Adding more sources will spread the reflections among them, making them have a less severe impact (they average out the result of each specific reflection). For instance: sub A has a reflection at say 7 ms, sub B has one no earlier than 11 ms, Sub B is going to average out the reflection dip of sub A (and vise versa). Add more subs, and add DSP and you have the recipe for a multi sub setup having a much more steady low end. Forget about the size of the subs making all the difference, it's the differing locations that matters.
(yes, Sd area or volume displacement is good to have for other reasons
)
(yes, Sd area or volume displacement is good to have for other reasons
)
I made one sim with 1 point, and another sim with 15 points spread at 1"
The graph changes dramatically... I thought colors were supposed to be spl but if the Radiation mass is close enough related maybe they plot similar? I think this is volume matched, the outer shell is about at the same line.
If this could be looked at as a representation of volume matched, high and low sd....The reflective character is going to be quite different. I have no way to connect this to a real life situation as far as how much input the sources are running at.....I mean is this 1000w or 1w....Amplitude had to be lowered to the larger radiation source to match spl at the border line.....but no idea the radiation size to expect vs watts, so its kinda generalized and hard to relate to a real life situation.
Last edited:
So multiple subs, but how to glue everything together into a coherent system, given the long tractrix horns?Adding more sources will spread the reflections among them, making them have a less severe impact (they average out the result of each specific reflection). For instance: sub A has a reflection at say 7 ms, sub B has one no earlier than 11 ms, Sub B is going to average out the reflection dip of sub A (and vise versa). Add more subs, and add DSP and you have the recipe for a multi sub setup having a much more steady low end. Forget about the size of the subs making all the difference, it's the differing locations that matters.
(yes, Sd area or volume displacement is good to have for other reasons
Personally, I'd be inclined to place 4 small, bass (tapped) horns in the corners of the room.
I'd enclose the tractrix horns between 2 front-loaded 15's, or put them on top.
Like this, but bigger and deeper.
Hi,
jeah we have different thought process on it. If you draw similar images through all boundaries in room, from the source through first specular reflections to listener, you get much more lines than directly from the source regardless of source size, as long as it radiates sound to boundaries, iow long wavelength. There is six boundaries in a room and you get ~six times more lines through the boundaries than direct sound. And thats just the first reflections. You would have to reduce sound towards boundaries to get them reduced relative to direct sound. You forget to think time and directivity.
Count in time
Here is thought experiment based on how I see the issue and why you would have to first figure out psychoacoustics of it, or how your body perceives the sound in low frequencies in a room, to get bottom of it. I haven't studied the subject more than two days here on this thread anyway, here is the thought experiment:Think you have all practically all walls as sound source like in your picture, hundred woofers, true multisub. Lets take ~50Hz sound, whose wavelength is ~6 meters and lets take room so that you are 3 meters away from any wall, its big living room for example. Now, subs start to emit to all directions. The ones that are not in corner have first reflections like any speaker would, much less than the 3 meters path length difference to direct sound, say 1,5meters. So, at only 1,5meters, or 1/4wl into the direct sound we start to get contribution of the room. Now lets progress it further in time, more and more reflections come and when "sub wall" has bounced from the opposite side of room its perhaps 6 meters behind direct sound, one wavelength, and starts to interfere with the direct sound as well and all reflections prior, we now have all the direct sound bounced at least once. But the thing goes on, perhaps for duration of two, three, four, ten, more wavelengths / bounces. The question is now how you perceive it, does the first "pristine" 1/4wl duration impression dominate the perception or the later sound field with all the nulls and peaks resulting from standing waves, and you happen to be in for one for the remainder of the sound duration?
Imagine you happen to be at a null spot, does the body experience the initial wavefront hit of 1,5meters, 1/4wl? Perhaps if reflection free duration was more, perhaps 1/2wl we have a chance to get the peak sound pressure there is? but if it goes null then, whats the impression after all do we experience great impact of brief bass, or nothing, a weak bass as we are on a null?
What if we were located at a modal peak instead? we get the direct sound hit and then crazy 10-20db peak after that. Which one we now perceive, the intitial direct sound or a great Boom after that?
What if the sound we imagined was part of melody and the next note has different wavelength and now we have reverse situation, peak instead of null or vice versa? My intuition says then the initial "direct sound" becomes meaningless in comparison and the nulls and peaks dominate the perception, perceived quality of bass.
If all the walls was source like in the thought experiment perhaps the modes are averaged and we get reduced peaks and dips, and perceive fine sound quality, but its not due to direct sound got somehow more or better, but evened out modal behaviour, or whats the correct word. Perhaps its better to tune the system with DSP to further optimize, perhaps only few subs would be enough for same perceived effect so we don't need hundred.
I'm just hobbyist like you, showing different thought process how to think about the subject. I'm not sure what are you getting at as Geddes and others have demonstrated how to get better bass in small room, not sure if us hobbyist have anything to add to it other than just trying to figure out how it plays out.
Last edited:
The last thing I would do is ace all of them equal distance to walls etc. Or floors for that matter.
Correct, ultimately the room (size) and location of the (main) system is dominant.
We can theorize endlessly about the best solutions to deal with this. In the end (at least for me) it's the experience that matters.
Recently, at the audio show, I've listened to many different systems, including high tech (cardioid) setups, with and without multiple subs etc.
By far the best sounding systems were all basic 2 ways (as in the video above), free standing without subs.
The absolute best were also of the bass reflex type.
I see no one can or has explained why when I increased point sources in the sim the polar changes in a way that looks very similar to how I suggested that direct energy increases... the larger red area suggest an area in the sound field that remains closer to the velocity of the point source "surface".... Do tell 🤔🤔🤔
Apologies if this has been referred to before, but this looks the business; from the highly regarded designer Curt Campbell:
http://speakerdesignworks.com/Invictus_9.html
Geoff
http://speakerdesignworks.com/Invictus_9.html
Geoff
I cant replicate your results, did you manipulate the sliders on the bottom? they are not visible in your screenshots.
If I try to do it it looks like attached, intensity increases as expected. Polar pattern is omni in both cases because size of the array is very small compared to wavelength, with your screenshots and mine. This means sound towards boundaries is same as towards listener, in other words relation of direct and indirect sound stays the same. Things just get louder as you increase Sd, more sources in the example, until size of the array gets closer to wavelength. 50Hz is almost 7meters, ~21' and if you increase spacing of the 15 sources you start to get dipole/bipole response, increased directivity index, which is less power into room compared to direct sound. You must increase directivity to reduce sound in room in relation to direct sound.
Attachments
Last edited:
Regardless the initial Db it should reach a similar space in the sound field when volume matched. I have some theories of why the left is better but why voice them, the science is out of reach. The most I can do is try and measure a small and large woofer in a room. The way i see it, GD will improve with the larger woofer but better yet a measurement at the face of the driver vs one taken at distance, would be a better test. Which one deviates the most loses?
Oh you hate learning new things now? Just because no one seems to be well versed in topic of radiation mass size, I should just drop it? I have a passion to learn, and its an interesting part of loudspeaker design that you don't see people talking about. Thanks to being some what alone on exploring the topic, I actually ran my first Akabak sim....progress. I also am realizing the above simulation is useful because of the strong relationship between spl and velocity. How aren't you interested to know the impact of this aspect on your system?
Because I use the Sd area I need and several separate sources to cover the bass part I want. Worst I could do is have a single source play that same part, however large it may be. By having multiple sources play the same low notes, I can simply follow the advice of someone smarter than me that already told us how to do it after he had studied it extensively and was even willing to share it with us.
