To put it rather more bluntly: sound pressure is a vector not a magnitude.
As I understood it, spl in a room ceases to be a vector below a certain F, and that is the problem with reproducing low bass. Building a pressure container and modulating the absolute pressure inside is the task. That's where the inches of marble come in handy. I guess for the rest of us we are looking for a mild impression of this without shaking the house to bits. Or a sub with a panic button is one way, recognising it is fun but you really don't want it for normal listening or programme. Different in a stadium where it presumably still is a vector.
If you want to understand what's possible in terms of "house shaking", with sufficient expertise and resources thrown at it, there's this: Why We Need Sound Above 120 Decibels
If you start with Hendrix I would agree, but if you try an Amadeus String Quartet playing Beethoven string Quartets then you are completely mistaken and out of your depth.
Have you personally set up listening to live also recording a live event then comparing using a double blind audience process ? at to what is live and what is recorded.
It is no time soon that 100% authentic percieved live sound will be available to the public
No, but what I very deliberately do whenever I have a chance, is to 'recalibrate' my hearing by sidling up to musicians playing live instruments, purely acoustically - say within a few feet of them - and trying to form a strong impression, memory of the quality of the sound. I have no interest in FR and dispersion of sound, for me it's all about the texture, the subjective impact of the sound ...
Yes, I would agree the vast majority of systems are hopeless at delivering that 'texture' at realistic volumes - mostly because it is "hard" to do. What people don't realise is that if a system can get Hendrix right, then it can also do a string quartet with correct intensity - and vice versa, the same factors are in play. You see, when listening to Hendrix what I'm really listening to is the drum kit, the shimmer of the cymbals - that has to be rendered completely cleanly, realistically - not easy for the amp to do while at the same time pumping out the growling Marshall amp ...
I am not sure where this is going. This blog is about a new speaker project around a 3 way horn system to replace Ariel.
Well, IME a speaker well sorted out, and well driven is able to reproduce music from every point of spectrum in a satisfying manner ... a fairly reasonable goal, I would have thought ...
I think it turned into something like a direct radiator bass and mid with a ribbon tweeter with a horn loaded upper midrange low treble driver added in there for good measure. I didn't realize this thread is actually going on 7 years old. 😱 Still no BTA
Yeah, I guess this is about right. 7 year itch. Like married life.
It has set me thinking I will get 1 off horn and some drivers for an audition with a mix of music. This will be enough to tell me if I am a horny. Panels are out at the present time. Neo's may change that later sometime. Probably try the ribbon in its own right and maybe as a super with the horn. More or less as you have suggested.
Maybe Lynn has some new ideas for crystallizing the project with all the recent postings.
Married life is funny- I'm enjoying it a lot more at 9 years than I did as a newlywed.
Regarding completion date- Lynn does this on his own terms, and doesn't appear to feel pressure to "get it done"- we rib him but it's a philosophical difference between him and some of the rest here- who are more goal-oriented WRT diy speakers. He has a clear goal in mind but casts a very wide net in evaluating the path. That's why this thread is so interesting- we talk about everything from mutual coupling to box alignments to dynamics and tonality, and with an experienced listener's perspective that's sometimes lacking on this engineering-oriented site.
You have to be careful there. Wrong horn, or worse yet - wrong implementation - and you'll hate it. You may end up being one of those guys who goes all over the web saying "Horns Suck!" when in fact you just heard a bad setup. 😉
The crossover will play a major roll in this.
Pano
You should repeat this comment on every page where people are going to make their own horn systems. Working with horns is extremely difficult even for the best among us. I could not do an adequate job with just calculations. It wasn't until I got a computer to do the crossover that everything came out right, but then I had to write the software to do that because nothing out there could do it right. I have heard more bad horn setups than good ones. Bad seems to be the standard and good is the exception.
You miss the fact that home hi-fi doesn't need to be played at 'realistic volume' because you ( me ) miss the sight of the venue so the brain cannot make the ratio intensity/ distance
Oh- I meant 'live volume', but some may confuse the perceived sound that produce that emotion with the real etc etc.😉
Good audio reproduction works at any sound volume, it can be whisper quiet or starting to rattle the windows - what alters, subjectively, is the intensity, or impact of the sound -- for me, the volume control should be almost irrelevant, it's merely a way of adjusting the level for convenience to suit where one's listening from, and the mood of the moment.
Less than optimum audio usually has one or some of the following characteristics: it only sounds 'right' at a certain volume, softer it's bland, louder it starts becoming irritating; or, it sounds 'good' from a distance but the closer you get to the speakers the more various unrealistic or unpleasant artifacts are noticeable; or, it sounds great at low key, 'nice' levels but once a certain volume level is reached this quality starts to disintegrate and a steady progression into becoming a typical PA system starts to take place.
Less than optimum audio usually has one or some of the following characteristics: it only sounds 'right' at a certain volume, softer it's bland, louder it starts becoming irritating; or, it sounds 'good' from a distance but the closer you get to the speakers the more various unrealistic or unpleasant artifacts are noticeable; or, it sounds great at low key, 'nice' levels but once a certain volume level is reached this quality starts to disintegrate and a steady progression into becoming a typical PA system starts to take place.
The sound at the recent RMAF underlined this point. There were a handful of simple 2-way direct-radiators that sounded pretty decent, but the more complex the system, the worse the sound.
The most expensive and absurd-looking speakers, as a rule, sounded the worst, and in a way that's hard to describe, except weird-sounding.
I keep being drawn back to a basically 2-way system because it's hard enough to do a 2-way; Dr. Geddes' point about crossovers cannot be repeated enough. I have a minor degree-of-freedom because the AH425/Radian 745Neo (whether aluminum or beryllium) does not require in-band equalization, and the Altec/GPA 416 and the AH425/Radian 745Neo have smooth off-axis behavior in the 500 Hz to 1 kHz region, which simplifies the crossover design.
Things get really, really hard if there are nulls in the polar pattern (from the drivers) or significant ripples anywhere close to the crossover region. My hat's off to Dr. Geddes for one of the most sophisticated systems I know of.
If you read the Karna amplifier page you'll see a set of guidelines, not a kit, and I warn the reader this is only for experienced builders that have already built several amplifiers of their own. If you want a kit, look elsewhere, and you want transistors, there's an entire forum here at DIY just for you.
The 20-year-old Ariel is a far simpler system; it was aimed at owners of direct-heated triode amplifiers who wanted a loudspeaker with adequate efficiency and balance similar to personal favorites of mine, the Quad ESL57 and Spendor/BBC monitors. The starting point was a minimonitor with expanded dynamic range, and improved bass from the twin transmission lines in the enclosure.
I'm satisfied with the results of the Bjorn Kolbrek/Jmmlc-designed and Martin Seddon-built AH425 Azurahorn. Very flat across a 50-degree arc, with no zeros in the response on or off-axis, and very fast decay characteristics, comparable to ribbons or the better grade of soft-dome tweeters. It was originally designed for the Altec/GPA 288, and sounds very good with that driver, but I prefer the Radian 745Neo.
The two big issues with a straight bass horn are: the very large size required for a 60 Hz cutoff, and the time delays associated with a long horn. A corner or wall location is not possible in my house, and if it doesn't fit in my living room, I'm not doing it. The time-delay issue is non-trivial as well. The existing 700 Hz lowpass filter delays the bass channel by about 8 to 9 inches, and by coincidence, this leaves the lower edge of the AH425 hanging slightly over the front edge of the bass cabinet, right where I want it.
A bass horn creates additional time delay for the bass channel. The three solutions are:
1) Move the AH425/Radian745Neo back by 2 or more feet.
2) Leave the AH425/Radian745Neo where it is, and use digital time correction to delay the MF horn system (multi-amping and digital crossovers are implied in this solution).
3) Ignore the problem and adjust the frequency-based EQ in the crossover.
Solution 1 makes a big speaker bigger. Solution 2 defeats the original purpose of a loudspeaker intended for purist all-triode triode amplification. Solution 3 is the original, old-school Klipsch approach ... ignore the time delay, and glory in the low IM distortion of an all-horn system. That has implications for stereo image quality.
I am looking more closely at a 2-woofer approach, which increases efficiency by 3 dB and headroom by 6 dB, as well as offering a method to smooth out the null created by the floor bounce. (If the centerline of the woofer is about 18" off the floor, the null falls right at the crossover frequency, 700 Hz.) Yes, there are the additional reflections off the rear wall and the closest side wall, but in most loudspeaker locations, the floor bounce arrives first, is the strongest reflection, and offers the least opportunity for acoustical treatments.
Last edited:
I wish people wouldn't do crappy web pages for this sort of thing - Firefox went away and sulked for about 5 minutes, trying to retrieve all the detail on the first page ...