I can think of some pros n cons for both ideas already, no need to get into it...but loading is one way to lower extension on a waveguide...the only other options begin at the driver...or like an altec 604 you can put the wave guide in front of a large woofer to sort of achieve this goal of extension on axis with the waveguide (now waveguide size is an issue)
My point is no option is without sacrifice... the only other option is in adding another woofer (for extension) which is a sacrifice in being another source off axis to the others...
My point is no option is without sacrifice... the only other option is in adding another woofer (for extension) which is a sacrifice in being another source off axis to the others...
Last edited:
I can see the allure of it, but there is still a problem that needs to be addressed and it can be a particularly difficult one to deal with.
The octave or octave and a half below 600Hz might get rough due to discrete modes from earlier reflections. Sometimes the full EQ correction (EQing to the room response) is not satisfactory with these and a partial compensation is the only thing that helps short of an acoustic solution.
So what is it we are seeing here? Is it that we are still able to discern a secondary source in this range, or does it simply come down to DI? Perhaps it's a combination.. perhaps positioning a speaker to gain a favourable mix of resonances may not be as simple as just looking at how smooth the plots become, and the result may still not be optimal.
In either case however, maybe with dumb luck, the waveguiding of this range seems to offer a potential solution.
The octave or octave and a half below 600Hz might get rough due to discrete modes from earlier reflections. Sometimes the full EQ correction (EQing to the room response) is not satisfactory with these and a partial compensation is the only thing that helps short of an acoustic solution.
So what is it we are seeing here? Is it that we are still able to discern a secondary source in this range, or does it simply come down to DI? Perhaps it's a combination.. perhaps positioning a speaker to gain a favourable mix of resonances may not be as simple as just looking at how smooth the plots become, and the result may still not be optimal.
In either case however, maybe with dumb luck, the waveguiding of this range seems to offer a potential solution.
I've found that higher directivity down low, somewhat ironically, helps more indoors than outdoors. And i mean helps a lot.
Without trying to forecast what to expect, I've simply increased the size of horns size by size....and compared.
My experience.... is make no mistake, lower pattern control indoors rocks.
(and matters less outdoors)
But getting there is totally about putting performance over aesthetics, imo.
My best pattern control comes from 48x29 inch horns...very big in a regular room for sure, but oh so worth it sound wise..(to my ears)....
Next step is 60"w x ?h" horns. (5x5 B-Birch limitation Lol)
Without trying to forecast what to expect, I've simply increased the size of horns size by size....and compared.
My experience.... is make no mistake, lower pattern control indoors rocks.
(and matters less outdoors)
But getting there is totally about putting performance over aesthetics, imo.
My best pattern control comes from 48x29 inch horns...very big in a regular room for sure, but oh so worth it sound wise..(to my ears)....
Next step is 60"w x ?h" horns. (5x5 B-Birch limitation Lol)
Last edited:
To avoid confusion, maybe I should clarify this. I didn't mean the directivity of the extension ("below 600 Hz"), but all the rest. Because this is what happens when you make a horn a higher acoustic load down to hundreds of Hz - you seriously degrade all its performace directivity-wise.
This is also what I meant when I said "I have yet to see a high-loading horn that I would consider useful." - High-loading horns beam badly and there seems to be no way around.
Last edited:
Ath allows to easily design all of this (only not all at once - at least I haven't figured out how to do it).
To me, "less mouth reflection" and good off-axis response is almost one and the same thing. I virtually don't care about loading - there has been always more than enough output for what I needed or wanted to achieve. I can't stand HF beaming so that's NOGO for me anyway.
To me, "less mouth reflection" and good off-axis response is almost one and the same thing. I virtually don't care about loading - there has been always more than enough output for what I needed or wanted to achieve. I can't stand HF beaming so that's NOGO for me anyway.
Last edited:
To get lower distortion there is possibility to change the system design or swap in better driver. But nothing can substitute the directivity waveguide provides, so choosing loading over constant directivity for the waveguide function seems like a compromise in wrong place. Of course one can do it if need to stick to a driver or a system which requires loading.
Last edited:
^^ I've found many things (audio) get a lot easier to figure out and then choose between options by thinking through the trade-offs 
Good compromise results when the trade-off is not valuable for given design goal or can be achieved through other compromises. Often the trade-offs can be pushed to cost or aesthetics, which is a win when audio quality is the main goal. Every person / project might have different priority between the cost / quality / aesthetics so loading might suit some project. Better drivers and bigger systems cost more, so maybe sacrificing the waveguide for cheaper system cost can be a good choice for some case.
Good compromise results when the trade-off is not valuable for given design goal or can be achieved through other compromises. Often the trade-offs can be pushed to cost or aesthetics, which is a win when audio quality is the main goal. Every person / project might have different priority between the cost / quality / aesthetics so loading might suit some project. Better drivers and bigger systems cost more, so maybe sacrificing the waveguide for cheaper system cost can be a good choice for some case.
Last edited:
I think that once one would like to have loading, it starts to get obvious that a horn or WG is only really good, for all parameters, for 3 octaves (said a reliable source). One can try all sorts of things but thats a reality as I have understood it. When one try to go beyond 3 octaves out of a "horn", probably it is the best thing to optimise on directivity.
Now let's return back for a while to the non-flatness (higher modes) of the input wavefront - its measurement and investigation.
I have the plane wave tube with the mics but I'm quite pessimistic about the outcome due to the fact how fast do the evanescent modes decay. Maybe what I could do "instead" would be to try to simulate it first - try different non-flat wavefronts and see what effect do various higher order modes have on the polar response of a waveguide. Could this tell us something new or interesting?
Maybe we could then somehow fit weights of the modes to match what is actually measured on the waveguide and get a description of the wavefront this way.
I have the plane wave tube with the mics but I'm quite pessimistic about the outcome due to the fact how fast do the evanescent modes decay. Maybe what I could do "instead" would be to try to simulate it first - try different non-flat wavefronts and see what effect do various higher order modes have on the polar response of a waveguide. Could this tell us something new or interesting?
Maybe we could then somehow fit weights of the modes to match what is actually measured on the waveguide and get a description of the wavefront this way.
Last edited:
From this perspective Quad esl57 measures very bad but sound is very pleasant. I think overall design matters.
Not only is this a good idea, but I think that I suggested it a long time ago as an alternative to using a PWT (although in the non-PWT case very good polar resolution is required.)
You have to use the "duct" modes and not the membrane modes - they have different boundary conditions. Do each mode, then since it's a linear system the polar response as measured can be used to extract each throat mode thus determining the shape of the throat wavefront.
As to the "loading" discussion, I also think most of it is nonsense as I am sure you all are aware. "Lower distortion"!? Really!? I thought that we had moved past that by now.
I stick to the battles I can win...which is; loading adds bandwidth to the Axis on which the loading takes place...options are limited as to how to get lower bandwidth extension on the same axis as the compression driver acting as tweeter...loading is one. Driver options for extending bandwidth lower in the register can only get you so far before loading becomes necessary to go any lower....on the same Axis.....