffraction effects are different to each direction.
 
I'm not just talking about the bandpass characteristics of speakers. Every deviation from a ruler flat frequency response, in a reflection free enviroment, is caused by some kind of a filter.
How else can it happen? Magic does not exist, its just how nature works.

You can think this is a horrible thing, but its not.
You can add all these kind of filters together, to make a speaker with no resonanses. Example: diffraction products, acousical filter, can be compensated for with electrical filters. Its standard procedure for every capabele speaker designer.

Except for the fact that that is not what Dave was referring to. He was referring to additional components used to modify a response from a given characteristic. Simple as that.