Originally Posted by SteveCallas
200-250hz isn't as low as some of the other 3 ways out there, but still a bit low. A single quality 8" might well be capable of doing some really nice things down to even 50hz, but with really low crossovers, more and more 3 ways are actually like 2 ways in a standard 80hz crossover system. In a really low crossed 3 way, the woofers contribution to the cost isn't even with their contribution to the playback, sometimes only a single octave.
This is not any type of critique on your speakers, I've never heard them, just an observation on a design trend I am seeing more and more.
Originally Posted by SteveCallas
A ~200Hz highpass on an 8" midrange is "a bit low"?
Only if one assumes a highpass on the mains
In a standard 80hz crossover system, at what point do some of these 3 ways really even benefit from being 3 ways? The new SVS speakers cross over to dual 8" woofers at 160hz - just one octave of range for what probably amounts to a lot of cost. Similar, but not as bad, on these. As for highpass on the mains, it is extremely difficult for a LCR to handle bass as well as one or a few stand alone subs, for a variety of reasons.
Not trying to stir anything, just wondering why it's done. How about a tweeter handling 3 octaves, a mid(s) handling 3 octaves, a woofer(s) handling 3 octaves, and a sub(s) handling 3 ocaves. That will cover from 20khz to 10hz, each driver or set of drivers contributing equally - dynamics, sensitivity, and distortion would all benefit - just as a generalization.
I see you've recently turned your interest from big subwoofers to speakers. Perfectly valid questions there.
There are really 3 or more factors contributing to such choices. As is always the case, how a designer prioritizes different performance aspects plays a big role as well. DS-21 likes the approach Earl Gedees suggests which while insightful, I don't feel is the best use of resources and devices in real vs. theoretical systems.
Here are the questions I would pose that were not included in your considerations:
1) How does the 3 way design benefit the midrange?
2) How does the 3 way design impact off-axis frequency response with respect to consistency in response and off axis hot-spots (aka lobes)?
3) What are more likely characteristics of the sound being reproduced with respect to peak demands, average levels, and crest factor vs. bandwidth?
Addressing question 1 I would flip your question of what the woofers add in a 3 way to look at what does the high passing of the midrange benefit to the midrange reproduction? Just as with subwoofers, excursion and power is key. The midrange is directly responsible for maintaining clarity and differentiating distinct sounds from each other in complex music and movie scenes. When the lower frequency content in their range push midrange drivers beyond their most linear limits the rest of the range gets modulated by the driver non-linearity resulting in both compression and distortion. Just as a compact bookshelf will suddenly sound more effortless and clean when crossed to a subwoofer, a midrange will be improved when it is always operating well within its linear limits. For the simple case of a sealed midrange, every octave of lower extension requires 4x the excursion or gives you 12dB less output potential for each halving of frequency. Obviously what is considered "low" for a given driver depends greatly on size, excursion capability, and sensitivity. While the 8" in the Catalyst is linear a little further, just +/-5mm of excursion can produce 128dB @1m half-space around 300Hz.
Issue #2 above is most often what conflicts with #1. If it was only about efficiency and output we would all use pro 12-15" drivers and huge horns. That can be a viable solution as well, but it becomes important to pay attention to beaming of each part and how they interact off axis due to the spacing between parts. In simplified terms, for a given crossover frequency & type, the greater the spacing between elements, the more significant the off axis gyrations in response. As the crossover frequency is shifted lower for a given spacing the off axis gyrations are reduced. Off axis peaks go away once the array or column of drivers is less than about 1/2 wavelength. Of course crossovers are not brick walls, and there are many shapes possible with many different interactions, so rules of thumb really are just guidelines to steer detailed models and real measurements to confirm them. With lower crossover points it is possible to transition from one device to the next with the behavior staying much closer to a single point source vs. a mass of separate elements. In the case of the Catalysts, moving the crossover to the woofers up about 2x to 500Hz would create off axis notches in the axis of the woofers(vertically) and likely some peaks nearly straight up and down which would then bounce around the room and further color the sound. Most speakers do have such off axis holes and peaks in response, with the worst offenders typically being described as tricky to set up or very sensitive to placement. Remember that in real listening rooms we are listening to the sound bounced around the room and back to us as we are the direct sound from the speakers.
There are some cases where designs will use this narrowing for certain goals such as with line sources, but the behavior is only textbook or ideal for a limited range, where the hurdles in using them come from the behavior at the transition points from different modes of operation and to higher or lower frequency devices in the system. Smooth transitions off axis combined with plenty of dynamic capability and clarity on axis was my goal for the Catalyst 12C. That is not to say less well behaved solutions can't or won't sound good, but they will have a more complex and less predictable interaction with real listening rooms.
Once #1 & 2 are considered, then there is the question of dividing up the frequency range to balance both needs. While in most cases we do settle around an 80Hz crossover, the Catalyst 12C with it's 12" woofers was designed with 50-100Hz crossovers in mind, and I've set up systems crossing them anywhere from 40-100Hz based on measurements in-room and the requirements of the space. A 24' x 34' x 12' room requires much more midbass headroom than a small living room. More important is the reality that the 50Hz to 300 or 500Hz range can be very heavy in content and dynamic peaks. This is a range where much of the perceived power and "big sound" comes from. If this range is compressed you loose all the impact and attack of a kick drum or bass guitar with a thinner sound resulting. While a range of 80-250Hz might seem like a waste for big woofers, watching the woofers dance during demo music and movie tracks tells a different story. Even with 60-80Hz crossovers and some spirited listening levels I've seen dual 12" woofers jump 10mm each way on transients. A single 8" woofer that also needs to cover the midrange will never match that type of air movement, and certainly never do so without compromising and modulating the all important vocal range. Of course this also circles back to the differences of typical playback in a dedicated basement theater vs. a living room setup. With treatment, sound isolation and often larger spaces it is common to easily listen 6-15dB louder than what would seem appropriate in a living room/great room.
In the end it's always a balancing act with driver sizes, headroom, off axis response and of course fidelity of the parts being used. I hope that clarifies the choices behind the Catalyst a little.