Jump to content

Room modes: Horn loaded vs direct radiator subwoofer


Tizman

Recommended Posts

Hi All.

In my research on building a pair of subwoofers for an upcoming project, I have come across conflicting opinions on the impact that the type of subwoofers used have on room modes.  Basically, some have said that because horn loaded subs are directional, room modes are less of an issue than they would be in the case of omni-directional direct radiator subs.  My intention was originally to build a pair of straight 30HZ horns to combine with a pair of Jubilee speakers.  My plan then went to using a single 30HZ sub.  This was based on opinions that two horns would be serious overkill, and that the directional nature of the bass from a 30HZ horn would minimize issues with room modes and subwoofer placement as long as the horn was pointed at the listening area.  I have read some of the literature that relates to subwoofer placement in rooms, and much of it recommends two or more subs placed using various strategies, but everything I have found relates to normal subwoofers, not 20 foot long horn subs with 50 square foot mouths.  So what is the deal with the directional nature of straight horn subwoofers and room modes?  I have attached a rough drawing of the room and three possible setups.  The drawing is rough, but in each scenario the Jubilees are placed in the same position, and the subs in the corners, or in the middle as indicated.  As always, your input is much appreciated!  Cheers, Tiz.  

 

IMG_0363.jpg

Link to comment
Share on other sites

I believe that you're confusing the phenomenon of directivity with loading the horn diaphragm through proximity to a room boundary. 

 

In directivity, the horn is larger than the 1/2 wavelength distance across the mouth of the horn (vertically and horizontally).   If the exiting acoustic wave has a longer wavelength (i.e., lower frequency) than 1/2 wavelength, you lose directivity control, and the exiting wave quickly expands to cover a full 360 degrees in three-dimension space.  So directivity is not really a factor in placement of subwoofers in rooms where all subwoofers--including horn-loaded ones of a mouth size that's less than a metre in minimum dimensions immediately lose directivity below ~170 Hz. 

 

Contrasting directivity, the next subject is the loading of the subwoofer diaphragms with a combination of a horn (or perhaps there isn't a horn) with the secondary boundary gain of the room's walls, ceiling, and floor due to placement of the mouth of the horn in a position so that the existing waves can pick up the boundaries of the room as if it is a horn extension.  This occurs whether or not the subwoofer (or woofer in a bass bin that's covering higher frequencies than subwoofer frequencies) has a horn or not.  All you have to do is to place the mouth of the subwoofer or the diaphragm of the direct-radiating woofer near a room boundary.  It then picks up boundary gain, in the same way that a corner horn picks up boundary gain.

 

full-space.jpghalf-space.jpg

quarter-space.jpgeighth-space.jpg

 

The degree of gain picked up by the woofer is dependent on whether or not there is a significant gap between the mouth of the horn/ or diaphragm of a direct radiating woofer and the room's boundaries. Every halving of the space of that the woofer or horn-loaded subwoofer plays into, there is a corresponding rise in SPL (sound pressure level, or apparent loudness) of 6 dB.  That's a full 18 dB of gain for a corner-located woofer or subwoofer over the full-space condition.  That's a huge amount, and it is the reason why we have corner horn loudspeakers.

 

However, if you move the mouth of a horn away from the room's boundaries, you delay the effects of boundary gain until the wavelengths of the exiting acoustic waves are within 1/4 wavelength of the room's boundaries.  So you produce a "suck out" in the frequency response of the subwoofer or woofer (with or without horns) by not using the boundary gain effectively (which is pronounced in the Klipschorn bass bin due to the higher frequencies involved, effectively requiring no gaps between the back of the Khorn bass bin cabinet and the adjacent corner walls, otherwise large a suck-out in SPL will occur at 200-300 Hz). 

 

speaker-boundary-interference-response.j

 

If you place a very long/straight horn subwoofer in an enclosed space, the problem is that it is difficult to get the horn's mouth close enough to a room boundary to not have boundary gain "suck outs".  So the solution is to place the long, straight horn in the room with the mouth facing a wall, the floor, or the ceiling, and well within 1/4 wavelength of the uppermost frequency that the subwoofer is handling--thus negating the reason for having a straight horn in the first place.  So the obvious solution is to use a folded or curved horn subwoofer in the first place, and place the mouth of the horn within a foot or so of at least the intersection of a wall and the floor or ceiling.  The room corner is even better, in that you pick up another 6 dB of boundary gain over a two-boundary intersection (i.e., wall-floor or wall-ceiling).  So you put everything into the room at the right places, and you EQ the peaks in response flat again (measured nominally at the listening positions in the room), and you have horn-loaded bass.

 

Chris

  • Like 1
Link to comment
Share on other sites

To add to what Chris has said:

 

You are talking about a 7 foot by 7 foot mouth if you have a 50 square foot mouth.  Very generally a radiator stops being omni-directional and starts being directional where the size is 1.0 or 0.5 wavelengths.   A 7 foot acoustic wavelength corresponds to about 160 Hz.  Therefore you are not getting directional at all down at the "sub" freqs. 

 

This involves a concept which must be grasped -- as you go lower in freq the size of the speaker shrinks in terms of wavelength!  Turn the knob on that function generator and your speaker gets bigger or smaller!

  

FWIW and IIRC the Bell Labs bass horn in Klipsch Papers had a mouth of about 5 foot by 5 foot and a cutoff of 32 Hz. Approximately 10 feet long with an 8 inch driver.  Accch.  Such time alignment potential.   This BTW is a re-entrant horn like a bullhorn.

 

 The Jubilee has about a 5 square foot mouth and a cut off of 32 Hz if you ignore the initial flare.  The MCM has a 32 Hz flare.

 

Nelson Pass has described "The Claw" bass horn he made during college.  I haven't seen a photo or sketch. He did say that no matter which direction he pointed it, the police showed up.  Maybe that is an unintended lesson on directional properties, or lack thereof. 

 

Let me suggest you read Don Keele's "Whats So Sacred . . . " on his website.  It addresses mid-range CD horns but has calculations and a nomograph on the effect of mouth size, exit angle and directivity, and frequency.  The more narrow the beam you wish to have, the larger the mouth.  The lower the freq you want, the larger the mouth. 

 

And I don't know how this will work out in your room.    You can't avoid standing waves caused by reflections from the opposite wall even if  you can keep sound off the wall behind the speaker. 

 

WMcD

  • Like 2
Link to comment
Share on other sites

ChrisA and WMcD:  Thank you very much for your replies.

 

In my drawing, I show the wall at the bottom creating a room of 8' X 30'.  This was for the purpose of showing a specific example.  However, this room is not built yet, and I do have a clean 30' X 50' room to work with.  I can put the horn entirely behind this wall, or fold the horn somewhat so it fits in the corner.  Either way, I can have the mouth come out at the boundary, rather than at some point beyond the boundary, in order to avoid back wall cancellations as referred to by ChrisA....  

5 hours ago, Chris A said:

If you place a very long/straight horn subwoofer in an enclosed space, the problem is that it is difficult to get the horn's mouth close enough to a room boundary to not have boundary gain "suck outs". 

 

 

 

4 hours ago, Chris A said:

So directivity is not really a factor in placement of subwoofers in rooms where all subwoofers--including horn-loaded ones of a mouth size that's less than a metre in minimum dimensions immediately lose directivity below ~170 Hz. 

1 meter is roughly a half wavelength of 170 HZ.  The horn mouth that I can use in this room can be as much as 3 meters high, by whatever width I choose.  If I make it also 3 meters wide, the half wavelength is around 57 HZ for both width and height.  57 HZ not a subwoofer frequency, but if this system was high-passed with a steep filter at 57 HZ, what would the result be with respect to room modes?  Would a single horn subwoofer placed in the middle pose more problems than one in each corner with respect to room modes?  I could also I make the horn mouth 3 meters high by 5 meters wide, for example, if that helps the improve the directivity at a lower frequency.  3 meters is 57 HZ and 5 meters is 34 HZ half wavelength respectively.

 

3 hours ago, WMcD said:

The lower the freq you want, the larger the mouth. 

 

And I don't know how this will work out in your room.    You can't avoid standing waves caused by reflections from the opposite wall even if  you can keep sound off the wall behind the speaker. 

I can make a very big mouth for this horn.  If I use a single horn in the middle of the room, as in example "B", it can be 30' wide as in example "C".  30' is roughly 9 meters, which corresponds to 19 HZ half wavelength.  

 

Attached is another drawing that takes into account the responses of Chris A and WMcD.  In these drawings, assume the use of horn mouths that are a minimum of 3 meters in height and width.  Example "C" uses the entire width of the room for the horn mouth, and the entire height, but could use a smaller portion of the width if that will result in a better outcome.  "C" might seem crazy, but it is in fact entirely doable.  Once again, your comments and insight are much appreciated.  

IMG_0365.jpg

Link to comment
Share on other sites

Below the Schroeder frequency of a room...

 

fs = 2000 ( T60 / V )1/2 [Hz], where T60 is the reverberation time in seconds (at the Schroeder frequency) and V is the room volume (in cubic meters)

 

...the concept of directivity has little meaning--certainly anything below 80 Hz in a home-sized listening room, which is the typical crossover point for many HT subwoofers . 

 

Could you elaborate why you feel that directivity is required below 100 Hz in a home-sized listening room?

 

Jubilee bass bins maintain their directivity down to 100 Hz due to their mouth size dimensions, at which point the walls, floor and ceiling of the room form the secondary horn, which extend the low frequency output of the Jubilee bass bins down to 31 Hz (-3 dB). 

 

229302952_JubileebassbinSPLandPhaseResponse.thumb.jpg.901bf364dc0920e94dc9aecdc9501307.jpg

 

If you wish to add a subwoofer or subwoofers, the target crossover point (in my experience) is 30-40 Hz, i.e., the subwoofer(s) need to attenuate their output to below that frequency band.  The Jubilee bass bins roll off at ~15 dB/octave, such they require no electrical high pass filters to roll off their output below 31 Hz.

 

Chris

  • Like 1
Link to comment
Share on other sites

6 hours ago, Chris A said:

Could you elaborate why you feel that directivity is required below 100 Hz in a home-sized listening room?

I don’t think that this is the case.  What I am trying to figure out is if having horn subwoofers that have directivity at these lower frequencies changes any requirements for the setup of my room.  Specifically, will using a horn subwoofer with directivity down to 20HZ allow for a setup that uses one such horn instead of two.  How will room modes be affected by using one as opposed to two?  In a general  sense, how does a horn subwoofer that is directional to 20HZ and whose mouth is flush with the front wall behave differently with respect to room modes as compared to a regular direct radiator sub? 

Link to comment
Share on other sites

53 minutes ago, Tizman said:

What I am trying to figure out is if having horn subwoofers that have directivity at these lower frequencies changes any requirements for the setup of my room. 

My experience is that once a full wavelength of sound cannot fit into the minimum dimension of the room, the concept of directivity has no real effect.  For an 8' ceiling, that would be 142 Hz, for 9', that's 126 Hz, etc.  While there may be instances where you might be able to tell if a horn subwoofer in one corner isn't playing, but another subwoofer in another corner is playing, but that's sort of a gross level of "directivity detection" that goes with human hearing abilities.  Once you're below 80 Hz, directivity in a home-sized listening room doesn't have much evidence or meaning.

 

53 minutes ago, Tizman said:

Specifically, will using a horn subwoofer with directivity down to 20HZ allow for a setup that uses one such horn instead of two. 

Not in my experience, because you still have to deal with room modes.  If you want to wash out the effects of room modes in a home-sized listening room, there is something called a "double bass array" (DBA) that requires two sets of subwoofers on opposing walls (i.e., front wall and rear wall) that are set up to be out of phase so that the rear subwoofer array cancels the bass wavelengths generated by the front-of-room subwoofer array.  Here is a Wikipedia page on the subject:

 

https://en.wikipedia.org/wiki/Double_bass_array

 

This was a concept that I first saw about 10 years ago on the AVSforum, but since then all the figures and pictures in that thread have broken links, so the usefulness of that thread is now very low. Here is the original (in German) paper that I translated in 2011 on the subject from the original authors:

 

https://community.klipsch.com/applications/core/interface/file/attachment.php?id=64885

 

53 minutes ago, Tizman said:

How will room modes be affected by using one as opposed to two?

You'll still have severe room modes, i.e., either you're going to have nulls along the mid-points of the room (all mid-wall positions and along centerline of the room), or you're going to have nulls along the corners and toward the outside of centerline, and the efficiency of the subwoofers used in this configuration will be significantly attenuated relative to the first condition mentioned (IIRC--it's a 12 dB penalty). One or the other. 

 

To not have deep bass nulls in the room requires a double bass array.  To minimize the nulls requires at least two subwoofers (three are better).  [I use three subwoofers (effectively) in my room, two on the opposite front corners of the room, and one in the center (actually it's the K-402-MEH, but it has -3 dB response at 16 Hz).]

 

53 minutes ago, Tizman said:

In a general  sense, how does a horn subwoofer that is directional to 20HZ and whose mouth is flush with the front wall behave differently with respect to room modes as compared to a regular direct radiator sub? 

It doesn't--except if you use a double bass array (like having straight horns on opposing ends of the room--wired out of phase).  I'd recommend the DBA instead of two long horns on opposite ends of the room.  I'd also recommend using something like four TH-SPUD designs on the front wall, and another four TH-SPUDs on the rear wall--wired out of phase from the front wall phase.  It would take up much less space and be much easier to build.  That would require 16 Tang Band 8" woofers (W8-740Ps) at about $100 each woofer to fully populate those SPUDs.

 

Chris

  • Like 1
Link to comment
Share on other sites

I'm glad Chris is doing the heavy lifting here.

 

I believe that a horn with output at 20 Hz has not been built, ever.  If it were it would be larger than a house.  Even then its polar pattern would be omnidirectional at sub frequencies unless it is the size of a warehouse.  Of course it is not going to fit in your large living room.

 

WMcD

 

 

  • Like 1
Link to comment
Share on other sites

I always thought that anything other than a full size horn was a compromise, with the various negatives associated to this compromise.  My understanding is that the only reason that full size horns are not built is because they are too big for most situations, not because they are not the best way to go SQ wise.  I am in the unusual situation of being able to build a full size horn.  Or two if that will give a better result.  I feel that this is a good thing.

Link to comment
Share on other sites

9 hours ago, WMcD said:

I believe that a horn with output at 20 Hz has not been built, ever.  If it were it would be larger than a house.

Yes.  A proper straight 20 HZ horn is house size.  If I understand this correctly, and I may not, lots of horns have output at 20 HZ, but do not control directivity at that frequency.  Given my ceiling height of 10 feet/3 meters as a limitation, I can build a horn where directivity is controlled to about 57 HZ in the vertical.  Horizontally, I can build a horn where directivity is controlled to around 19 HZ.  Either way, there will be lots of output at 20HZ.

 

11 hours ago, Chris A said:

I'd also recommend using something like four TH-SPUD designs on the front wall, and another four TH-SPUDs on the rear wall--wired out of phase from the front wall phase.  It would take up much less space and be much easier to build.  That would require 16 Tang Band 8" woofers (W8-740Ps) at about $100 each woofer to fully populate those SPUDs.

I understand that there may be better ways to do this, and thanks ChrisA for your suggestion that I look at DBA and the SPUDS.  I will do so.


My goal here is to get the best sound possible in this room, and have it all be horn loaded.  This project will prioritize SQ above any other criteria.  There will be extensive acoustic treatments used in the room.  For example, if I need to take up  10’ at the other end of the room for bass traps, it will be done.  The goal is to have as large a sweet spot as possible, with the best SQ possible.  All other criteria are far less important.  

Link to comment
Share on other sites

13 hours ago, Chris A said:

If you want to wash out the effects of room modes in a home-sized listening room, there is something called a "double bass array" (DBA)

Thanks for the reference Chris.  I'm not sure how effective I can make it given that the room will contain 60 people or so, and the furniture required to seat these people, in addition to a lot of other stuff.  I may have to make do with a very large bass absorber on the back wall, even though the DBA would be an elegant and far less intrusive solution.  60 people would also make for a pretty good bass absorber.  A very inconsistent one though....

Link to comment
Share on other sites

15 hours ago, Chris A said:

My experience is that once a full wavelength of sound cannot fit into the minimum dimension of the room, the concept of directivity has no real effect.  For an 8' ceiling, that would be 142 Hz, for 9', that's 126 Hz, etc.  While there may be instances where you might be able to tell if a horn subwoofer in one corner isn't playing, but another subwoofer in another corner is playing, but that's sort of a gross level of "directivity detection" that goes with human hearing abilities.  Once you're below 80 Hz, directivity in a home-sized listening room doesn't have much evidence or meaning.

 

Chris

 

Respectfully, I just don't agree with this. Sometimes all this theory is superseded by actual experiences. I had to move my SVS DR sealed sub to a corner behind a chair because my new cornwalls are pretty wide (space limitations). My sub is crossed @ 40 hz. I can tell where that sub is all day, its location when reproducing sound. It's performance was much better in the front corner projecting diagonally into the room with the narrower RP-280 front mains.

 

More proof.......how many users advocate a pair of subs versus a single in some scenarios? 

If the unit output volume is sufficient and clean, why would anybody need two subs in their room if Chris's theory above is true?

Link to comment
Share on other sites

On 1/31/2020 at 7:56 AM, Chris A said:

 

In directivity, the horn is larger than the 1/2 wavelength distance across the mouth of the horn (vertically and horizontally).   If the exiting acoustic wave has a longer wavelength (i.e., lower frequency) than 1/2 wavelength, you lose directivity control, and the exiting wave quickly expands to cover a full 360 degrees in three-dimension space.  So directivity is not really a factor in placement of subwoofers in rooms where all subwoofers--including horn-loaded ones of a mouth size that's less than a metre in minimum dimensions immediately lose directivity below ~170 Hz. 

 

 

 

See above.........so is it < 170 or 80 hz?

Link to comment
Share on other sites

15 hours ago, WMcD said:

I'm glad Chris is doing the heavy lifting here.

I believe that I've lifted my end of the load by now...

 

Before I retire from this thread, I'd offer one point: there is a big difference between acoustic directivity and locatability of bass response below the room's Schroeder frequency.  Below the Schroeder frequency, sound waves don't "travel", but rather inhabit corridors or patterns of space within the room, locked in their geometric places at various frequencies--sort of like looking at a drum head vibrate in various modes or patterns.

 

And with that...

 

Chris

Link to comment
Share on other sites

Coming to this thread late. Reading it for the first time, I think I see a couple of points that have been stated indirectly, but never explicitly.

 

When any of the room dimensions are "small" compared to the wavelength, the room becomes either a waveguide (if the other dimensions are "very large" compared to the wavelength) or a resonant cavity (the more general case). As such, the type of generator (i.e. horn or direct radiator) of the waves is almost irrelevant -- the propagation characteristics and standing wave pattern are controlled by the geometry of the room. Given the 10' ceiling height of the room in question, this starts to occur around 56 Hz. 

 

Also, the directivity of a horn really only has meaning in the far field. In the near field things can be very different. Given the situation described here, the audience is definitely in the near field. They're basically inside the horn.

Link to comment
Share on other sites

3 hours ago, Edgar said:

As such, the type of generator (i.e. horn or direct radiator) of the waves is almost irrelevant -- the propagation characteristics and standing wave pattern are controlled by the geometry of the room.

Thanks for the clarification.  It would appear that the solution is room treatments or a DBA.  On ChrisA’s suggestion, I read up on DBAs.  My concern with them, based on my reading, is that there will be so much stuff in the room that will interfere with the operation of a DBA that it may not work properly.  This is based on what I read, as I have no real world experience with them.  Perhaps a DBA would work out to be the best real world solution..  It certainly appears to be the lease intrusive and most elegant solution.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...