Stig Erik;
have you tried to have the Seas 8" hanging 2 and 2, side by side downwards, insted of 4 downwards?
Anyone know whether this would seriously mess up the Frd?
They would probably have to be hang as close to each other as possible i the horisontal plane, perhaps with a little baffel in between the two horisontal ones(?)
(I ask because this would give the option of introducing more drivers without increasing the height of the speaker)
I.e. this:
OO
OO
in stead of
O
O
O
O
have you tried to have the Seas 8" hanging 2 and 2, side by side downwards, insted of 4 downwards?
Anyone know whether this would seriously mess up the Frd?
They would probably have to be hang as close to each other as possible i the horisontal plane, perhaps with a little baffel in between the two horisontal ones(?)
(I ask because this would give the option of introducing more drivers without increasing the height of the speaker)
I.e. this:
OO
OO
in stead of
O
O
O
O
Re: the Neo3... horses for courses. 
SL was just setting the Neo3 up for failure as an object lesson in power handling.
I use a Neo8 down to 500Hz (6th order HP), and am very pleased with it -- great bang for the buck. However, I'm pretty sure that if I were to pump 30 or 50 watt 500Hz tone bursts through it unprotected, it too would probably squeal like a pig. But, this would all be academic, since at 50 watts, my 6V6 tubes would be squealing too, and probably melting as well.
SL was just setting the Neo3 up for failure as an object lesson in power handling. I use a Neo8 down to 500Hz (6th order HP), and am very pleased with it -- great bang for the buck. However, I'm pretty sure that if I were to pump 30 or 50 watt 500Hz tone bursts through it unprotected, it too would probably squeal like a pig.
But, this would all be academic, since at 50 watts, my 6V6 tubes would be squealing too, and probably melting as well. No, havn't tried it. I didnt want my speaker to be that wide.
I guess it could work fine, but the dipole peak might be shifted downwards into the 500 Hz area, which could be a problem if you indend to XO above 500.
Last edited:
The latest update here now is some more work on the room acoustics.
The general level of early reflections and decay in my room is very low, but there was still some early reflections that could be treated. It turned out that I did get a reflection from the front wall behind the speakers, although the wall is damped with acoustic tiles covered with curtains. The source of the problem was the curtains, they were slightly reflective. I solved it very simply - just hang them in folds rather than straight....
Finally, there is the floor reflection, which is hard to cure. Who wants a thick absorber on their floor? I dont! I've managed to suppress it well enough with two thick carpets, not perfect but better than nothing.
And now - a ETC (Energy Time Curve) of my room, in the listening position. The first pulse is the speaker, all the rest are room reflections. As a point of reference, the second graph is a typical living room. As you can see, the decay is much longer, and there are very strong early reflections (and it sounds like craaap).
The general level of early reflections and decay in my room is very low, but there was still some early reflections that could be treated. It turned out that I did get a reflection from the front wall behind the speakers, although the wall is damped with acoustic tiles covered with curtains. The source of the problem was the curtains, they were slightly reflective. I solved it very simply - just hang them in folds rather than straight....
Finally, there is the floor reflection, which is hard to cure. Who wants a thick absorber on their floor? I dont! I've managed to suppress it well enough with two thick carpets, not perfect but better than nothing.
And now - a ETC (Energy Time Curve) of my room, in the listening position. The first pulse is the speaker, all the rest are room reflections. As a point of reference, the second graph is a typical living room. As you can see, the decay is much longer, and there are very strong early reflections (and it sounds like craaap).
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.
Last edited:
So I have tried no baffle!
Hi
Now I have tried a "no baffle" speaker. And I think I like the result.
I followed the guidance of Stig Eric when it comes to set the DSP
of the Hypex AS2.100
So far I added and changed a few things.
1. Beyma 18G550 -16 dB
2. Tang Band 4W 1320 -1 dB
Unfortunately I only have 1 Beyma 18G550 at he moment so the other side only has a Scan Speak 25W/8565 woofer wich I cut of at 100 Hz.
So the set up is not perfect, far from perfect.
But it still sounds amazing. At least for the first 10 min. then the tweeter channel of the Hypex on the left side stopped playing.
So for the last 5 days I have been listening to a system where only the midrange is identical in left and right side.
But still even with this amputated system I am convinced that my future lies in no baffle speakers.
One thing I don't understand Stig Eric is how you can listen at such a close distance to your speakers.
I prefere at least 3 meters distance.
I think it is because i like the illusion of listening to the band playing in the recording room, instead of having the band playing in my own room. If you know what I mean. Beeing there instead of them beeing here.
Best regards
Uwe
Hi
Now I have tried a "no baffle" speaker. And I think I like the result.
I followed the guidance of Stig Eric when it comes to set the DSP
of the Hypex AS2.100
So far I added and changed a few things.
1. Beyma 18G550 -16 dB
2. Tang Band 4W 1320 -1 dB
Unfortunately I only have 1 Beyma 18G550 at he moment so the other side only has a Scan Speak 25W/8565 woofer wich I cut of at 100 Hz.
So the set up is not perfect, far from perfect.
But it still sounds amazing. At least for the first 10 min. then the tweeter channel of the Hypex on the left side stopped playing.
So for the last 5 days I have been listening to a system where only the midrange is identical in left and right side.
But still even with this amputated system I am convinced that my future lies in no baffle speakers.
One thing I don't understand Stig Eric is how you can listen at such a close distance to your speakers.
I prefere at least 3 meters distance.
I think it is because i like the illusion of listening to the band playing in the recording room, instead of having the band playing in my own room. If you know what I mean. Beeing there instead of them beeing here.
Best regards
Uwe
Being in the US I picked up some 18SWS1100s, which arrived yesterday. In nude dipole with equalization in place SPL holds up to 35Hz (speced Fs is 39Hz) and I'm measuring THD at the ambient noise floor of my environment at typical listening levels (-40dB THD at 35Hz, dropping to -55dB around 80Hz). Will take a closer look once the cones are more thoroughly broken in.
Having only four amp channels at the moment I've been doing something similar with the 18WS1100s. Works surprisingly well. Though no so well as to impair my motivation to get some more power amps built.
Good to hear that you like no-baffle speakers!
I too want to be "transfered" into the music event/recording rather than make it sound like its in my room. I dont want to hear my own room, I want to hear the "room" (real or artificial room) thats on the recording. To really hear that, you must remove yourself as much as possible from your own room. That means both bringing down the reflection and decay of your room, as well as increase the level of the direct sound from the speakers by shortening the listening distance.
I would love to have 3 meter listening distance as well, but my room is not big enough for that unfortunately.
Stig Erik,
If I understand it correctly, you want to hear what the loudspeaker is doing, minimizing reflections. Well, then maybe dipoles aren't the best solution for you. Dipoles do have greater directivity than most types of loudspeakers, but they still inherently make use of the room acoustics. It seems to me that what you really want is speaker that has much stronger directivity than a dipole. The name 'Geddes' springs to mind .
Have you ever contemplated designing a system with a horn/WG for the mids and highs, dipole for the midbass and closed subs for deep bass?
If I understand it correctly, you want to hear what the loudspeaker is doing, minimizing reflections. Well, then maybe dipoles aren't the best solution for you. Dipoles do have greater directivity than most types of loudspeakers, but they still inherently make use of the room acoustics. It seems to me that what you really want is speaker that has much stronger directivity than a dipole. The name 'Geddes' springs to mind
. Have you ever contemplated designing a system with a horn/WG for the mids and highs, dipole for the midbass and closed subs for deep bass?
We cannot have a totally dead room, so room reflections and "reverb" will still be there to some degree. I think its very important that the spatial content ("frequency response") of the room reflections and decay is somewhat close to the on-axis response. With Constant Directivity speakers and proper room treatment this is possible to achive.
The "Geddes type" of speaker does not have Constant Directivity, like a proper dipole can (almost) have.
I dont quite understand what you mean by saying that dipoles "inherently make use of the room acoustics". Please elaborate.
The "Geddes type" of speaker does not have Constant Directivity, like a proper dipole can (almost) have.
I dont quite understand what you mean by saying that dipoles "inherently make use of the room acoustics". Please elaborate.
I agree a dipole could potentially be more CD than Geddes' speakers, but you will need even more different driver sizes than you are currently using. Moreover physics and available driver sizes limit the maximum frequency to which the speaker can be CD to about 8 khz. Still, you should ask yourself just how CD a speaker really has to be.
The Summa is omni up to about 300 hz, above which frequency directivity increases rapidly up to about 500 hz. Above that directivity slowly increases a little more up to a little above 1 khz, to stay more or less constant above that. I believe CD to be more important at higher frequencies and an even power-response to be the most determining factor for sound quality at lower frequencies. So maybe a Geddes type speaker isn't that bad .
You designed your listening room and equipment for relatively strong direct sound and for strongly damped reflections. A speaker with a higher DI will give you more of what you are trying to accomplish.
A dipole sends an equal amount of sound energy to its rear as to its front. The sound sent to the rear doesn't contribute to the direct sound, but instead increases the total level of the reflections. That works against your efforts to dampen all reflections, especially those on the front wall. It would be more convenient if the speaker didn't radiate to its rear in the first place. You could be accomplished by means of a waveguide.
I am not suggesting you buy the Summa, I am suggesting you designed a speaker that combined the best of both worlds: a large WG from maybe 700 hz up to 20 khz, to maybe cardioid below that, and dipole and omni below that. Just some suggestions. Of course you'd have to think it through thoroughly, but I think it could lead to great results. Of course I am not implying your current dipoles don't already sound superb, it's just that the design of your loudspeakers and the acoustic treatment of your room seem a bit contradictory.
ESP15 freq. resp. plots.
The Summa is omni up to about 300 hz, above which frequency directivity increases rapidly up to about 500 hz. Above that directivity slowly increases a little more up to a little above 1 khz, to stay more or less constant above that. I believe CD to be more important at higher frequencies and an even power-response to be the most determining factor for sound quality at lower frequencies. So maybe a Geddes type speaker isn't that bad
.You designed your listening room and equipment for relatively strong direct sound and for strongly damped reflections. A speaker with a higher DI will give you more of what you are trying to accomplish.
A dipole sends an equal amount of sound energy to its rear as to its front. The sound sent to the rear doesn't contribute to the direct sound, but instead increases the total level of the reflections. That works against your efforts to dampen all reflections, especially those on the front wall. It would be more convenient if the speaker didn't radiate to its rear in the first place. You could be accomplished by means of a waveguide.
I am not suggesting you buy the Summa, I am suggesting you designed a speaker that combined the best of both worlds: a large WG from maybe 700 hz up to 20 khz, to maybe cardioid below that, and dipole and omni below that. Just some suggestions. Of course you'd have to think it through thoroughly, but I think it could lead to great results. Of course I am not implying your current dipoles don't already sound superb, it's just that the design of your loudspeakers and the acoustic treatment of your room seem a bit contradictory.
ESP15 freq. resp. plots.
An externally hosted image should be here but it was not working when we last tested it.
The only way to really find out of things is to build it, measure and listen. Now I'm not going to do that right now, as I am very satisfied with the system I have. I'm sure the Summa is a brilliant speaker though.
My current 4-way dipoles are CD up to about 5 kHz, where the tweeter starts to get lobes because of its width. CD all the way up is very difficult, if not impossible with a dipole. Waveguides do have an advantage there.
From my own experience dipoles have their biggest advantage in the bass region.. much less problems with room modes, actually the whole decay in the bass range is lower in level and shorter than with omni subwoofers, and frequency response is flat and depend very little on placement within the room. Omni's are a total mess here. One thing is for sure - I will never never ever go back to omni in the bass region. Period.
The rear wave is a "bonus" with dipoles that I would like to not have. I have killed the rear radiation faily well since the entire front end of my room is damped, so I dont think the rear radiation contributes to room ambience in my setup, or at least very little.
My current 4-way dipoles are CD up to about 5 kHz, where the tweeter starts to get lobes because of its width. CD all the way up is very difficult, if not impossible with a dipole. Waveguides do have an advantage there.
From my own experience dipoles have their biggest advantage in the bass region.. much less problems with room modes, actually the whole decay in the bass range is lower in level and shorter than with omni subwoofers, and frequency response is flat and depend very little on placement within the room. Omni's are a total mess here. One thing is for sure - I will never never ever go back to omni in the bass region. Period.
The rear wave is a "bonus" with dipoles that I would like to not have. I have killed the rear radiation faily well since the entire front end of my room is damped, so I dont think the rear radiation contributes to room ambience in my setup, or at least very little.
Stig's Dipole line array work i think deserves further exloration. Why do most speaker systems ignore the area above the HF device. Clearly there's a boundy above just as significant as the one below?
The discussions between Stig and keyser make interesting points on both sides of the tracks so to speak. For me and my space, directivity via waveguides has been the most successful to my ears....which don't seem to appreciate a 'live' sounding space anymore. Dipole rear wave radiation has proven too complicated for 'ME' and my brain has a hard time sorting out the info, resulting in fatigue quite quickly.
Possibly the right solution might be for heavy absorbtion/treatments behind a dipole and to the outside allowing for 'just' the cancellations to occur leaving only the highly directive remaining material to reach the listener. I haven't explored the issue yet and not sure i'm going to as the CD/WG combo i'm using for the HF is producing some of the most stunning detail and clarity i've ever heard and i'd be hard pressed to part with it.
Now the lower midrange is another story that i've yet to work out. There's been some implementation of the hybrid concept (Emerald Physics) that seems to work well and yet others have stated a strong feeling of disconnect between the octaves as i would imagine room reflections are impacting the listening experience. These subjective reviews seem to fault the WG/CD instead of the OB/Dipole part of the system.......which i find confusing,
Clearly it's difficult to maintain directivity much below 1khz with a single point source driver such as that in the Gedlee and other WG designs so how to we/I overcome this. Dipole bass from 250hz down certainly helps to clean up the bottom octaves with less excitation of room resonances....But what about the 250-1khz range? Is the answer dipole implentation there as well?......In my case i'd be back to the same reflected material that 'escapes' cancellation, reverberating back to the listening position...without treatments which at these frequencies are much harder to control.
I say maybe the line array concept might accomplish a balance between the two, which considers the floor and ceiling as significant boundries/reflections as well. Maybe Aperiodic, well damped enclosures for as close to the 'no box' sound that has attracted people to OB/Dipole designs in the first place. It's a tough pill to swallow but i think there's work to be done on both sides.
The discussions between Stig and keyser make interesting points on both sides of the tracks so to speak. For me and my space, directivity via waveguides has been the most successful to my ears....which don't seem to appreciate a 'live' sounding space anymore. Dipole rear wave radiation has proven too complicated for 'ME' and my brain has a hard time sorting out the info, resulting in fatigue quite quickly.
Possibly the right solution might be for heavy absorbtion/treatments behind a dipole and to the outside allowing for 'just' the cancellations to occur leaving only the highly directive remaining material to reach the listener. I haven't explored the issue yet and not sure i'm going to as the CD/WG combo i'm using for the HF is producing some of the most stunning detail and clarity i've ever heard and i'd be hard pressed to part with it.
Now the lower midrange is another story that i've yet to work out. There's been some implementation of the hybrid concept (Emerald Physics) that seems to work well and yet others have stated a strong feeling of disconnect between the octaves as i would imagine room reflections are impacting the listening experience. These subjective reviews seem to fault the WG/CD instead of the OB/Dipole part of the system.......which i find confusing,
Clearly it's difficult to maintain directivity much below 1khz with a single point source driver such as that in the Gedlee and other WG designs so how to we/I overcome this. Dipole bass from 250hz down certainly helps to clean up the bottom octaves with less excitation of room resonances....But what about the 250-1khz range? Is the answer dipole implentation there as well?......In my case i'd be back to the same reflected material that 'escapes' cancellation, reverberating back to the listening position...without treatments which at these frequencies are much harder to control.
I say maybe the line array concept might accomplish a balance between the two, which considers the floor and ceiling as significant boundries/reflections as well. Maybe Aperiodic, well damped enclosures for as close to the 'no box' sound that has attracted people to OB/Dipole designs in the first place. It's a tough pill to swallow but i think there's work to be done on both sides.
Last edited:
Impulse?
Did you ever post an impulse that we can compare to the twest820 project? Does your system show the extreme pre-ringing he is getting? I know that you posted a large group of various measurements of the nearly final configuration but I can't find them by searching.
.
http://www.diyaudio.com/forums/multi-way/175745-three-way-nude-swinging-dipole-thread-4.html
.
Did you ever post an impulse that we can compare to the twest820 project? Does your system show the extreme pre-ringing he is getting? I know that you posted a large group of various measurements of the nearly final configuration but I can't find them by searching.
.
http://www.diyaudio.com/forums/multi-way/175745-three-way-nude-swinging-dipole-thread-4.html
.
No, I never published an impulse response measurement. I will look into that this week, as I am also working on improving both the impulse response and phase response (which is the same thing really).
I dont see any severe pre-ringing with the filter software I'm using now (PLparEQ), but I do remember that Waves LinEq was pretty bad in this respect.
I dont see any severe pre-ringing with the filter software I'm using now (PLparEQ), but I do remember that Waves LinEq was pretty bad in this respect.
Impulse response:
The small peak at 8 ms is the remains of the floor bounce.
Phase response, smoothed for clarity:
Linear Phase filters and EQ have some advantages, as you can see.
The low and top end rolls off below 20 Hz and above 20 Hz, which is the reason for the phase shifts at each end I guess.
Dont know why it shifts somewhat around 150 Hz. Tried to test different XO slopes and delay without much difference.
The small peak at 8 ms is the remains of the floor bounce.
An externally hosted image should be here but it was not working when we last tested it.
Phase response, smoothed for clarity:
An externally hosted image should be here but it was not working when we last tested it.
Linear Phase filters and EQ have some advantages, as you can see.
The low and top end rolls off below 20 Hz and above 20 Hz, which is the reason for the phase shifts at each end I guess.
Dont know why it shifts somewhat around 150 Hz. Tried to test different XO slopes and delay without much difference.
Any analog
My post disappeared so I will throw it up again. I guess I am just used to looking at impulses from passive cross speakers.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.