Rerurn to Romy the Cat's Site
In the Forum: Horn-Loaded Speakers
In the Thread: Bill Gaw: over 50 years of high-end audio experience and time aligned horns.
Post Subject: Alternative History FictionPosted by IronManIV on: 6/15/2025
Can we imagine what Hilliard, Lansing, Olson, Thuras and Wente would have done if we had offered them a technology that could time align the amplified signals for the Shearer horn? What transducer and horn topology might they have built or kept in production if they had that. Would they care that it was digital?
It would seam that digital signal processing and horns were made for each other. Time alignment in the analog domain for horns is not for the faint of heart or real estate constrained. Just trying out a large horn in one's living room could be a daunting task.
I am quite amazed at what can be done with a $600 3 channel Hypex D-Class amp and DSP processor. Add a Hypex plate amp for subs and bass, and now 4 channels with full DSP control all the way through FIR filters of modest size. And, the revelation that all the relationships between Xover slopes and phase slippage and group delay are 1 to 1 the same as what passive filters are doing. The Laplace and Fourier transforms are inviolate. So, a design can be ported, even piece by piece back into analog topology. Except the interaction between the amplifier and filter to the reactive component of the voice coil impedance. There is still plenty of fun to be had for analog enthusiasts to do complex impedance optimization for ideal driver behavior and the work of people like Harvey Fletcher on long distance telephone lines can still inspire us. If only we could do the math like these Phd's who laid the groundwork for our methods.
Tools like Aura and Trinnov etc. which can bring raw processing power to do statistical averaging of polar responses, delays and comb filtering to trim delays and slopes to a least error total room or total power response reverberant field are quite amazing feats of technology. It is not surprising the results might be truly emergent or unexpected.
Jean Michel Le Cleach suggested in 2004 at ETF2004 that the off axis, reflected or ambient sound was important in crossover design and suggested 3rd order slopes in phase quadrature with additional delay of the woofer channel could improve sound. And as far back as 1986 Vanderkooy and Lipshitz recognized that better power response and sound quality could be achieved by some sacrifice of on axis response flatness. These folks had to propose these methods with rigorous mathematical models since without them there would be no way to implement the methods in the early dsp solutions. A phd in math or physics was almost necessary to propose these solution academically.
Today we have machines that can iteratively design slopes for a given room and drivers without any restriction to symmetrical models and analytical solutions like Butterworth, LW or Bessel. The Aura or Trinnov machines can move the slopes around to infinite shapes including assymetrical arrangements and hunt for the best overall solution for the reverberant field and if necessary run the whole kit through an all pass FIR to flatten phase. (analog fanatics have had coronaries at this point thinking of all the signal manipulation
None of this would be very interesting for the horn enthusiast or hobbyist, except that smart guys in Europe and Scandinavia (contributions from Amerika are few as we've decided that higher education is a conspiracy) are writing incredibly good software that incorporates cutting edge signal processing and acoustics research into easy to use tools that are often FREE. Tools like REW, VituixCAD, Rephase, etc. The hobbyist can do a poor man's Aura with these tools and several Hypex amps.
Of course we would not know these tools might yield such pleasant results without the open minded observations from good sound fanatics like Romy and Bill who can compare these results to the most sophisticated analog systems. What a pleasure it is to get these reports from the frontier for without it, we'd be lost in an exponentially growing field of practical options.
JamieRerurn to Romy the Cat's Site