Greetings.
In am contemplating building an omnidirectional speaker using conical dispersers below downward facing drivers. My design will probably be two way with widerange HF and woofer with crossover about 500Hz. Has anyone experience with conical dispersers? What are the difficulties?
Your comments appreciated.
In am contemplating building an omnidirectional speaker using conical dispersers below downward facing drivers. My design will probably be two way with widerange HF and woofer with crossover about 500Hz. Has anyone experience with conical dispersers? What are the difficulties?
Your comments appreciated.
How do you plan on making the speaker stands? Presumably they'll be in the path of the sound, so maybe they could be profiled in a way that gives the system a bit of horn loading?
Hi CeramicMan.
I have seen designs using a 360 degree horn like you seem to be suggesting. I reckon that I should try to avoid horn effects since horn reinforcement would be in a restricted frequency range with amplitude ripples outside that range. I am not a horn expert and so I intend to keep it simple. I am aiming at diffusion rather than reinforcement, and wonder if there are difficulties in achieving that.
What do you think?
I have seen designs using a 360 degree horn like you seem to be suggesting. I reckon that I should try to avoid horn effects since horn reinforcement would be in a restricted frequency range with amplitude ripples outside that range. I am not a horn expert and so I intend to keep it simple. I am aiming at diffusion rather than reinforcement, and wonder if there are difficulties in achieving that.
What do you think?
I'd probably go for two sections. E.g.: midwoofer with a dispersion cone underneath it, and a similar system for the tweeter placed on top of the midwoofer box.
The stands could be a tripod-like design, or maybe they could converge onto the cone?
You'd have to make some drawings to see what angles the reflections would go in. And you might find that some other shape is better suited rather than a cone, e.g.: a bullet tip?
The stands could be a tripod-like design, or maybe they could converge onto the cone?
You'd have to make some drawings to see what angles the reflections would go in. And you might find that some other shape is better suited rather than a cone, e.g.: a bullet tip?
First, I'm not expert but have built one omni design and I'm working ont he second and third ones.
First, you are essentially listening 90º off axis to the driver. This means you will have a very quick natural rool off with most drivers. The diffuser can help flatten this out and extend it a bit. A diffuser, combined with a waveguide, can do an even better job of flattening and extending the response.
You want to think about the wavelengths at your contemplated crossover points. Diffusers/dfelectors do little to nothing down low because the wavelengths are too long. For example the woofer you crossover to at 500 hz will not be helped by a diffuser, except perhaps to smooth out its roll off up high, well above the crossover point.
The spacing between the diffuser and the driver will be critical and have a major effect on it's response. IMHO, do not even try this kind of project unless you have full measurement capabilities and can test lots of different scenarios until you find the FR you are looking for.
My experimentation seems to indicate that concave cone diffusers work best. I have them to have better FR than dome shapes or straight cone shapes.
Below are a couple pics of the projects I have done or are in the prototype stage. The first one features a downfiring 12" sub, an up and downfiring midwoofer, and a front firing tweeter. It is only omnidirectional up the the crossover to the tweeter, however I believe the front firing tweeter improves the imaging.
The second one is fully omnidirectional, at least horizontally9the one one the right in the pic). It features up and downfiring woofers, an upfiring dome midrange, which is in a modified waveguide and fires up into a concave cone diffuser, and an upfiring tweeter, which is in a modified waveguide and recessed in the diffuser for the mid. the tweeter fires into another concave cone diffuser.
Also, there are two threads on the process I went through to get these done over at HTGuide that you might find helpful in terms of understanding what diffusers, etc. will do to driver response. There are lots of measuements, graphs, etc. detailing my journey.
There is little to no information out there in the DIY community on these kinds of speakers and they are very challenging to get right. Good Luck!
http://htguide.com/forum/showthread.php4?t=22445&page=6&pp=35&highlight=omnidirectional
http://htguide.com/forum/showthread.php4?t=25074&highlight=omnidirectional
First, you are essentially listening 90º off axis to the driver. This means you will have a very quick natural rool off with most drivers. The diffuser can help flatten this out and extend it a bit. A diffuser, combined with a waveguide, can do an even better job of flattening and extending the response.
You want to think about the wavelengths at your contemplated crossover points. Diffusers/dfelectors do little to nothing down low because the wavelengths are too long. For example the woofer you crossover to at 500 hz will not be helped by a diffuser, except perhaps to smooth out its roll off up high, well above the crossover point.
The spacing between the diffuser and the driver will be critical and have a major effect on it's response. IMHO, do not even try this kind of project unless you have full measurement capabilities and can test lots of different scenarios until you find the FR you are looking for.
My experimentation seems to indicate that concave cone diffusers work best. I have them to have better FR than dome shapes or straight cone shapes.
Below are a couple pics of the projects I have done or are in the prototype stage. The first one features a downfiring 12" sub, an up and downfiring midwoofer, and a front firing tweeter. It is only omnidirectional up the the crossover to the tweeter, however I believe the front firing tweeter improves the imaging.
The second one is fully omnidirectional, at least horizontally9the one one the right in the pic). It features up and downfiring woofers, an upfiring dome midrange, which is in a modified waveguide and fires up into a concave cone diffuser, and an upfiring tweeter, which is in a modified waveguide and recessed in the diffuser for the mid. the tweeter fires into another concave cone diffuser.
Also, there are two threads on the process I went through to get these done over at HTGuide that you might find helpful in terms of understanding what diffusers, etc. will do to driver response. There are lots of measuements, graphs, etc. detailing my journey.
There is little to no information out there in the DIY community on these kinds of speakers and they are very challenging to get right. Good Luck!
http://htguide.com/forum/showthread.php4?t=22445&page=6&pp=35&highlight=omnidirectional
http://htguide.com/forum/showthread.php4?t=25074&highlight=omnidirectional
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Re: pics... So essentially it looks like a form of horn-loading anyway, right? The profile is essentially rotated around a circle.
The second pic or the prototype is essentially 360º horn loading, yes.
It's the same concept that Duevel uses in their commercial omni's.
It's the same concept that Duevel uses in their commercial omni's.
Post #5
First, I'm not expert but have built one omni design and I'm working ont he second and third ones...
Hi dlneubec
Congratulations for going and doing it. There is a feast of information in your thread which I will digest later.
Thanks.
AA
First, I'm not expert but have built one omni design and I'm working ont he second and third ones...
Hi dlneubec
Congratulations for going and doing it. There is a feast of information in your thread which I will digest later.
Thanks.
AA
Could you guys explain to me the hole point about omnidirectional ( well i know they aren;t doing it seriously, but you guys are actually trying to get to it)
Is it because you like to have a lot of reflections?
ambiance?
i've owned a few Mirage OM loudspeakers, and though the quality of the drivers and the system was very high,
they were introducing a lot of problems in small rooms with too much reflections soo close to the original signal
is it for dispersion / off axis listening ?
and how does your "deflectors" work with different frequency ranges ? i would tend to think that a certain mecanical deflector could only work with a limited frequency range ( lower would be not affected and higher wouldn'thave time to take a serious radius/angle change ?? )
Is it because you like to have a lot of reflections?
ambiance?
i've owned a few Mirage OM loudspeakers, and though the quality of the drivers and the system was very high,
they were introducing a lot of problems in small rooms with too much reflections soo close to the original signal
is it for dispersion / off axis listening ?
and how does your "deflectors" work with different frequency ranges ? i would tend to think that a certain mecanical deflector could only work with a limited frequency range ( lower would be not affected and higher wouldn'thave time to take a serious radius/angle change ?? )
Some people (me including) believe there is more about radials than lots of reflected sound. Here is some discussion about it:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=100356
http://www.diyaudio.com/forums/showthread.php?s=&threadid=100356
Could you guys explain to me...
There are a few potential benefits. Whether they prove to be so remains to be seen.
1/
I have read quite a few articles about the reflected sound being part of our normal experience in a room - eg two people in conversation. The logic is that loudspeakers ought to bounce sound off the walls in the same way becase our ear 'expects' it and therefore it will sound more natural.
2/
There is a view that the HF and MF drivers ought to have similar dispersion so that the reflected sound won't be imbalanced. In practise this is hard to achieve. HF drivers beam more than MF drivers, and LF drivers are always nearly omnidirectional. Therefore reflections from the side walls will have excessive MF and LF energy. Our ears are confused by this imbalance and the result is listening fatigue.
Having an omnidirectional speaker would solve this uneven dispersion problem (at least in the horizontal plane)
3/
I once read an article in a HiFi magazine where the authors turned the two speakers around to face the backwall and also placed big cushions between the speakers and the listener. They found that it sounded quite good and the stereo image was still fine.
Your experience with Mirage OM speakers is interesting. How did they sound in larger rooms?
As regards deflector frequency characteristics I expect that an infinitely large hard solid will reflect all frequencies equally. I reckon if the reflector is a few wavelengths in size it will reflect all sound above that frequency. I expect there would be diffractions from the perimeter of the deflector to complicate matters.
All this is semi educated guesswork on my part since I haven't built any speakers of this type yet. All experience (both good and bad) from fellow experimenters will be of great value to me.
There are a few potential benefits. Whether they prove to be so remains to be seen.
1/
I have read quite a few articles about the reflected sound being part of our normal experience in a room - eg two people in conversation. The logic is that loudspeakers ought to bounce sound off the walls in the same way becase our ear 'expects' it and therefore it will sound more natural.
2/
There is a view that the HF and MF drivers ought to have similar dispersion so that the reflected sound won't be imbalanced. In practise this is hard to achieve. HF drivers beam more than MF drivers, and LF drivers are always nearly omnidirectional. Therefore reflections from the side walls will have excessive MF and LF energy. Our ears are confused by this imbalance and the result is listening fatigue.
Having an omnidirectional speaker would solve this uneven dispersion problem (at least in the horizontal plane)
3/
I once read an article in a HiFi magazine where the authors turned the two speakers around to face the backwall and also placed big cushions between the speakers and the listener. They found that it sounded quite good and the stereo image was still fine.
Your experience with Mirage OM speakers is interesting. How did they sound in larger rooms?
As regards deflector frequency characteristics I expect that an infinitely large hard solid will reflect all frequencies equally. I reckon if the reflector is a few wavelengths in size it will reflect all sound above that frequency. I expect there would be diffractions from the perimeter of the deflector to complicate matters.
All this is semi educated guesswork on my part since I haven't built any speakers of this type yet. All experience (both good and bad) from fellow experimenters will be of great value to me.
Ok
my two cents on that for now ..
Sound that is recorded on a track,
isn't always supposed to be played as we do ..
because it depends on the recording situation
example ..if they use seperate anechoic or almost tracks for everysound ( like signer voice close to mic )
then yes, we need to emulate the way it would sound within a certain room..but then we aren't supposed to be able to reproduce their situation since there was not an actual situation where the mic was recording bouncing sounds and all ( unless mixing does this on purpose with machines, wich i know nothing of ...do they do that kind of stuff ?? )
Then there is situations like a band is playing in a room where the mic gets the sound of everywhere including reflections up to a certain level
or orchestral recording where everything bouces from everywhere with echoes and such
how are we supposed to be able to reproduce all of the different situations ?
Then, i do not have the knowledge, but i would tend to logically think that sounds produced by instruments in real world do have different dispersion ..
a trompet would probably radiate more like a horn
loudpseaker than a dipole baffle
where drums might be more like an ob with wings ...
again all those things should influenced the recorded sounds nah ?
so how can we assume that using omnipolar dispertion will reproduce anything like the recording situation ?
since every listening room will influence differently the final result , and since it might not just be needed at all
to get the best reproduction of the recorded sound
i might be completly aside of the track here,
since i have very limited knowledge in recording/mixing technicalities ...but this is my logical reasonning
my two cents on that for now ..
Sound that is recorded on a track,
isn't always supposed to be played as we do ..
because it depends on the recording situation
example ..if they use seperate anechoic or almost tracks for everysound ( like signer voice close to mic )
then yes, we need to emulate the way it would sound within a certain room..but then we aren't supposed to be able to reproduce their situation since there was not an actual situation where the mic was recording bouncing sounds and all ( unless mixing does this on purpose with machines, wich i know nothing of ...do they do that kind of stuff ?? )
Then there is situations like a band is playing in a room where the mic gets the sound of everywhere including reflections up to a certain level
or orchestral recording where everything bouces from everywhere with echoes and such
how are we supposed to be able to reproduce all of the different situations ?
Then, i do not have the knowledge, but i would tend to logically think that sounds produced by instruments in real world do have different dispersion ..
a trompet would probably radiate more like a horn
loudpseaker than a dipole baffle
where drums might be more like an ob with wings ...
again all those things should influenced the recorded sounds nah ?
so how can we assume that using omnipolar dispertion will reproduce anything like the recording situation ?
since every listening room will influence differently the final result , and since it might not just be needed at all
to get the best reproduction of the recorded sound
i might be completly aside of the track here,
since i have very limited knowledge in recording/mixing technicalities ...but this is my logical reasonning
JinMTVT I'll answer your question with a question: why would you think the standard loudspeaker that is called a monopole but which has dispersion that varies wildly from highly directive to nearly omnipolar, and back, several times, depending on the drivers and network be any better or even "the best" way to reproduce music or the recording situation?
that is not what i think at all
i am just questioning myself about sending sound waves all over the place in all sorts of different rooms
getting infinitly variable results
doesn't it incur more time smeared sound
and alot more of early reflections than regular monopole/dipole setup ?
i am just questioning myself about sending sound waves all over the place in all sorts of different rooms
getting infinitly variable results
doesn't it incur more time smeared sound
and alot more of early reflections than regular monopole/dipole setup ?
JinMTVT said:
I hope I didn't offend you. Sorry if I did.
I have the same concerns as you regarding the dispersion and *very* early reflections. I don't know the answer but I would like to hear a good omni and see if it is deterimental to SQ. Considering the poor acoustics of the average room I wonder if the str
not at all!!! 🙂
it is sometime hard to carry emotions or the lack of on a forum 😛
Sound reproduction is a very very complex system
because it involves soo many variables
and we always try to reproduce sound that was produced by different objects in different spaces,
in our home with very different transducers ..
it is very hard for me ( maybe because i have no engineering background ) to try understand even the basic principles ...but my logical sense tells me there is something wrong in the way we are trying to reproduce the sound VS the way it was captured
it is sometime hard to carry emotions or the lack of on a forum 😛
Sound reproduction is a very very complex system
because it involves soo many variables
and we always try to reproduce sound that was produced by different objects in different spaces,
in our home with very different transducers ..
it is very hard for me ( maybe because i have no engineering background ) to try understand even the basic principles ...but my logical sense tells me there is something wrong in the way we are trying to reproduce the sound VS the way it was captured
I'm not sure early reflections are as much a problem as many folks think. However, my first project is omni up to about 1.5khz where it crosses over to a direct firing tweeter, so from there up is more like a traditional speaker.My speakers are about 36" from the front wall and at least 36" from the side walls. My listening position is about 12' away in a 15'x19' room. My second project is a single prototype and more of a true omni, with 360º dispersion and I can't compare it, since there is only one at this time.
IMO, there is no early reflection problem with smearing the image. The imaging and clarity are excellent. The soundstage is considerably wider and deeper than any of the traditional box speakers I've built, and also than my open baffle dipoles. It projects considerably behind the speakers and seemingly outside the room on some recordings.
To me, open baffle dipole was a big improvement over standard direct firing boxes and this imrpoves on that, to my ear anyway. The midrange simply seems more open, less like a veil has been thrown over the speakers, which is what some traditional box speakers give me the feeling of.
IMO, there is no early reflection problem with smearing the image. The imaging and clarity are excellent. The soundstage is considerably wider and deeper than any of the traditional box speakers I've built, and also than my open baffle dipoles. It projects considerably behind the speakers and seemingly outside the room on some recordings.
To me, open baffle dipole was a big improvement over standard direct firing boxes and this imrpoves on that, to my ear anyway. The midrange simply seems more open, less like a veil has been thrown over the speakers, which is what some traditional box speakers give me the feeling of.
My comment above seems to have been cutoff. I meant to say: Considering the poor acoustics of the average room I wonder if the strength of the omni idea is that it's extremely wide and even dispersion has a more positive effect on SQ than the possible *very* early reflections have a negative effect?
My experience with omnis tells me that sound quality can be a big problem with too many quick delayed reflections ... the waves that hit the back wall do usually take almost double time as the front one to get back to the listener ( depending on loudspeaker placement vs backwall )
it can't be good to have strong early reflections and it adds up to the primary sound too much where it is important to keep the details at ...
I have also experience the wide soundstage with the Mirage Om's in my small room .quite impressive on some recordings.. but is a really big soundstage something to trade off for sound quality ???
Even dipole radiating loudspeakers share some of the same difficulties of smearing in small rooms , though lessen by the lack of side waves ....
We coul always treat the backwall and side walls
( sides should always be neway .. )
but then what would be the point in making omni-directional drivers?
also keep in mind that the more you diffuse the waves,
the less powerfull they get at your listening area,
thus you have to augment the amplification to get the same volume ...wich usually brings distortion and such
have you tried to use your omnis in close range/nearfield
i didn't try it myself..would have to try it soon
just for the kicks..i believe that if the sound improves alot but getting closer to the drivers ( not so close that you can't hear any reflections .. ) you could find the perfect spot where omni radiating loudspeakers should work better than any of other systems
( less amplification required in close range, attenuation of the reflected waves VS direct sound ...)try it and let us know what difference you get from seating half the usual distance..then 1/4 of it
it can't be good to have strong early reflections and it adds up to the primary sound too much where it is important to keep the details at ...
I have also experience the wide soundstage with the Mirage Om's in my small room .quite impressive on some recordings.. but is a really big soundstage something to trade off for sound quality ???
Even dipole radiating loudspeakers share some of the same difficulties of smearing in small rooms , though lessen by the lack of side waves ....
We coul always treat the backwall and side walls
( sides should always be neway .. )
but then what would be the point in making omni-directional drivers?
also keep in mind that the more you diffuse the waves,
the less powerfull they get at your listening area,
thus you have to augment the amplification to get the same volume ...wich usually brings distortion and such
have you tried to use your omnis in close range/nearfield
i didn't try it myself..would have to try it soon
just for the kicks..i believe that if the sound improves alot but getting closer to the drivers ( not so close that you can't hear any reflections .. ) you could find the perfect spot where omni radiating loudspeakers should work better than any of other systems
( less amplification required in close range, attenuation of the reflected waves VS direct sound ...)try it and let us know what difference you get from seating half the usual distance..then 1/4 of it
- Status
- Not open for further replies.