Today I removed one of my speakers (W5-1611SAF). I wanted to measure them, to be put in a new enclosure. Next to the speaker cab is some sort of book basket with a rug on top, right up against the wall, so I put it there because it's soft and safe for the driver. It was pointing up. Out of curiosity I played some stuff with it while it's out in the open.
Amazingly, the speaker "disappeared". Of course it sounded crap, but it was eerily invisible. I had to be mere inches from it to recognize it as the sounds source. I've only read about speakers doing so and never actually heard it in real life. I did a similar thing a year ago when burning them in on my desk, but I didn't notice this effect. I would remember them if there was anything as intriguing as this. Tried the same thing to the other, hoping to find out what it'll be like in stereo, but because I couldn't exactly mirror the condition of the other one it didn't work. The other was easily localizable.
I'm interested in knowing why that works. Also, I would imagine that's one of the biggest reason as to how the Pluto disappears so well, or so they say.
Amazingly, the speaker "disappeared". Of course it sounded crap, but it was eerily invisible. I had to be mere inches from it to recognize it as the sounds source. I've only read about speakers doing so and never actually heard it in real life. I did a similar thing a year ago when burning them in on my desk, but I didn't notice this effect. I would remember them if there was anything as intriguing as this. Tried the same thing to the other, hoping to find out what it'll be like in stereo, but because I couldn't exactly mirror the condition of the other one it didn't work. The other was easily localizable.
I'm interested in knowing why that works. Also, I would imagine that's one of the biggest reason as to how the Pluto disappears so well, or so they say.
When a conventional basket-type cone driver is operating in free air, it becomes a dipole radiator (not a very good one as you noticed, but still). As such, there is a plane along and near which you would be in a null between the front and back radiation, and being opposite polarities the two would cancel. This plane would be oriented the same as an imaginary baffle extending in all directions. In this region, you'd hear very little direct sound from the driver; mostly just reflected sounds from other parts of the room.
Was your driver sitting on its back on the magnet, without much reflective surface behind it, at or near ear level? These conditions would make it "disappear" over a fairly large listening area. Kind of a cool effect, huh?
-- Jim
P.S. I don't think this is what people mean when they describe Pluto as "disappearing;" rather I believe they are referring to its sound quality being accurate enough that it has very little sonic "signature" of its own, so one tends to hear only the music and not the loudspeaker. (My speakers do this too - it's fun!
)
Last edited:
It was sitting on its back on the magnet, and I would imagine the rug though not very soft would ne be a very reflective surface. It is ear level if I'm sitting on the floor, but I tried walking around and the speaker still reacts the same. The illusion does disappear if I put my head at any height above the driver, or at an angle close to it.
Interesting. I wonder if I can recreate that effect, but with a more correct sound? It would be very difficult to get it accurate for music, but it could work for home theatre somewhat.
Interesting. I wonder if I can recreate that effect, but with a more correct sound? It would be very difficult to get it accurate for music, but it could work for home theatre somewhat.
Maybe this will provide some inspiration: http://www.diyaudio.com/forums/multi-way/121385-advantages-floor-coupled-up-firing-speakers.html
wesayso;
You are better off tooting your own horn!
You've pointed the OP off into an endless bog. Such speakers disappear in sea of reflections that have higher levels of spectrum >2kHz than direct sound. Above 2kHz directional cues rapidly are transitioning from timing based to intensity based.
Speaker's that disappear in sense of Pluto have smooth frequency response on and off axis, behave as single source, and no sibilance. They also tend to have minimal baffle area and shape. Poor baffles can behave as secondary source. This may be heard when listening to single speaker from <1m and moving slowly head up, down, and side to side. With disappearing speaker sound appears from single point that remains directly behind speaker with small head movements. With speaker that doesn't disappear, small head movements reveal sound that doesn't remain directly behind speaker, this may occur with different parts of the spectrum of the sound shifting forwards, backwards, upwards or downwards.
Classic example is a two way box speaker with woofer and 1" tweeter with typical crossover >2kHz. At crossover point speaker is two resolvable sources, and above crossover point baffle reflections change tonality and timing cues from tweeter with small head movements.
In comparison small full range speaker typically images much better than classic monkey coffin. Here a lot depends on cone profile and damping characteristics as to how well it disappears. Problems crop up when trying to get low frequencies. This manifests in a variety of ways. Tall, narrow, and shallow pencil type tower enclosures are prone to early internal reflections that radiate back through driver and otherwise modify cone motion that lead to loss single source behavior. The lighter and less self damped the cone is, the greater the loss of coherency.
Larger full range drivers operate with break up modes that cause loss of coherency as well the back wave problems, and a tendency for sibilance.
With coherent speakers room behavior still plays a big role in disappearing act.
You've discovered this with your Towers. (I'll honk your horn for you!) You have noted that with some recordings that sound appears outside of listening triangle. This happens because of strong and highly coherent reflection from side wall. The reflection mixes with direct sound to form new virtual source. You've hidden a damping panel on the offending side wall that helps, but perhaps a small, <1 foor wide wing wall of damping panel place close to side wall reflection point would capture, and otherwise reflect much of the sound back to speaker and front wall. Easy enough to try, but I understand it could be a hard sell to your commander and chief of room decor if it works.
FirstOff;
Pluto is easy to build/clone, and will give you the disappearing experience.
You are better off tooting your own horn!
You've pointed the OP off into an endless bog. Such speakers disappear in sea of reflections that have higher levels of spectrum >2kHz than direct sound. Above 2kHz directional cues rapidly are transitioning from timing based to intensity based.
Speaker's that disappear in sense of Pluto have smooth frequency response on and off axis, behave as single source, and no sibilance. They also tend to have minimal baffle area and shape. Poor baffles can behave as secondary source. This may be heard when listening to single speaker from <1m and moving slowly head up, down, and side to side. With disappearing speaker sound appears from single point that remains directly behind speaker with small head movements. With speaker that doesn't disappear, small head movements reveal sound that doesn't remain directly behind speaker, this may occur with different parts of the spectrum of the sound shifting forwards, backwards, upwards or downwards.
Classic example is a two way box speaker with woofer and 1" tweeter with typical crossover >2kHz. At crossover point speaker is two resolvable sources, and above crossover point baffle reflections change tonality and timing cues from tweeter with small head movements.
In comparison small full range speaker typically images much better than classic monkey coffin. Here a lot depends on cone profile and damping characteristics as to how well it disappears. Problems crop up when trying to get low frequencies. This manifests in a variety of ways. Tall, narrow, and shallow pencil type tower enclosures are prone to early internal reflections that radiate back through driver and otherwise modify cone motion that lead to loss single source behavior. The lighter and less self damped the cone is, the greater the loss of coherency.
Larger full range drivers operate with break up modes that cause loss of coherency as well the back wave problems, and a tendency for sibilance.
With coherent speakers room behavior still plays a big role in disappearing act.
You've discovered this with your Towers. (I'll honk your horn for you!) You have noted that with some recordings that sound appears outside of listening triangle. This happens because of strong and highly coherent reflection from side wall. The reflection mixes with direct sound to form new virtual source. You've hidden a damping panel on the offending side wall that helps, but perhaps a small, <1 foor wide wing wall of damping panel place close to side wall reflection point would capture, and otherwise reflect much of the sound back to speaker and front wall. Easy enough to try, but I understand it could be a hard sell to your commander and chief of room decor if it works.
FirstOff;
Pluto is easy to build/clone, and will give you the disappearing experience.
Lol, I didn't want to do that .
But wider than the speaker placement is pretty "normal" in my setup(*). And it only started after I placed that damping panel (covering the first reflection). The other (right) channel has no side wall at all and still the stage is wider than the speaker placement there. I believe phase and level play a role here in steering the sound. Augmented by mid/side processing creating a bit of crosstalk compensation.
Seeing that both sides behave very similar, one with a side wall and one without it has to be something other than a side reflection. The damping panel is placed at the first reflection point on the left side. Using side walls to widen the stage has limits. It would do the same trick on every track. My stage actually varies with the music played. Most panned pop music pretty much stays within the triangle though. Mono is really mono, coming from the front only.
With stereo, eyes closed and I cannot accurately point out where the speakers are on most songs, judging by the sound heard at that moment. As it should be I.M.H.O. First time I heard it like that was in my car, and that resulted in the build of the towers to be able to do that trick in my house. Minimising first reflections is key. All that is left are instruments and voices floating.
The disappearing works very well with the two towers. On some recordings phase trickery can place the sound way to the sides, not wide but toward you in front of the speaker.
You can only do that if the first clues to speaker placement are missing, usually those clues are floor and ceiling reflections. Those are minimised with line arrays. Cover the rest of the first reflections the best you can and you hear a believable stage.
Aiming the speakers lets me fine tune the shape of the stage. Pretty hard to choose what's right though. I cross them a bit behind the listening position while in my car I cross in front. That last trick (crossing in front) makes a wider sweet spot but I cannot get away with that without creating an argument in the house.
I agree with the narrow/minimalistic baffle theory though. I chose the round shape with minimal baffle width, enhanced by the waveguide like placement of drivers to have a thin line of sound. At least the best I could do within reason.
But the thread I pointed at talks some more about the up firing speakers. I remember putting in a link to something Tom Danley said about an experiment with up firing speakers.
These posts fit in this thread as well:
Source: http://www.diyaudio.com/forums/multi-way/236009-hearing-future-loudspeaker-design-18.html#post3499478
(*) It's not always wider, sometimes its like the sound coming from in front of the speaker. Hard to explain but exactly like I hoped it would be. You get a mixed queue from the room (with damped reflection points) and the ones in the recording. By faking a room queue behind me I hope to diminish the left over queue's making it feel like you're in a bigger space. A trick that used to be common in studios to create more creature comfort.
Not used that often anymore I believe because it made every recording sound nicer than it actually was. Not good in studio's, very good in homes I say.
The back reflected sound is delayed in time long enough to be uncorrelated to the first sound but short enough not to be heard as a reflection. Usually within 15 to 25 ms. Preferably later than real cues in the recording room.
.But wider than the speaker placement is pretty "normal" in my setup(*). And it only started after I placed that damping panel (covering the first reflection). The other (right) channel has no side wall at all and still the stage is wider than the speaker placement there. I believe phase and level play a role here in steering the sound. Augmented by mid/side processing creating a bit of crosstalk compensation.
Seeing that both sides behave very similar, one with a side wall and one without it has to be something other than a side reflection. The damping panel is placed at the first reflection point on the left side. Using side walls to widen the stage has limits. It would do the same trick on every track. My stage actually varies with the music played. Most panned pop music pretty much stays within the triangle though. Mono is really mono, coming from the front only.
With stereo, eyes closed and I cannot accurately point out where the speakers are on most songs, judging by the sound heard at that moment. As it should be I.M.H.O. First time I heard it like that was in my car, and that resulted in the build of the towers to be able to do that trick in my house. Minimising first reflections is key. All that is left are instruments and voices floating.
The disappearing works very well with the two towers. On some recordings phase trickery can place the sound way to the sides, not wide but toward you in front of the speaker.
You can only do that if the first clues to speaker placement are missing, usually those clues are floor and ceiling reflections. Those are minimised with line arrays. Cover the rest of the first reflections the best you can and you hear a believable stage.
Aiming the speakers lets me fine tune the shape of the stage. Pretty hard to choose what's right though. I cross them a bit behind the listening position while in my car I cross in front. That last trick (crossing in front) makes a wider sweet spot but I cannot get away with that without creating an argument in the house
.I agree with the narrow/minimalistic baffle theory though. I chose the round shape with minimal baffle width, enhanced by the waveguide like placement of drivers to have a thin line of sound. At least the best I could do within reason.
But the thread I pointed at talks some more about the up firing speakers. I remember putting in a link to something Tom Danley said about an experiment with up firing speakers.
These posts fit in this thread as well:
Source: http://www.diyaudio.com/forums/multi-way/236009-hearing-future-loudspeaker-design-14.html#post3497251Tom Danley said:
Tom Danley said:
Source: http://www.diyaudio.com/forums/multi-way/236009-hearing-future-loudspeaker-design-18.html#post3499478
(*) It's not always wider, sometimes its like the sound coming from in front of the speaker. Hard to explain but exactly like I hoped it would be. You get a mixed queue from the room (with damped reflection points) and the ones in the recording. By faking a room queue behind me I hope to diminish the left over queue's making it feel like you're in a bigger space. A trick that used to be common in studios to create more creature comfort.
Not used that often anymore I believe because it made every recording sound nicer than it actually was. Not good in studio's, very good in homes I say
.The back reflected sound is delayed in time long enough to be uncorrelated to the first sound but short enough not to be heard as a reflection. Usually within 15 to 25 ms. Preferably later than real cues in the recording room.
Last edited:
I would take whatever info I get and put the pieces together, and you know, give it my best shot in getting the right picture, hopefully. I remember an Instructable I read a couple of years ago of a 3-way that's named after some parasitic worm (something that sounds like 'tube'). Minimal baffle area, properly designed crossover, seems to do the trick according to the builder. That, the Pluto, the Towers, seems to be happening a bit more often with low-profile enclosures.
I have been longing to replicate a Pluto when I have my hands on a decent DSP, because I want to know what all the fuss was about.
I have been longing to replicate a Pluto when I have my hands on a decent DSP, because I want to know what all the fuss was about.
I spend some time investigating the Haas effect for my next experiment, which will be based on the Haas Kicker idea.
I think I figured out why my arrays do what they do. My strongest early reflection is at about 7 ms from the back wall. Largely dependant on the aiming of the arrays.
After thinking about it I'm pretty sure it is this reflection that causes the widening effect I get in my particular room/listening position.
As said, aiming the arrays I can dial in that effect Due to having an absorption panel behind the listening position. I see the changes in amplitude in my measurements of that 7 ms reflection in relation to toe-in of the speakers. So now I'm pretty sure it's not a side wall reflection of any kind which was already hard for me to believe due to the asymmetric setup. Knowing what caused it I can take advantage of it for my up coming test with virtually created reflections from behind.
Check out this video and this particular Haas effect will become more clear:
https://www.youtube.com/watch?v=UQOkSF8auFc
Just wanted to share my thoughts on this here as it might help others that read the above.
The proposed test I intend to play with is partly based on the above theory. Check out my latest rambling on that on my own thread: Haas kicker testing, the idea
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.