Jump to content

Dual 15 inch Corner Bass Bin - Cornscala dbb


jwc

Recommended Posts

  • Replies 349
  • Created
  • Last Reply

Top Posters In This Topic

----------------

On 7/11/2005 6:49:37 PM DrWho wrote:

----------------

On 7/11/2005 5:22:56 PM IB Slammin wrote:

Guess I am missing it. Why does the verticle narrow? Are the upper waves(for lack of better words) from the top woofer interrupted by the lower woofer?

Thanks,

----------------

Well I don't fully understand the smaller details, but the basic idea is that you're introducing comb-filtering in the higher frequencies as you move off-axis. In the horizontal plane, you can move around and you're always the same distance from each driver. But in the vertical plane, as you move up and down you are now introducing different distances between each driver:

Fig1.gif"

-http://www.prosoundweb.com/live/articles/jbrusi/pa.php

So basically, the different path distances will cause cancellation at frequencies that get shifted a distance of 1/2 wavelength. Frequencies that get shifted slightly less and slightly more also experience this cancellation.

This picture does a good job of showing a 3-D model:

Figure6.jpg"

-http://www.prosoundweb.com/studyhall/am/lobes/lobes2.php

(Note that each balloon plot is for ONE SINGLE frequency and there are two plots shown. Notice that the points of cancellation aren't always the same for different frequencies).

When you look at every graph at the same time, the end result of all the summations and whatnot is a polar response that has narrowed...because everything closer to on-axis tends to remain in phase. The following picture is a bit of an extreme example, but it gets the idea across. Notice the narrow beam going right down the middle.

Fig10sm.jpg"

(Click For Larger Picture)

Here are some array guidelines:

1. For most venues, sub arrays should be stacked vertically to narrow vertical coverage and maintain horizontal coverage.

2. If subs are to be split horizontally, they should be as far apart as possible (>5 wavelengths).

3. If subs MUST be arrayed horizontally, con sider the Bessel array.

4. Cardioid subwoofers can be useful for control ling rear radiation.

5. Remember that the polars are frequency dependent, and a good polar at one frequency will produce the opposite effect at twice (or half) that frequency. For this reason, it may be necessary to use multi-way subwoofer systems with each passband optimized for the desired polar.

6. The room’s response will interact with the subwoofer array response, and can nullify the ben efits of the array.

7. A ridgid boundary near a subwoofer (floor, wall, ceiling or combination) will produce a reflection that can be treated as another low frequency source. The princip of subwoofer spacing apply, and the best coupling takes place within 1/8 wavelength of the sur face, producing a mirror image that couples with the sound source as though separated by 1/4 wavelength.

One thing not explicitly stated is that keeping the speakers within a 1/4 wavelength of the highest frequency to be reproduced exhibit no comb-filtering issues. If you think about it, two sources have the biggest discrepancy when you are 180 degrees off axis (off to the side or up above in this situation). If the most out of phase they can get is 90 degrees (1/4 wavelength) then the most cancellation you get is -3dB at the highest frequency...all the other frequencies will "couple" and approach that 3dB gain you get from using two drivers.

So here are a bunch of links for anyone interested:

-http://www.prosoundweb.com/install/synaudcon/tt26_3/tt26_3.php

-http://www.prosoundweb.com/studyhall/am/lobes/lobes.php

-http://www.prosoundweb.com/live/articles/jbrusi/pa.php

---------------

Doc, I am a simple old construction worker. I do appreciate your efforts and with all due respect; I still can't see it.. other than applied in "free space". If I had a chamber,or if my listening enviornonment were outside...... I consider everything to be "on axis" bcause I can adjust it. Toe in or out. I can adjust ME to the sweet spot. Please considder the enviornonment of real world listening. Even in a large room, four CWs, six LaS, two cornerhorns and four LaS, six Cornerhorns, ...... Please describe what is "on or off axis" with SPLs of over 120db.

Ahha, my pure little two channel life.......so uncomplicated....

Regards,

Terry

2.gif2.gif

Link to comment
Share on other sites

----------------

On 7/11/2005 7:48:40 PM jwcullison wrote:

I haven't seen woofers placed horizontal. However, I did consider such a cabinet. It would be very wide at least 37-38 inches. I even calculated some cabinets where they weren't horizontal or vertical but more oblique from each other. This concept made it hard to build two different enclosures in one cabinet.

----------------

Yikes, horizontal would have been worse than going vertical (so it's a good thing you went the way you did).

----------------

I haven't considered the 2.5 crossover. Please feel free to explain some advantages there.

----------------

How far apart are the center points of your woofers? I made a rough guess of about 18 inches which puts your 1/4 wavelength at about 200Hz...so you're getting a full octave of sound (400Hz crossover right?) where you're starting to get beaming. The beaming in the vertical plane might actually be a good thing though because it would start to reduce floor and cieling reflections above 200Hz. However, 15" woofers tend to be narrower in the upper frequencies already, so it's probably getting a bit thinned out.

A 2.5 way crossover would make it so that you only have one woofer playing from let's say 50Hz to 400Hz...so at 50Hz, the other driver starts to kick in and help fill out the low end. The idea here is to pick the low pass for the secondary driver such that you can control the polar response of the woofer system, while also trying to maintain the same 6dB increase in efficiency. I was talking to some of the klipsch guys about crossover design at the pilgrimage and they were talking about goals they had for the off-axis response of the speaker because such a large portion of the sound comes from the room. Basically you want the off-axis response to be flat too and you want it to be smooth too (ie avoiding sharp transitions between drivers)

For example, you wouldn't want a speaker that is perfectly 90x45 at every frequency except a small region where it's 90x270...if you make the on-axis response flat you're going to be hearing all those frequencies that disperse at 270 as if they were way louder because you will be hearing the off-axis reflections. Likewise the opposite is true, if you get a lot of narrowing over certain frequencies then it's going to sound like the speaker is lacking in those areas. Ironically the 200-400Hz issue with this design happens to be a region where you can drop the signal 3-6dB and hardly notice it (or sometimes it will even sound cleaner/less muddy).

Anyways, I didn't mean to get too far into these concepts because in all actuality we're not talking that big of a difference. Probably on the order of 3dB from the reflections which will be partially masked by the on-axis signal. I'm not exactly sure what goes into a 2.5 way crossover, but I was thinking that it would come closer to the original sound of the cornwall (aka, trying to maintain the same off-axis response as the original).

As Hurd already mentioned, we were talking about the exact same speaker concept over msn (though I was redoing the ports to make use of the entire height of the cabinet, but it was the same tuning). Also, for easier speaker placement I was thinking it would be good to make the back corner of the speaker a bit narrower so that you can still tuck it all the way back into the corner and still be able to change the angle a bit...perhaps going so far as to make the speaker a triangle shape instead of the house shape (less parallel boundaries in the cabinet).

Sorry for getting a bit off topic with the whole dispersion thing. I honestly don't know how big of an influence it would have so I just kind threw it out there 2.gif It must be at least somewhat important if klipsch is spending the time to measure and tweak the polar response.

Link to comment
Share on other sites

Seems we've been writing at the same time 2.gif

----------------

Doc, I am a simple old construction worker. I do appreciate your efforts and with all due respect; I still can't see it.. other than applied in "free space". If I had a chamber,or if my listening enviornonment were outside...... I consider everything to be "on axis" bcause I can adjust it. Toe in or out. I can adjust ME to the sweet spot. Please considder the enviornonment of real world listening. Even in a large room, four CWs, six LaS, two cornerhorns and four LaS, six Cornerhorns, ...... Please describe what is "on or off axis" with SPLs of over 120db.

----------------

Right, if you were outside and you could adjust yourself to be on-axis then yes I'm just blowing steam. However, when inside the room the off-axis response of the speaker is bouncing off the floor, cieling, and walls which contributes to the overall sound of the speaker. Because the intent was to maintain the original voicing of the cornwall, I figured I'd mention that we were changing the off-axis response of the speaker. Whether or not this is for better or for worse will be dependant on the room the speaker is in. Essentially this design reduces reflections that occur in the 200-400Hz octave...so the speaker will sound shy in these frequencies when compared to the cornwall.

The ONLY reason I brought it up is because I found it interesting when the klipsch guys were talking about it...I have no idea how large the impact might be, but I'm guessing rather small (so perhaps I am just blabbing for no reason).

Link to comment
Share on other sites

Doc,

You science into this goes beyond something I didn't consider.

Yes the drivers are 18 inches apart (nice guess).

The original plan actually was to make a triangle backend of a cabinet. Infact, I have those plans here with the correct volume and porting. Have you ever tried to calculate the volume of a triangular prism. It is very hard but was done.

I retired the triangle backend because it gives less capabilities to "toe" and that extra volume to make the "point" in the back was almost negligable. This bass bin will sit flush in the corner if you want. There was more to designing this than you may think. The part of the cabinet that starts into the trapezoid is exactly a 45 degree angle. This allows flush placement with the corners. It was hard to get the volume of a trapeziod prism using a 45 and 135 degree angles. I had to relearn geometry to figure this out.

HOWEVER, you probably won't end up makin it flush with the corner. As with the Cornwall, it sounds better off the wall. The back trapeziod shape however lends itself to lose some of the volume into the corner. 3.622 cubic foot to be exact that is a hidden in the corner.

jc

Link to comment
Share on other sites

----------------

On 7/11/2005 9:24:11 PM colterphoto1 wrote:

K33 are 8 ohm woofers, two in parallel is 4 ohms. Where is the problem?

----------------

Michael, that isn't correct. K77 8 ohm, K55 16 ohm, K33 4 ohm

Two K33's in parallel is 2 ohms

jc2.gif

Link to comment
Share on other sites

When I said "nevermind" earlier, the distance between the woofers and the resultant comb filtering that would take place was the thing I was going to mention, but JC worked so hard, and they looked so nice -- I just didn't want to burst his bubble. I thought about the tapered array thing Saturday night, but I wasn't sure where to roll off the first driver -- 50Hz seems kind of low -- I don't see how he can still get 6db of gain when the second driver isn't doing all that much. At any rate, those are some excellent posts Dr. Who, thanks.

JC, you really need to get an RTA in there so you can see what's going on with that thing. Also, when you finish that top, take advantage of the fact that you can now align the voicecoils.

Link to comment
Share on other sites

Yes, that's the other thing. I would wire them in series and go with a single inductor. The value will go down to 1.25mH.

It's actually 13 ohms for the Atlas driver JC, however, if you are on tap 4, the reflected impedance back to the amplifier is 26 ohms.

Link to comment
Share on other sites

You're right Dean. I messed up the K55 ohm deally.

I didn't want to run each woofer "differently" because I don't want to take away fom my dB gain. Maybe it won't but it seems to me that it would. So, again, my 1st goal with this was to keep the Cornwall sound and get the dB gain.

The series hook up is a very good idea with the one inductor value cut in half. I bet I would get the same "sound" but the dB would decrease. I started with the two inductors as above because I have those laying around.

Dean, what would be the advantage of a series hook up and what do you mean by align the voice coils. Please let me know because I might need to change my drawings of the HF section. Thanks man.

Your input and Dr Who's is very helpful. I somewhat understand what you are saying. I didn't consider those points when building. You don't have to do the "nevermind" thing with me. I want you opinion even if it deflates my balloon. I think the project so far is a big success but there is always room for improvement. That is the reason for the early post with an incomplete project. I wanted opinions. I didn't do this approach with the Cornscalas. I finished them then posted on the forum. Infact. I sanded and painted them before I ever gave em a test run.

I was listening to it (mono) a good bit last night. I don't think I'll change the bass bin design. It just sounds too good. I will build the other one starting later in the month.

Dean, you may need to be a little more specific with me regarding what I need to do to "test" this setup. I read a bunch for this project but didn't get much into testing. I probably need to order some equip from mouser or Parts Express.

jc

Link to comment
Share on other sites

The distance between the two woofers at the frequencies they operate in will not cause appreciable comb filtering either.

You will have better horizintal imaging with the woofers and horns in a vertical alignment. Vertical imaging will be a bit narrower.

There are some formulas on how far apart the 15 inch drivers can be based on the wavelength of the highest freq. they will reproduce, but I can't locate them at the moment. It was some information I cam across on the JBL site.

Link to comment
Share on other sites

----------------

On 7/12/2005 10:00:03 AM Marvel wrote:

There are some formulas on how far apart the 15 inch drivers can be based on the wavelength of the highest freq. they will reproduce, but I can't locate them at the moment. It was some information I cam across on the JBL site.

----------------

The biggest acceptable distance between two drivers (from center point to center point) is 1/4 the wavelength of the highest frequency to be reproduced. (18" is roughly 1/4 the wavelength of 200Hz)

Link to comment
Share on other sites

I have some interesting information regarding this comb filtering debate.

Its 1115pm Georgia time. I just got back from Atlanta. Here is why.

About 50 people were invited to a lecture/event at one of the McIntosh dealers in Atlanta. Come to find out, McIntosh had one of their design engineers there talking about the technology in their equip.

Towards the end of his lecture he went on to say that MAC is no longer building the conventional speakers mainly because of the power handling requirement of their amplifiers. The biggest problem was their tweeters. Multiple tweeters are needed in their speakers. There ia big problem with dispersion, cancelling out, interference with multiple tweeters. They now have a patented design with 5 tweeters that doesn't do this.

ANYWAY

I B-lined this guy at the end of the lecture as well as everyone else there. The questions to him were the typical HIFI questions. Kinda boring. His eyes got big and he got excited when I floated this question.

"If you had the problem with multiple tweeters than how come you don't have that problem with dual woofers". He said "you do"!

I then went on and explained my design of the two 15" woofers with my network details. He then said "you don't have a problem". He went on to explain that the comb filtering isn't audible at low frequencies. The characteristic "muddy midrange" you get with this effect will occur more around 1000Hz. My current setup is crossed at 200hZ. He then went on to give me examples of how this would occur at other frequencies.

I think this seems to be an issue with dual woofers crossed higher in which there is mid bass coming out of them. It's really a problem with 2 way designs with two woofers. Maybe that is why the 2.5 way crossover got its name instead of 3.5 way crossover. This is just speculation here.

He didn't feel that I should use a different crossover frequency for each woofer. He rattled off a bunch of stuff designed with the same XO frequncy that I had never heard of.

He could have just been nice to me. Not sure.

I did check out the JBL site which says:

http://www.jblpro.com/pub/technote/fat_wht.pdf#search='comb%20filtering%20jbl'

By the way: This forum has turned me into a sick man. While at work I'm thinkin about this stuff all the time. Heck, I dream about ridiculous audio concepts. To make things worse, I auditioned some of my Diane Krall at this McIntosh dealer. There were several setups in there similar to mine. This is no joke. Even a pair of McIntosh speakers of $7000 or the B & W's of the same expense just don't carry the sound of what is in my listening room. I watched guys ew and ah tonight and I was floored. I wasn't impressed with any speaker there.

So what does this do.....I just don't see how I could ever buy another pair of HiFi speakers.

Good Night14.gif

Link to comment
Share on other sites

"The series hook up is a very good idea with the one inductor value cut in half."

A) One 4 ohm woofer and a 2.5mH inductor 600hz (Cornwall).

B)Two 4 ohm woofers and a 1.25mH inductor 600hz (2 ohm parallel connection).

C)Two 4 ohm woofers and a 5mH inductor 600hz (8 ohm series connection).

2.83V sensitivity around 100hz

A)101dB

B)107dB

C)101dB

Link to comment
Share on other sites

I think I need to quit posting until I catch up on sleep.

Dennis, he's using the K-401 and wants to use 400Hz. Actually, wired in series at 8 ohms, he could just use the single 2.5mH (500Hz). IOW's, do nothing.

Link to comment
Share on other sites

----------------

On 7/13/2005 1:05:22 AM djk wrote:

"The series hook up is a very good idea with the one inductor value cut in half."

A) One 4 ohm woofer and a 2.5mH inductor 600hz (Cornwall).

B)Two 4 ohm woofers and a 1.25mH inductor 600hz (2 ohm parallel connection).

C)Two 4 ohm woofers and a 5mH inductor 600hz (8 ohm series connection).

2.83V sensitivity around 100hz

A)101dB

B)107dB

C)101dB

----------------

Isn't Two 4 ohm woofers and two 2.5mH inductors (one to each woofer) the same as your example B? In otherwords, each woofer has its own inductor and wired individually into the network (no parallel or series).

Someone give me advice on some simple test equip I can purchase to "test" this thing.

Link to comment
Share on other sites

Isobarik? My understanding is that you want to increase the LF output to match the mid and highs. 2 woofers would do that in a regular mount as long as they are in parallel so that each outputs the same as a single woofer, thus doubling (adding 3 db) to the SPL. Of course your amp has to be able to handle the lower impedance. Isobarik mount uses one woofer to make the other woofer feel that it is in a much bigger cabinet thus reducing resonant frequency vs a sealed cabinet. The drawback is that you loose 3 db of output! This is the reverse of what you want to achieve. Isobarik are really only used to reduce cabinet size. I do not think that this is your issue.

Link to comment
Share on other sites

----------------

On 7/13/2005 7:50:10 AM jwcullison wrote:

Someone give me advice on some simple test equip I can purchase to "test" this thing.

----------------

Do you have a laptop or a computer nearby? You can get a software RTA relatively cheap:

http://www.trueaudio.com/rta_abt1.htm

And then I believe Behringer sells a testing mic for like $60 or something like that which is flat from 10Hz-30kHz or something close. I think formica has one actually, perhaps he'll chime in. Ok, here's a link:

http://www.behringer.com/ECM8000/index.cfm?lang=ENG

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...