HK filters and numerous amps are unnecessary. In order for HK filters to work the way they are intended, you first have to linearize a minimum phase device. In contrast, I've already shown how to do the same thing with passive filters (and a single amp) and minimum phase target response curves in my thread on fractal arrays, which are generalized (can incorporate vertical and/or horizontal directivity).
Fractal Array Straight CBT with Passive XOs and no EQ
It's slow to start because bad assumptions were made by those trying to verify the design concept, but it gets there eventually. There's even a how-to on post 58 Fractal Array Straight CBT with Passive XOs and no EQ
I certainly don't have that problem with my arrays. My arrays are nested, not side-by-side. My tweeter is about 2.5" tall, and the array grows from there. Many have doubted the coherence of the array... until they heard it.
I agree.
My current project is based on a similar idea. Multiple amps and HK filters would be nice, but I can get very close using passive 3rd order filters.
No worries - I am not talking about a beginner's project at all. I am talking about an option to provide CD (=title of the thread), matched along with the intriguing option to cancel/attenuate floor/ceiling reflections as well. Because of bbutterfield's / brett's comments.
Horbach-Keele has been developed for professional use (Harman/JBL), has been patented and will for most hobbyist's motifs be a too expensive/complex way to provide CD, or in the words of the authors, a constant beam within a vertical array. But despite the patent and the complexity, as a non-commercial DIY-er you are free to try this one out at home. I dont know of any commercial implementations outside of JBL by now, which may be attributed to the fact that there is this patent.
And by the way(s), my test array is not (symmetric) 4-ways WMMTMMW, it's an asymmetric 5-ways WMMTMMW[W], and should be even a much huger 6-ways if you want a really deep bass extension with dipole characteristic. The 5th way is a non-CD, 65Hz-140Hz W-Dipole-Bass on the floor, which will not cause any floor reflection due to the vicinity to the floor. So this approach might be vertically CD from 150Hz to 1.5kHz, above 1.5kHz you may no more deal with a dipole configuration for CD, but with a suitable waveguide. In the transition region from 1.5kHz to 3...4kHz you might expect some beam irregularities, though. I did not measure that.
A boxed monopole configuration (as in the original paper) would make bass extension much easier and the look much more conventional.
No problem, linearizing the minimum phase devices is easily done by DSP. Have a look e.g. at Uli Brüggemann's Acourate, which also comes along with a straightforward and elegant HK filters generating option (Introseite - AudioVero).
I have read your thread, and I did appreciate all the simulations. Still: Do you have set up a working prototype, and can you post some measurements to check the real outcome? Would be most interesting. I have the strongest interest in archieving similar results as in HK, but with a leaner approach!
[Off-topic answer:] Right. You are welcome to post in open forum mode, e.g. visible to all whatever you would like to in reply to my input(s). It might hopefully enrich the discussion and be of interest to all others also.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.