Jump to content

A K-402-Based Full-Range Multiple-Entry Horn


Chris A

Recommended Posts

 

Chris,

 

I looked at my 402s last night and the approximate area you drilled the holes and mounted the woofers.  It appears the flat surface is only about 10 inches meaning the woofer is bigger than the mounting surface.........discharging into the box over the exterior top and bottom of the horn.  Is this an issue?

 

Does Danley seal his drivers to their mounting surface so they discharge only into the holes?

Yes on the mid and high frequency drivers the woofers however are not sealed in the back as you can see in the photo the midrange's are sealed  these is a SPL td-1 that I refurbished it is the early version of the SH46.

 

 

That all mounts into a sealed box though, doesn't it?

 

Bruce

Link to comment
Share on other sites

Perhaps if I understood how to use a phase plug with this type of multiple-entry horn design, I might put that on my list of things to do. 

 

This is venturing outside the realm of what I've played with, but I had a few thoughts on the phase plug:

 

1) The volume inside the "pyramid" created by the shape of the driver diaphragm reacts with the "port" of the throat opening to create a low pass filter. The bigger the volume and the smaller the hole, the lower in frequency the low pass occurs. You can see this in the hornresp model by adjusting VTC. You'll note the high frequency corner moves around. Start with an arbitrarily high VTC and you'll notice an early HF rolloff. As you decrease VTC, the HF rolloff happens higher and higher. Sometimes you you can get to a point where you hit the peak corner, and then it starts going down again. With the offset horns, you'll have a secondary reflection off the throat messing with the total phase seen at the port opening - I've always wondered if a "tapped horn" approach would be an effective solution to extend the HF bandwidth a bit further: Try to balance the reactance of the throat chamber to the "reactance" of the reflection. I wonder if Danley hasn't alluded to that in the past.

 

2) Roy talks about extending the horn through the phase plug. I'm not sure how much another 2" or so matters from the perspective of the bandwidth you're using, but with an offset throat you will have some phase misalignment between the two ends of the driver. (not talking about the offset from the throat of the K402, but the offset inside the "VTC" where one end of the driver is closer to the throat than the other side). Adding a ridge inside the throat to extend the path length of the close end of the driver will improve the phase coherency. There will always be some pathlength differences, so again this is about extending the HF bandwidth and its importance is relative to the bandwidth you need from each portion of the system.

 

3) At higher SPLs (although I think it applies to some extent at all SPLs), you're going to get more "port chuffing" due to the sharp transitions happening at the throat opening. Smoothing those out like a flared port will improve the laminar flow, and ultimately reduce distortion. You can see the woofer excursion increase quite a bit as you get below 60Hz, and the air velocity is going to follow a similar shape. One unfortunate thing about the synergy horn approach is that this air turbulence shares the same air that the HF information is travelling through. The air nonlinearity is definitely an exponential effect (so it gets exponentially less important as the SPL goes down). One of these days I'd like to get around to measuring the significance of this. I know it happens, I just don't know how bad it is. I know I've heard the effect when placing a classic port near a tweeter - it sounds very much like FMD, or adding a tremolo to the higher frequencies. Move the port away and the problem goes away.

 

Along the lines of 3, have you considered porting the LF driver to reduce its low frequency excursion requirements? It sounds like the horn portion has all that freedom with the rear volume, so maybe you could add some port holes and reduce the LF excursion - you could even target the same bandwidth. With a properly sized port, you can add a lot of EQ without increasing driver excursion - and a good port can be made to be fairly linear too.

 

 

Anyways, just thinking out loud since it seems you're interested in more experimentation.

 

 

Ok, one more thought since you brought up the other direct radiating bass bins from Klipsch. I've done a ton of measurements of all sorts of box shapes and sizes and I always run into standing waves happening inside the box. I've also run into reflected energy inside the cabinet bouncing its way out through the port openings. With every design, adding a carefully placed pillow or two has dramatically improved the sound of every ported cabinet I've played with. Even with a sealed cabinet I've seen improvements, and I think that is due to energy leaking through the diaphragm of the driver. Although the sound is naturally attenuated a lot by the driver, you still see it show up in the measurements. I didn't start correlating this behavior until I started doing anechoic measurements. The room totally swamps the frequency response accuracy, but I don't think all frequency response aberrations have equal audibility. In other words, room acoustics induced dips/peaks sound different than internal cabinet resonances / reflections...and the latter is more annoying in my opinion.

 

 

Btw, I'm not familiar with the SynTripP design. I see on Page 138 he's using a wooden phase plug:

http://www.diyaudio.com/forums/multi-way/264485-syntripp-2-way-2-part-virtual-single-point-source-horn-14.html#post4523799

That's not quite what I had in mind for your application, but then I don't know all your dimensions either.

Link to comment
Share on other sites

 

 

Chris,

 

I looked at my 402s last night and the approximate area you drilled the holes and mounted the woofers.  It appears the flat surface is only about 10 inches meaning the woofer is bigger than the mounting surface.........discharging into the box over the exterior top and bottom of the horn.  Is this an issue?

 

Does Danley seal his drivers to their mounting surface so they discharge only into the holes?

Yes on the mid and high frequency drivers the woofers however are not sealed in the back as you can see in the photo the midrange's are sealed  these is a SPL td-1 that I refurbished it is the early version of the SH46.

 

 

That all mounts into a sealed box though, doesn't it?

 

Bruce

 

Yes on the td-1 however some of the Synergy horns have ports.

  • Like 1
Link to comment
Share on other sites

Yes, cast aluminum would be a nice material to use for a single integrated casting.  However, I can candidly say that's a bit out of my current manufacturing capabilities here.  But it certainly would have the desired material characteristics if you've got access to that sort of manufacturing capability.  You'd have to machine the throat area unless you're using an investment casting.  There are other materials that also will work that don't require access to a foundry, and machining.

 

I'm already using a "powered electronic crossover", although it's a general purpose unit.  Xilica used to market OEM crossovers for dedicated use that would be part of a powered loudspeaker approach, but I've found that most hi-fi enthusiasts want to be able to choose their own amplifiers, hence the use of general purpose loudspeaker crossovers just downstream of your preamp and just upstream of your power amplifiers.  It's the most flexible and straightforward approach, I've found.  That's what I use for each speaker in my 5.2 array, and I can tell you that has been a real strength of my setup to be able to dial in each loudspeaker after any changes are made. [bTW: I don't recommend using miniDSP, Behringer, or dbx units due to fidelity and/or SNR issues. ]

 

Surrounds are actually more critical than I believed, as recently as last year when I bi-amped and calibrated mine.  I'd go with the same configuration all the way around since I do listen to multi-channel music that uses each channel independently of the others, i.e. all 5 channels reproduce with the same fidelity and coverage angles, like THX and Dolby have advocated in the past for HTs.  Using surrounds with different coverage angles than the fronts usually results in a timbre shift, I've found.  It's pretty important to match timbres and coverage angles.  YMMV.

 

But if you wish to keep things simple, a stereo pair of these loudspeakers would work, bi-amped, situated in a listening room just like La Scalas or Cornwalls along a wall or in the corners.  That's pretty simple.

 

Chris

Which Xilica are you using?

Link to comment
Share on other sites

Chris,

I read through most of the Bill Waslo stuff on the DIY Sound Group forums. They don't want to go any larger than an 8 inch woofer because they say there is not enough room on the horn (maybe up to a 10 inch). I think they need to do as you did and use a larger horn and cross lower. They certainly aren't going as large as a K402. :o

 

Bruce

I had posted on there about it too, and I told them I could go as large as a 3' x4' horn and they still said a 10" wouldn't work...obviously it would though.

Link to comment
Share on other sites

Chris,

Did you make a filler plug or whatever they call it for the woofers when you mounted them on the K402 sides? Attached is a pic of one on that SynTripP build witha10 inch woofer.

 

 

post-5045-0-97500000-1452794773_thumb.jp

  • Like 1
Link to comment
Share on other sites

Chris,

Did you make a filler plug or whatever they call it for the woofers when you mounted them on the K402 sides? Attached is a pic of one on that SynTripP build witha10 inch woofer.

So a plug blocks all the woofer except for the slots... What would happen if you made a cone that started as the size of the woofer, and on the other end was the size of the slot (reverse horn)?

Link to comment
Share on other sites

A foundry sand cast aluminum K402 with gasket and bosses for mounting the K33s would offer the stiffness desired. Before you know it a powered internal dedicated electronic crossover and amplification will be next. Three across the front and inexpensive powered surrounds,,,

Not a bad idea, if I do say so myself. I already have all that, except, the surrounds aren't self powered and everything is passive.

Edited by ClaudeJ1
  • Like 1
Link to comment
Share on other sites

It appears Danley is quite a way out there on the horn with his woofer ports.  Nowhere near the throat.  If the box itself is sealed then the woofers are sealed.

Was referring to the sealed back part of the driver on the midrange it would be bad if they weren't sealed they would be small passive radiators. :D

Edited by kg4guy
  • Like 1
Link to comment
Share on other sites

Which Xilica are you using?
I'm not, I was referring to the XM-2040, I believe, which was discontinued.  The party has already come and gone, unfortunately.  Bummer.
Link to comment
Share on other sites

Chris,

Did you make a filler plug or whatever they call it for the woofers when you mounted them on the K402 sides? Attached is a pic of one on that SynTripP build with a10 inch woofer.

 

Not yet.  I actually spent some time analyzing it but I figured that I could easily add that feature once I got everything going on the basic design.  That's the major reason why I didn't already epoxy the woofer mounting pads down to the horn body.  As it turns out, the basic design doesn't need it currently, but nevertheless it may help by its addition.  I found that the upper frequency response and phase of the woofer was just about perfect without it.  Note that the performance of the horn itself is very insensitive to changes outside the horn through-ports. 

 

What I've found is that the simplification of the multi-entry horn (unity and synergy-like designs) that eliminate the midrange drivers and instead cross in a 2-way directly to the woofers has significant advantages in terms of complexity and apparent freedom from diffraction issues.  What is interesting to me are the number of people at home designing for the following two performance parameters:

 

1) extremely high SPL capability (really not needed for home hi-fi), thus requiring the use of midrange cone drivers,

2) extremely complex and difficult to implement passive crossovers to handle dual crossover frequencies in the horn, and do passive EQ of its frequency response.

 

Both of these pursuits result in very compromised and time-consuming designs, IMO.   Neither of them are necessary. 

 

Almost no one is designing their DIY multiple-entry horns to handle the mid-bass region and lower into the bass region--as Danley has been doing consistently.  They instead use direct-radiating woofers...which for me kill the deal.

 

YMMV.

 

Chris

Edited by Chris A
  • Like 1
Link to comment
Share on other sites

It also looks like Danley does some fancy corner mounting of the drivers so they discharge out of corners of the horn (2 ports each)

 

Those are bass reflex ports.  Since I don't prefer the sound of ported boxes, and I'm not designing for extremely high SPL capabilities that are required of PA designs, I currently have forgone those extra through-holes in the horn for the sake of fidelity.  YMMV.

 

If it means anything, they certainly can be added later simply by adding port tubes to the back side of the horn and cutting four more holes through the K-402 horn.

 

Chris

  • Like 1
Link to comment
Share on other sites

Chris,

Did you make a filler plug or whatever they call it for the woofers when you mounted them on the K402 sides? Attached is a pic of one on that SynTripP build witha10 inch woofer.

 

Note that the SH96 box only uses small recesses on their woofer mounting surfaces:

 

post-26262-0-30020000-1452805715_thumb.j

 

Chris

  • Like 1
Link to comment
Share on other sites

...I had a few thoughts on the phase plug:

 

1) The volume inside the "pyramid" created by the shape of the driver diaphragm reacts with the "port" of the throat opening to create a low pass filter. The bigger the volume and the smaller the hole, the lower in frequency the low pass occurs...

 

With the offset horns, you'll have a secondary reflection off the throat messing with the total phase seen at the port opening - I've always wondered if a "tapped horn" approach would be an effective solution to extend the HF bandwidth a bit further: Try to balance the reactance of the throat chamber to the "reactance" of the reflection. I wonder if Danley hasn't alluded to that in the past.

 

2) Roy talks about extending the horn through the phase plug. I'm not sure how much another 2" or so matters from the perspective of the bandwidth you're using, but with an offset throat you will have some phase misalignment between the two ends of the driver.

 

(not talking about the offset from the throat of the K402, but the offset inside the "VTC" where one end of the driver is closer to the throat than the other side). Adding a ridge inside the throat to extend the path length of the close end of the driver will improve the phase coherency. There will always be some pathlength differences, so again this is about extending the HF bandwidth and its importance is relative to the bandwidth you need from each portion of the system.

 

3) At higher SPLs (although I think it applies to some extent at all SPLs), you're going to get more "port chuffing" due to the sharp transitions happening at the throat opening. Smoothing those out like a flared port will improve the laminar flow, and ultimately reduce distortion. You can see the woofer excursion increase quite a bit as you get below 60Hz, and the air velocity is going to follow a similar shape. One unfortunate thing about the synergy horn approach is that this air turbulence shares the same air that the HF information is travelling through. The air nonlinearity is definitely an exponential effect (so it gets exponentially less important as the SPL goes down). One of these days I'd like to get around to measuring the significance of this. I know it happens, I just don't know how bad it is. I know I've heard the effect when placing a classic port near a tweeter - it sounds very much like FMD, or adding a tremolo to the higher frequencies. Move the port away and the problem goes away.

 

4) Along the lines of 3, have you considered porting the LF driver to reduce its low frequency excursion requirements? It sounds like the horn portion has all that freedom with the rear volume, so maybe you could add some port holes and reduce the LF excursion - you could even target the same bandwidth. With a properly sized port, you can add a lot of EQ without increasing driver excursion - and a good port can be made to be fairly linear too...

 

5) Ok, one more thought since you brought up the other direct radiating bass bins from Klipsch. I've done a ton of measurements of all sorts of box shapes and sizes and I always run into standing waves happening inside the box. I've also run into reflected energy inside the cabinet bouncing its way out through the port openings. With every design, adding a carefully placed pillow or two has dramatically improved the sound of every ported cabinet I've played with. Even with a sealed cabinet I've seen improvements, and I think that is due to energy leaking through the diaphragm of the driver. Although the sound is naturally attenuated a lot by the driver, you still see it show up in the measurements. I didn't start correlating this behavior until I started doing anechoic measurements. The room totally swamps the frequency response accuracy, but I don't think all frequency response aberrations have equal audibility. In other words, room acoustics induced dips/peaks sound different than internal cabinet resonances / reflections...and the latter is more annoying in my opinion.

 

Btw, I'm not familiar with the SynTripP design. I see on Page 138 he's using a wooden phase plug:

http://www.diyaudio.com/forums/multi-way/264485-syntripp-2-way-2-part-virtual-single-point-source-horn-14.html#post4523799

That's not quite what I had in mind for your application, but then I don't know all your dimensions either.

 

 

Much here to respond to.  I'll briefly address each point, and perhaps come back and amplify my thoughts later:

 

1) I agree that there is a phase issue within the woofer compressed volume - with a 12 inch diameter and crossing at 475 Hz, you're getting close to a half wavelength.  While I didn't see anything in the data for the bass performance indicating phase issues, I'll certainly put it on my to-do list to try out a "phase plug" on the woofer compressed volume to see if there are any changes. 

 

2) That can be addressed by the approach taken in #1.

 

3)  I already rounded the ports on the back compression side of the woofers for the reasons you highlight.  That is pretty well discussed in the multiple-entry horn literature online. Good point, though.

 

4) I addressed that above. When I get to the point where I have a horn to spare, I'll try cutting ports.  I believe that Danley is using ports that are tuned to the region around 100-200 Hz, just looking at the pictures and doing the simple calculations.

 

5) I've yet to worry about the batting material inside the assembly cabinet.  Once I get it apart again, I'll try adding to it/moving it around. 

 

Chris

  • Like 1
Link to comment
Share on other sites

 

Chris,

Did you make a filler plug or whatever they call it for the woofers when you mounted them on the K402 sides? Attached is a pic of one on that SynTripP build witha10 inch woofer.

 

Note that the SH96 box only uses small recesses on their woofer mounting surfaces:

 

attachicon.gifpst003.jpg

 

Chris

 

This is not a SH96 cabinet the midrange holes are in the wrong place there are 6 midrange speakers in the SH96HO and the regular SH96 so I'm not sure that the woofer section is correct?

  • Like 1
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...