Thanks,
I would like to see the polar response of the Iwata.
This Tractix has not great pattern control.
Itawa 300 the tractrix is even better dispersion. Also a advantage of a short horn.
23" wide 18" long. 10khz 30degrees the tactrix does 40 degrees there.
23" wide 18" long. 10khz 30degrees the tactrix does 40 degrees there.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Hi Jack
By “not great” it sort of would depend on what you were seeking. As you know, all compression drivers have a power response which starts rolling off in the 2-4 KHz range (typical 1 inch driver) and so with “the right” curved wall horn, one can progressively narrow the dispersion pattern to compensate that and if one is lucky / careful one can end up with a measured “flat” response on axis by trading that flat response for constant directivity. This progressive narrowing (the sum of effect of H and V) “appears” to fix the drivers power response and is how many in hifi see it, rather that CD horns need EQ to fix them when in reality, to the degree a horn IS CD, the response reflects exactly the drivers true power response now un-altered by the horns directivity vs frequency..
That is fine if you are always on axis like many home systems but like in commercial sound where the audience plane can be larger than one or two sets of ears up high, one wants constant directivity so that the frequency response does not change with your position when within the horns intended pattern. Also a plus with constant directivity is the reverberant spectrum is essentially the same as the on axis response so that they are very listenable off axis while a typical large non-CD horn system can sound murky and dark off axis.
In commercial sound the reverberant sound can be primarily what limits one’s ability to understand spoken words etc and so here maximizing the directivity and extending it lower in frequency and eliminating the lobes and nulls that source to source interference produces, were what drove the Unity and then Synergy Horn designs. While acting like a single driver in time and space, preserving input waveshape over a broad band and gobs of headroom make them ideal for music, the size and problem of appearance as well as interest at work are issues. I wish they were interested in hifi at work (I could never go back to hifi speakers) because the first obvious thing you notice is there is only one source of sound, even when you stick you head in the horn mouth, you can’t hear separate low, mid or high drivers, just a source floating somewhere in front of you and in stereo in a room, the directivity makes it much like being outside or a window to another space. It’s pretty weird hearing that in a stadium haha but that’s what is wagging the dogs tail at work now ( a big stadium demo later this week).
Best,
Tom Danley
By “not great” it sort of would depend on what you were seeking. As you know, all compression drivers have a power response which starts rolling off in the 2-4 KHz range (typical 1 inch driver) and so with “the right” curved wall horn, one can progressively narrow the dispersion pattern to compensate that and if one is lucky / careful one can end up with a measured “flat” response on axis by trading that flat response for constant directivity. This progressive narrowing (the sum of effect of H and V) “appears” to fix the drivers power response and is how many in hifi see it, rather that CD horns need EQ to fix them when in reality, to the degree a horn IS CD, the response reflects exactly the drivers true power response now un-altered by the horns directivity vs frequency..
That is fine if you are always on axis like many home systems but like in commercial sound where the audience plane can be larger than one or two sets of ears up high, one wants constant directivity so that the frequency response does not change with your position when within the horns intended pattern. Also a plus with constant directivity is the reverberant spectrum is essentially the same as the on axis response so that they are very listenable off axis while a typical large non-CD horn system can sound murky and dark off axis.
In commercial sound the reverberant sound can be primarily what limits one’s ability to understand spoken words etc and so here maximizing the directivity and extending it lower in frequency and eliminating the lobes and nulls that source to source interference produces, were what drove the Unity and then Synergy Horn designs. While acting like a single driver in time and space, preserving input waveshape over a broad band and gobs of headroom make them ideal for music, the size and problem of appearance as well as interest at work are issues. I wish they were interested in hifi at work (I could never go back to hifi speakers) because the first obvious thing you notice is there is only one source of sound, even when you stick you head in the horn mouth, you can’t hear separate low, mid or high drivers, just a source floating somewhere in front of you and in stereo in a room, the directivity makes it much like being outside or a window to another space. It’s pretty weird hearing that in a stadium haha but that’s what is wagging the dogs tail at work now ( a big stadium demo later this week).
Best,
Tom Danley
I'll let Tom Danley chime in here after me but there is one thing that I keep seeing here in audiophile threads that I can't fathom in horn or waveguide design. Why are so many people so enamored of the round horns? I understand they are easy to construct but in a room such as a home they make very little sense. Why you would want the same vertical dispersions as horizontal is beyond me. You have created a problematic situation with as much reflected energy from the floor and ceiling as you have in the limited horizontal dispersion that you have now created. I would prefer a radial horn to that, even with it's lobing problems, or the Danley solution of at least a rectangular mouth that can have different vertical and horizontal dispersion dependent on use. In professional sound applications you could desire a different pattern than in a home as Tom has talked about, but at least there is some thought to the actual use in the space. Unless you want to tie yourself to a limited seating area and few listeners it just doesn't make sense to use round horns. Another problem with stacked round horns are the integration of the different centers being at the maximum distance apart. I'll stop here and see if Tom will chime in on my thoughts.
Helmuth,
If that was the case you would only need one playback device. That is far from reality and microphones do not normally have a perfectly round pick up pattern. What you are talking about is only a hypothetical argument, not based in any reality of recording at all. Where are you supposed to get this supposed source material. this is the problem with audiophiles, they live in a fantasy world that is just not real.
If that was the case you would only need one playback device. That is far from reality and microphones do not normally have a perfectly round pick up pattern. What you are talking about is only a hypothetical argument, not based in any reality of recording at all. Where are you supposed to get this supposed source material. this is the problem with audiophiles, they live in a fantasy world that is just not real.
In recordings they can do many things work and they do.
Omnidirectional
Supercardioid
Cardioid
Figure of Eight (bidirectional)
Omnidirectional
An externally hosted image should be here but it was not working when we last tested it.
Supercardioid
An externally hosted image should be here but it was not working when we last tested it.
Cardioid
An externally hosted image should be here but it was not working when we last tested it.
Figure of Eight (bidirectional)
An externally hosted image should be here but it was not working when we last tested it.
No, this is how mics work. And not "some graphic artist", but from the manufacturer themselves.
What do you perceive to be a falsehood about this?
Just as we have polar plots of speakers and horns the same goes for a microphone. They do not have these perfect polar responses. That is just not real or we would have perfect reproduction from a microphone. Why do you think that artist use selective microphones, they all sound different. if this was the case you would only have to have a couple of types of microphone as they would all be perfect. Stop living in dream land, look at a real response curve on any microphone and also its pickup patterns. Not what is being shown here. That is only an idealized situation, not representative of reality! This is why audiophile try to do some of the things they do, they believe we live in a perfect world with perfect reproduction devices. Just as that is not true in any loudspeaker the same goes for a microphone for the same reasons. The laws of physics are such that there are more things going on than an idealized concept. Those are just examples of the overall types of directivity patterns. I am not making this stuff up, go read a technical journal on microphone design or microphone analysis and you will find out what I am saying is true. I wish it wasn't but it is. Even laboratory microphones are often not as good as you seem to believe.
Sorry Pano....... I would make a perfect microphone for you if I could, but the rest of these guys need to have some lessons in reality! So much information is taken so out of context it is unbelievable what some believe. I know you are smarter than that so you can disregard my comments!
...
Hence, why back in the late 70's and during the 80's and early 90's, many pro reviewers used the measurements of equipment, especially speakers, as the absolute word on how they performed. Then they realized that many times the measurements looked great, but didn't tell the whole story. Some looked great, but actually didn't sound as good as it's measurements suggested, and others didn't look that good on paper, but were actually very good.Signal measurements are a great tool, and tell a big part of the story, but they have to be properly checked and interpreted.
Hi Kindhornman
Round horns have the up side of being fully symmetric and if small enough, can be turned on a wood lathe resulting in some fine looking horns. Being fully symmetric, they follow a simple progression of area’s and are what was used in the old days when one wanted the maximum “throw” (greatest energy in the smallest beam). As the hifi interest in horns has been in good part a result of retired commercial sound equipment finding it’s way into the home, it is perhaps something like country music was (just getting to the long hair phase a few years back).
Nothing is perfect, everything has limits, boundaries and especially when we are dealing with sound, we have a phenomena that exists in 3 dimensions while our conceptual “oscilloscope view” of signals is one dimensional, applicable for the electrical domain but far short in airborne sound..
With perfect symmetry, one finds any spatial problems are maximized.
If you do a high resolution polar plot across the pattern of round horn, you usually find some funny stuff up high. Some have a "black dot" or dead spot on axis or other stuff like that.
If you have anything other than simple axial mode radiation from the driver and at a dimension suitable for the frequency, those artifacts a focused at the center by that symmetry.
The earliest attempts to get “Constant directivity” was the multi cell horn, these were what I think were the most beautiful and sadly hard to build of the horn art. Given how bad the early CD horns often sounded, I am sure the move was mostly a cost to build based one.
Where the move to spatially designed horns began (away with the curved wall, pattern control designed) was here; #7 “What’s so sacred about exponential horns”.
AES Papers -- Official website of D.B.Keele
What you can see is that for a given horn wall angle and mouth dimension, there is a pattern control loss frequency where that wall angle no longer confines the radiation angle. Also, if you double the horn angle with the same mouth dimensions, that the pattern control loss point drops an octave.
The paradox of the asymmetric horn becomes clear when you say ok, I want to loose pattern control at the same frequency in H and V BUT I want to make the angles different. Loosing pattern control at different frequencies is not good because it can cause “pattern flip” where the formerly narrow angle is not the wide one. If you have ever seen a T-35 mounted up and down, it is because over most of it’s range the “wide” patterns is what would be the vertical and that because below about 10K it’s in pattern flip.
So lets say you had a horn that had a 90H by 30V degree pattern, if you made it the normal simple way of a simple conical horn, you get a height much smaller than the width. Do the math and you find the mouth has to be 3 times the width, not smaller in order not to have that problem.
That requirement lead to the diffraction slot horn, some of which had sonically disastrous shapes a result of trying to improve the driver loading and being on new ground.
It was a comment Don Davis had made at a synaudcon in the late 80’s about conical horns have good directivity but poor lf loading that popped into my head about 12 years later. I was starting over in the speaker biz and was faced with making a speaker for commercial and live sound. It dawned on me why the conical horn had poor lf loading. The impedance transformation that gives a horn it’s efficiency has a “high pass” corner to it based on how fast the area expands. For example, a 30Hz horn cannot double any faster than doubling it’s area about every 24 inches while a 300 hz horn doubles every 2.4 inches. When you plot the rate of expansion for a conical horn, you find it is very fast at the apex and progressively slows as you move towards the mouth. It would not load a driver well low in it's range. I thought what if I put mid drivers suitable for horn loading on the horn sides where the expansion was more like a midrange. That began the Unity horn work about 14 years ago and now Synergy Horns.
What I have found is that one can get away with some asymmetry on a simple conical horn but if a horn is to be used alone or has to be asymmetric, then you have to follow what Don’s rule implies.
For our big horns at work, the problem is a lot more complicated, the whole point is to produce the wave front of a single source and doing that is pretty easy when you can put the drivers less than ¼ wl apart but the highs were always a problem.
This was perhaps the most difficult technical problem I have dealt with in modern times, “how to add more than one hf driver and radiate as one into a horn”. I thought I saw how to do it while at a trade show and told my partner Mike “I think I see a way to do it” but it took 3 months of pondering, sketching and being afraid I had “mispoken” and a month going back and forth with our 3d cad guy before we had one..
Patent US20120328140 - Horn enclosure for combining sound output - Google Patents
In a couple of the J’s, the throat diffraction issue is dealt with by driving the throat in layers where the center radiation is first with the cells next to it delayed slightly etc, the wavefront is already “bent” in both H and V when it enters the last visible part of the horn.
Anyway, for the diy’r, the conical horn is also more picky about the driver internal geometry as the driver is partly in control of the pattern up high. One I have found to be very good on conical horns and can go unusually low is the BMS 4550, we use a good number of those at work.
Best,
Tom
Round horns have the up side of being fully symmetric and if small enough, can be turned on a wood lathe resulting in some fine looking horns. Being fully symmetric, they follow a simple progression of area’s and are what was used in the old days when one wanted the maximum “throw” (greatest energy in the smallest beam). As the hifi interest in horns has been in good part a result of retired commercial sound equipment finding it’s way into the home, it is perhaps something like country music was (just getting to the long hair phase a few years back).
Nothing is perfect, everything has limits, boundaries and especially when we are dealing with sound, we have a phenomena that exists in 3 dimensions while our conceptual “oscilloscope view” of signals is one dimensional, applicable for the electrical domain but far short in airborne sound..
With perfect symmetry, one finds any spatial problems are maximized.
If you do a high resolution polar plot across the pattern of round horn, you usually find some funny stuff up high. Some have a "black dot" or dead spot on axis or other stuff like that.
If you have anything other than simple axial mode radiation from the driver and at a dimension suitable for the frequency, those artifacts a focused at the center by that symmetry.
The earliest attempts to get “Constant directivity” was the multi cell horn, these were what I think were the most beautiful and sadly hard to build of the horn art. Given how bad the early CD horns often sounded, I am sure the move was mostly a cost to build based one.
Where the move to spatially designed horns began (away with the curved wall, pattern control designed) was here; #7 “What’s so sacred about exponential horns”.
AES Papers -- Official website of D.B.Keele
What you can see is that for a given horn wall angle and mouth dimension, there is a pattern control loss frequency where that wall angle no longer confines the radiation angle. Also, if you double the horn angle with the same mouth dimensions, that the pattern control loss point drops an octave.
The paradox of the asymmetric horn becomes clear when you say ok, I want to loose pattern control at the same frequency in H and V BUT I want to make the angles different. Loosing pattern control at different frequencies is not good because it can cause “pattern flip” where the formerly narrow angle is not the wide one. If you have ever seen a T-35 mounted up and down, it is because over most of it’s range the “wide” patterns is what would be the vertical and that because below about 10K it’s in pattern flip.
So lets say you had a horn that had a 90H by 30V degree pattern, if you made it the normal simple way of a simple conical horn, you get a height much smaller than the width. Do the math and you find the mouth has to be 3 times the width, not smaller in order not to have that problem.
That requirement lead to the diffraction slot horn, some of which had sonically disastrous shapes a result of trying to improve the driver loading and being on new ground.
It was a comment Don Davis had made at a synaudcon in the late 80’s about conical horns have good directivity but poor lf loading that popped into my head about 12 years later. I was starting over in the speaker biz and was faced with making a speaker for commercial and live sound. It dawned on me why the conical horn had poor lf loading. The impedance transformation that gives a horn it’s efficiency has a “high pass” corner to it based on how fast the area expands. For example, a 30Hz horn cannot double any faster than doubling it’s area about every 24 inches while a 300 hz horn doubles every 2.4 inches. When you plot the rate of expansion for a conical horn, you find it is very fast at the apex and progressively slows as you move towards the mouth. It would not load a driver well low in it's range. I thought what if I put mid drivers suitable for horn loading on the horn sides where the expansion was more like a midrange. That began the Unity horn work about 14 years ago and now Synergy Horns.
What I have found is that one can get away with some asymmetry on a simple conical horn but if a horn is to be used alone or has to be asymmetric, then you have to follow what Don’s rule implies.
For our big horns at work, the problem is a lot more complicated, the whole point is to produce the wave front of a single source and doing that is pretty easy when you can put the drivers less than ¼ wl apart but the highs were always a problem.
This was perhaps the most difficult technical problem I have dealt with in modern times, “how to add more than one hf driver and radiate as one into a horn”. I thought I saw how to do it while at a trade show and told my partner Mike “I think I see a way to do it” but it took 3 months of pondering, sketching and being afraid I had “mispoken” and a month going back and forth with our 3d cad guy before we had one..
Patent US20120328140 - Horn enclosure for combining sound output - Google Patents
In a couple of the J’s, the throat diffraction issue is dealt with by driving the throat in layers where the center radiation is first with the cells next to it delayed slightly etc, the wavefront is already “bent” in both H and V when it enters the last visible part of the horn.
Anyway, for the diy’r, the conical horn is also more picky about the driver internal geometry as the driver is partly in control of the pattern up high. One I have found to be very good on conical horns and can go unusually low is the BMS 4550, we use a good number of those at work.
Best,
Tom
Tom,
I appreciate your answer here. I too have been involved in the design of waveguides for a long time. There are many trade offs with any design, it is the balance that we are trying to achieve that gives us the best result for a given application. I understand how easy it is to create a round horn, things are just simpler for the mathematics and that seems to be a driver behind many of the designs. I know the Webster papers and also many early papers by Paul Klipsch. My original work was on the connection of the cone driver to the throat expansion area of a horn with a throat smaller than a cone drivers effective cone area. I can't say that I have heard your devices but I do understand the principal of what you are doing. We really are very much of the same thought in many areas of design. I guess I am just showing my prejudice against a round horn and its usefulness in a small room, especially with multiple horns and the combination of those horns. The floor reflections from a round horn just outweigh the simplicity of the design to me in a hi end system for a home. I grew up with PA applications and also horn systems in my house. I still have a pair of old Altec Barcelona speaker cabinets that I have been afraid to modify just because of their collector value.
Steven
I appreciate your answer here. I too have been involved in the design of waveguides for a long time. There are many trade offs with any design, it is the balance that we are trying to achieve that gives us the best result for a given application. I understand how easy it is to create a round horn, things are just simpler for the mathematics and that seems to be a driver behind many of the designs. I know the Webster papers and also many early papers by Paul Klipsch. My original work was on the connection of the cone driver to the throat expansion area of a horn with a throat smaller than a cone drivers effective cone area. I can't say that I have heard your devices but I do understand the principal of what you are doing. We really are very much of the same thought in many areas of design. I guess I am just showing my prejudice against a round horn and its usefulness in a small room, especially with multiple horns and the combination of those horns. The floor reflections from a round horn just outweigh the simplicity of the design to me in a hi end system for a home. I grew up with PA applications and also horn systems in my house. I still have a pair of old Altec Barcelona speaker cabinets that I have been afraid to modify just because of their collector value.
Steven
You are blowing high from the tower with your reality.Sorry Pano....... I would make a perfect microphone for you if I could, but the rest of these guys need to have some lessons in reality! So much information is taken so out of context it is unbelievable what some believe. I know you are smarter than that so you can disregard my comments!
Sure it is imperfect, only thing is to work to perfection what is in reach. And not be content with the imperfections that can be solved.
"Signal measurements are a great tool, and tell a big part of the story, but they have to be properly checked and interpreted."
So very true, two thumbs up! We have a winner here.
Thanks, in getting involved in DIY I've come to realize that I have to trust my gut feeling, even if sometimes wrong. And, after doing/analyzing all the mathematics and data, to trust my ears.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.