Jump to content

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


Chris A

Recommended Posts

6 minutes ago, Supersteff said:

 

Hi RandyH000

 

I think you mistake me for someone else, maybe StabMe

yeah , you're right ---- tx  --

Link to comment
Share on other sites

17 hours ago, gnarly said:

I'm pretty dang certain it isn't electrical xover induced delay

 

Hi Gnarly. I'm just guessing here.

 

Are FIR-filters always with the same delay? Maybe the lower the crossower-frequency the longer the needed delay, due to longer wavelenghts? My thought is, that the delay is given by the high-pass-crossower-frequency, in your case 100 Hz. You could check that hypotese, by creating high-pass-filters with say 200 Hz, and see if the deley gets shorter/lower or stays the same!?

 

I think that for lower frequencies you need more taps, and more taps make more delay. Is that correct?

 

Steffen

Link to comment
Share on other sites

4 minutes ago, Supersteff said:

Are FIR-filters always with the same delay?

 

No. FIR filters have whatever delay you design them to have.

 

Quote

Maybe the lower the crossower-frequency the longer the needed delay, due to longer wavelenghts?

 

For a linear-phase (constant delay) filter, the lower the frequency the greater the delay, but not due to the wavelength. It is due to the longer period of the waveform. Of course, the wavelength is also related to the period of the waveform, so in the end the effect is the same.

 

Quote

I think that for lower frequencies you need more taps, and more taps make more delay. Is that correct?

 

That is generally correct, but there are also many other things that affect the tap count.

Link to comment
Share on other sites

Posted (edited)
1 hour ago, Supersteff said:

 

Hi Gnarly. I'm just guessing here.

 

Are FIR-filters always with the same delay? Maybe the lower the crossower-frequency the longer the needed delay, due to longer wavelenghts? My thought is, that the delay is given by the high-pass-crossower-frequency, in your case 100 Hz. You could check that hypotese, by creating high-pass-filters with say 200 Hz, and see if the deley gets shorter/lower or stays the same!?

 

Hi Steffen,   like Edgar said, a FIR-filter can be made with whatever delay you want. 

 

The delay is simply a function of the size of the filter, it's sample rate, and where in the filter the impulse-peak is placed. (Linear-phase centers the impulse peak; IIR replication puts the peak at or very near the start)

 

Which makes the delay independent of xover frequency. Xover frequencies can be moved up or down with no effect on measured delay. 

So in the experiment you propose, delay will not change at all moving high pass from 100Hz to 200Hz.  A really helpful tuning property, huh? 😀

 

I've come to realize the difference in delays measured, using linear-phase tuning on a number of conventional multi-ways, has always appeared to equal the physical distance between acoustic centers. I say appeared because it has always looked close to what would be expected geometrically.

Not so with the MEHs....i can't wait for someone smarter than me to tell why MIDs closer to the mic, measure further away than the more distant CD.

 

 

 

Quote

 

I think that for lower frequencies you need more taps, and more taps make more delay. Is that correct?

 

Steffen

 

Yep, again like Edgar said.  Basically, it takes time to fix time.  So more taps are needed, doubling per every octave decrease. 

Higher order hpf's and lpf's, as well as higher-Q Eqs, also need relatively more taps.

Edited by gnarly
Link to comment
Share on other sites

I am not in the mood for going through all the posts.

Hypothetical question here: how much can be accomplished with MEH design fully active in an acoustically untreated room?

What are the parameters that matter most in such environment (i.e. ceiling height, room walls ratio etc.)?

Materials of walls are made of (brick, wood, combination of both, something else)?

  • Like 1
Link to comment
Share on other sites

I am not in the mood for going through all the posts.
Hypothetical question here: how much can be accomplished with MEH design fully active in an acoustically untreated room?
What are the parameters that matter most in such environment (i.e. ceiling height, room walls ratio etc.)?
Materials of walls are made of (brick, wood, combination of both, something else)?
Mine sound pretty damn good in a brick room, stone floors, an d a concave brick ceiling. All I have is a large rug., and no other treatments. Now in my room I am going to make acoustic panels, as I have found in the past they can make a bigger improvement than buying new equipment. In addition they allow you to reach the full potential of your equipment. Plus a lot cheaper than buying more equipment.

Sent from my SM-T830 using Tapatalk

  • Like 2
Link to comment
Share on other sites

21 hours ago, parlophone1 said:

Hypothetical question here: how much can be accomplished with MEH design fully active in an acoustically untreated room?  What are the parameters that matter most in such environment (i.e. ceiling height, room walls ratio etc.)?  Materials of walls are made of (brick, wood, combination of both, something else)?

1) The MEH probably is able to handle room acoustics issues better than almost anything else (albeit, you may lose boundary gain if you move them away from the walls/floor at some frequency, and depending on the dimensions of the listening room).  Having directivity control and co-axial output without disruptions in the directivity down to 100 Hz (where boundary gain begins to kick in) is the reason for this.  Even more so than Jubilees, etc. (Perhaps more on this discussion to come.)

 

2) I think that the real issue is watching the reverberation times (RT30, in particular) between 100-250 Hz.  If you lose control of this band's total absorption (usually provided by bass traps), you get a booming sound when the SPL gets to some point--that you can't do anything about.

 

3) You don't want something like this: listening to a live rock concert in a place like the G. Rollie White Coliseum (the former basketball arena on the Texas A&M campus--a structure since torn down).  I never heard any FOH guys able to handle mid-bass "boom" in that horseshoe-shaped arena, because the entire back wall just behind the stage was corrugated steel with no absorption.  You have to control the ringing of these kind of surfaces.  If that back wall was covered with anything acoustically absorptive in the 100-250 Hz band, the problems would be been manageable, I believe. (Of course then they couldn't call it "the Holler House on the Brazos" if they did that, however.) 

 

5220f731d23e6.image.jpg

 

Deep bass can be a problem if you're up against solid brick masonry walls--like in a basement or a solid masonry structure with no wood, plaster lathe construction, or drywall--where the walls basically don't flex.  Most rooms rely on that flex to do membrane absorption below 70 Hz, and there aren't enough people in the audience to substitute for deep bass absorption.  If the walls don't flex at all, you've got a real problem, and it takes really big absorbers to do anything about it.  I think about the Beatles' most notable early venue--the Cavern Club--which was basically in the catacombs of Liverpool--and solid as a rock.  They apparently had to develop a "new sound" in order to avoid the booming sound, mostly devoid of Mr. McCartney's deep electric bass range (which they carried forward in their early recordings, and almost anything below 100 Hz was basically attenuated out):

 

1920px-The_Cavern_of_the_Cavern_Club_(cl

 

Chris

  • Like 3
Link to comment
Share on other sites

Anybody have recommendations for 15” 8ohm drivers for this MEH? I thought the Eminence Kappa 15a looked suitable. Fs 33, qt .31, xmas 4 mm. Very similar to 4ohm 15c

Any other contenders?

Thanks, Ted
 

Link to comment
Share on other sites

3 hours ago, VDS said:

Anybody have recommendations for 15” 8ohm drivers for this MEH? I thought the Eminence Kappa 15a looked suitable. Fs 33, qt .31, xmas 4 mm. Very similar to 4ohm 15c

Any other contenders?

Thanks, Ted
 

Looks like a possibility if you don't go with a Faital driver.

Link to comment
Share on other sites

On 3/30/2021 at 8:36 PM, Chris A said:

You'd be much better off using those 8-ohm woofers wired in parallel (i.e., 4 ohm load). That's probably the only thing that I'd currently change on the K-402-MEH prototype--to go to 8 ohm woofers wired in parallel.

 

I don't have any reservations about woofers unless the T/S parameters of the woofers are far away from the Crites 1526C or K-33 woofers (any model of K-33).  You're looking for 15" woofers that have a free air resonance around 20-35 Hz and Qts of about 0.3-0.7.  If you're inside those limits, the only other thing is the voice coil impedance. The voice coil impedance 8 ohms is better than 4 ohms or 16 ohms, in order to keep the required voltage levels from the amplifiers down, while making use of their increased current capabilities to 4-ohm loads.

 

The JBL 2226H has a free air resonance of 40 Hz, which is a little high, but it will probably be okay.  The Qts is 0.31, which is on the low side, which is probably good.

 

Maybe this answers your question? 

 

I believe there are many woofers out there that can be used in a MEH, especially considering that you need a DSP anyway to cross, boost and attenuate.

 

Steffen

  • Like 1
Link to comment
Share on other sites

9 hours ago, Chris A said:

Deep bass can be a problem if you're up against solid brick masonry walls--like in a basement or a solid masonry structure with no wood, plaster lathe construction, or drywall--where the walls basically don't flex.  Most rooms rely on that flex to do membrane absorption below 70 Hz, and there aren't enough people in the audience to substitute for deep bass absorption.  If the walls don't flex at all, you've got a real problem, and it takes really big absorbers to do anything about it.  I think about the Beatles' most notable early venue--the Cavern Club--which was basically in the catacombs of Liverpool--and solid as a rock.  They apparently had to develop a "new sound" in order to avoid the booming sound, mostly devoid of Mr. McCartney's deep electric bass range (which they carried forward in their early recordings, and almost anything below 100 Hz was basically attenuated out):

 

 

 

Chris

 

Well, I do not have solid bricks, but 15 inch thick clay blocks.

I do have a problem with bass but without measurements not aware how big that problem is.

Little experimentation with absorption made some improvements but not too much.

Any way, my tinkering about all that started when I saw a photo of listening room from Hope pilgrimage, that room with Jubilees seem pretty empty except for some foam absorbers on the walls. 

Link to comment
Share on other sites

1 hour ago, parlophone1 said:

Any way, my tinkering about all that started when I saw a photo of listening room from Hope pilgrimage, that room with Jubilees seem pretty empty except for some foam absorbers on the walls. 

I've been in that room next to the chamber listening to Jubilees over the course of a couple of days.  It's got a left-right asymmetric ceiling height and drywall construction, following the contour of the building's roof itself--which has a low pitch characteristic of metal-and-concrete floor construction industrial buildings of this part of the world.  It is fairly well damped in the bass bands--ostensibly through its inherent membrane absorption construction--and the left-right ceiling height difference wasn't a big remaining factor at midrange frequencies--apparently using foam-type absorption to control reflections, but placed asymmetrically within the room left-right. 

 

Roy mentioned that they did a lot of work in that room acoustically to get it right, but the resulting visual presence wasn't really indicative of that work, I have to say.  It looked like the placement of the absorption was almost random, and not nearly as much absorption as I would have guessed was necessary.  But listening along the side and rear walls (farthest from the loudspeakers) was not as problematic as I've experienced in other similar-sized rooms.  Pretty much the entire room was usable as listening positions.

 

Chris

  • Like 2
Link to comment
Share on other sites

This might be a over cautious question but. in a sealed design, like MEH, does Xmass of the woofer come into play more than in a ported box?  I thought I read that in a sealed box woofers will increase their “travel” as a piston.  Is this theoretical true, but not a real issue ?

Link to comment
Share on other sites

Xmax is much less useful in a sealed (acoustic suspension) design than in a ported enclosure design, precisely because the air inside the sealed box acts as an additional opposing spring force at all frequencies where larger movements of the woofer occur.  Xmax is a "large signal parameter" from the Thiele/Small (T/S) list of parameters, and its usefulness was created by the use of bass reflex designs (as are all parameters in the Thiele/Small papers), but not really sealed box (acoustic suspension) bass bins.

 

If a bass reflex box is given a strong bass signal below the port frequency, only the suspension in the woofer stops it from going over its travel limits, as there is no effective back air pressure from the box to resist the woofer motion (i.e., below the tuned port frequency). So it is very easy to "unload" a woofer from its normal position if trying to boost its output below the port resonant frequency (which is the reason why most manufacturers recommend high pass filters for their bass reflex-ported loudspeakers--which further raises the effects of the audible phase growth at low frequencies). 

 

The only reason why people use bass reflex is because they continue to use passive crossovers without EQ embedded in them (i.e., low-Q notch filters).  They use bass reflex to get a extra 3 dB of output in a very narrow frequency range just below the resonant frequency of the woofer/box system--without having to use correcting EQ in the passive crossover circuitry (which also decreases the effective sensitivity of the entire design using passives). 

 

But they pay a high price for this, however, and it generally shows up as increased phase and group delay response growth.  It also limits the low frequency extension of the reflex ported system vs. the sealed system, because as the frequency descends below the woofer+port resonant frequencies, the phase of the backwave coming out the port(s) begins to cancel the frontwave energy by increasing amounts.  In a closed box system, EQing the box SPL response below resonance doesn't result in this phase cancellation.  In fact, the sealed box system loses SPL at nominally 12 dB/octave below resonance, while the bass reflex loses SPL at 24 dB or more per octave below resonance.

 

Also, in order to get more SPL output below resonance, many/most loudspeaker manufacturers use higher "Q" woofers with higher mass, which also increases the phase and group delay growth growth, and limits high frequency output capability of the woofer/box system. 

 

So you pay a high price for that "extra" 3 dB of extremely limited bass extension from a bass reflex design.  I don't use bass reflex designs for these reasons, simply because I don't use passive crossovers anymore.  It's much better and easier just to add 3 dB (or more) of EQ boost below the sealed box's resonant frequency with a DSP crossover. 

 

Bass reflex also typically doesn't sound very good to my ears.

 

Chris

  • Like 1
Link to comment
Share on other sites

Posted (edited)
On 5/12/2021 at 11:03 AM, Chris A said:

Xmax is much less useful in a sealed (acoustic suspension) design than in a ported enclosure design, precisely because the air inside the sealed box acts as an additional opposing spring force at all frequencies where larger movements of the woofer occur.  Xmax is a "large signal parameter" from the Thiele/Small (T/S) list of parameters, and its usefulness was created by the use of bass reflex designs (as are all parameters in the Thiele/Small papers), but not really sealed box (acoustic suspension) bass bins.

 

If a bass reflex box is given a strong bass signal below the port frequency, only the suspension in the woofer stops it from going over its travel limits, as there is no effective back air pressure from the box to resist the woofer motion (i.e., below the tuned port frequency). So it is very easy to "unload" a woofer from its normal position if trying to boost its output below the port resonant frequency (which is the reason why most manufacturers recommend high pass filters for their bass reflex-ported loudspeakers--which further raises the effects of the audible phase growth at low frequencies). 

 

The only reason why people use bass reflex is because they continue to use passive crossovers without EQ embedded in them (i.e., low-Q notch filters).  They use bass reflex to get a extra 3 dB of output in a very narrow frequency range just below the resonant frequency of the woofer/box system--without having to use correcting EQ in the passive crossover circuitry (which also decreases the effective sensitivity of the entire design using passives). 

 

But they pay a high price for this, however, and it generally shows up as increased phase and group delay response growth.  It also limits the low frequency extension of the reflex ported system vs. the sealed system, because as the frequency descends below the woofer+port resonant frequencies, the phase of the backwave coming out the port(s) begins to cancel the frontwave energy by increasing amounts.  In a closed box system, EQing the box SPL response below resonance doesn't result in this phase cancellation.  In fact, the sealed box system loses SPL at nominally 3 dB/octave below resonance, while the bass reflex loses SPL at 6 dB or more per octave below resonance.

 

Also, in order to get more SPL output below resonance, many/most loudspeaker manufacturers use higher "Q" woofers with higher mass, which also increases the phase and group delay growth growth, and limits high frequency output capability of the woofer/box system. 

 

So you pay a high price for that "extra" 3 dB of extremely limited bass extension from a bass reflex design.  I don't use bass reflex designs for these reasons, simply because I don't use passive crossovers anymore.  It's much better and easier just to add 3 dB (or more) of EQ boost below the sealed box's resonant frequency with a DSP crossover. 

 

Bass reflex also typically doesn't sound very good to my ears.

 

Chris

 

Hi Chris, 

first, may i say yes to a possible prior question regarding my name....  my name is gnarly, mark100 on DIY, and Mark Wilkinson on prosound forums.

 

Thought I'd add a few thoughts re bass-reflex since I've used them at home (and for live sound).

For home, I've built sealed, bass-reflex, and push-pull slot load (PPSL) subs, all using the same 18" driver(s) for comparisons.

I chose the BMS 18n862 based off of Josh Ricci (data-bass) getting very good results using it, in both sealed and vented DIYs.

 

The sealed was certainly the easiest to build and get great results.  A no brainer really. 

Adding EQ to the bottom was a bit trickier than first appeared. I ended up using a Linkwitz transform ala the miniDSP route.

The tricky part was xmax  increased fast, given the gain it took to flatten response to as low as I wanted, 30Hz.

But as long as I keep the sealed at reasonable SPL, it is simply super. 

 

Then I built the bass-reflex.  Hornresp led me there, but it definitely still took some trial and error with  port area and length, to get to 30Hz and keep port velocity within bounds.

Must get right, takes work, but once accomplished, it rocks. Requires so much less amp power than the sealed, and sounds just as good to my ears.

 

Interestingly, I found the phase traces / group delay of the bass-reflex, and the sealed after the necessary EQ to get to the same f-3 response, were very similar.

Something we don't normally hear about.

 

Oh, i've never encountered anyone using a passive high-pass for a bass reflex to prevent it from overloading below tuning.  Come to think of it, I haven't heard of anyone using anything but active for bass-reflex....but I'm less in tune with home audio than live sound on this. Your post about folks using BR because they continue to use passives took me by big surprise.

 

I'll abbreviate any comments about the PPSL subs other than to say the motivation behind them was purely for cabinet vibration reduction, and the hope of needing less extensive internal  bracing.  Happy to say it worked awesomely!  Opposed subs is the only way I'll fly anymore. 

I've noticed the prelims on Danley Hyperion indicate dual opposed subs, covering 14-45Hz. 

It also appears he's probably going bass-reflex from the pict...as it looks like there's a hole in the bottom of the cabinet, hence the funky bottom plates and feet.

 

 

 

Edited by gnarly
  • Thanks 1
Link to comment
Share on other sites

1 hour ago, gnarly said:

The sealed was certainly the easiest to build and get great results.  A no brainer really. 

Adding EQ to the bottom was a bit trickier than first appeared. I ended up using a Linkwitz transform ala the miniDSP route.

The tricky part was xmax  increased fast, given the gain it took to flatten response to as low as I wanted, 30Hz.

But as long as I keep the sealed at reasonable SPL, it is simply super. 

 

Then I built the bass-reflex.  Hornresp led me there, but it definitely still took some trial and error with  port area and length, to get to 30Hz and keep port velocity within bounds.

Must get right, takes work, but once accomplished, it rocks. Requires so much less amp power than the sealed, and sounds just as good to my ears.

 

Interestingly, I found the phase traces / group delay of the bass-reflex, and the sealed after the necessary EQ to get to the same f-3 response, were very similar.

Something we don't normally hear about.

 

So much here to talk about! I modeled the 18n862 in WinISD, and the lowest cutoff frequency occurred with the QB3 alignment; 31.5 Hz in 171.15 liters. In that same volume you can fit two 18n862 drivers in a sealed box, and still achieve a Qtc of 0.737 before stuffing. If you then Linkwitz-Transform that for a Qtc of 0.707 at 31.5 Hz, the attached graphic shows the resulting frequency responses. (I'll continue in the next message.)

 

18n862 Frequency Response.png

  • Thanks 1
Link to comment
Share on other sites

Now we can compare the cone excursion of one driver in the vented box with two drivers in the sealed box. It is smaller at around 30 Hz in the vented box because that's what vented boxes do. But note how quickly it rises at lower frequencies.

 

18n862 Cone Excursion.png

  • Thanks 1
Link to comment
Share on other sites

Finally, let's look at impulse responses. The trace labeled Butterworth4 is the vented box. The one labeled Butterworth2 is the sealed box. The difference that you see is the difference between "slow bass" and "fast bass".

 

Vented Impulse Response.png

Sealed Impulse Response.png

  • Like 1
  • Thanks 1
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...