My next speaker will be a statement project. One intended to enter cost-no-object performance but retain a modest price. I'm building it because it doesn't exist in the market and creating my own is half the fun.
Design Goals
Very high output
Effortless dynamic performance
High efficiency
Controlled directivity
Flat on axis response
Smooth polar response transition
Neutral, relaxed and slightly warm sound
Wide sweet spot
Wide and deep immersive sound stage
Tweakability
Allow for upgrades
Sculptural design
Design concept
My goals dictate that this speaker must use a large midrange driver to provide directivity control and it must have high efficiency and high power handling. In particular, it needs low compression and a smooth extended response without harsh breakup or uncontrolled resonances which show up as peaks and dips in response. The ideal driver I have found is the Acoustic Elegance TD12M:
When comparing the parameters of this driver to other 12" drivers designed for midrange use, it becomes clear that it is exceptional. The structure is similar to a subwoofer, but it has a treated paper cone and accordian surround. There is a woofer version that has a rubber half roll surround more like what we see in typical hifi drivers, but it turns out this is quite a compromise, introducing resonances. An accordian surround is a better choice. Unlike the ugly foam gasket we have on many comparable drivers, we have a neat rubber gasket. Later I'll create another post to rave more about this driver - it has no equal.
The challenge it creates is that a crossover around 1.2k is needed to avoid beaming. Few tweeters will meet my output goals, and most of them will fail to perform well with such a low crossover. The only real option is a compression driver loaded into a waveguide. I expect the polar response will match around 1.2k.
The compression driver will be Beyma CP380M which is well regarded. The B&C DE250 was also considered an equivalent, but the cost is a lot more in Australia.
The waveguide will be a custom made oblate spheroid 15". I'm planning to turn it up on a lathe and from there I may create a fibreglass mould, or simply build them both on the lathe.
What about bass?
Bass will be handled by stereo 40 Hz bass horns (not yet built). In my room I expect them to cover 35 - 150 Hz with 97 - 100 db efficiency. Their corner location and stereo setup means they don't interfere with imaging, even with such a high crossover. They create more energy in the upper octave which is a problem area with room modes. I've also found I can smooth out a dip around 70 Hz if the mains extend to about 60 Hz.
So there will be some overlap between the mains and subs from 60 - 150 Hz. This is not a problem - in reality it is an advantage. Those who see this as doing it "the wrong way" in many cases don't consider the complexity of getting a smooth response in this region. For some insight, have a look at my in-room measurements. Forget the arguments that apply only in an anechoic chamber, in a real room, more bass sources below the Shroeder frequency is usually better.
What about the other goals?
The wide sweet spot and large immersive spacious sound stage will come from an open baffle design. This will limit the extension to around 150 Hz and I may include a monopole woofer to extend down to 60 Hz. The warm natural sound will come from some frequency response shaping. I have found that this works if the drivers are well behaved to begin with.
Crossover
This speaker will be an active design, powered by Behringer DCX 2496 digital active crossover. I will be creating a rotating measurement stand to measure polar response outdoors.
Amplifiers
Tweeter - modified DSE A2760 (80 watts into a tweeter with 108db 1w1m - that should be enough)
Midrange - Behringer Europower EP2500 (650w)
Updates
If you want to follow my progress, click the "follow" button on the right.
No comments:
Post a Comment
All comments are moderated.