• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Dipole vs Box speakers

jim1274

Active Member
Forum Donor
Joined
Dec 28, 2023
Messages
280
Likes
184
Ok—here is my first in room response curve sweep for the Enterprise Omnis. I used the mic in my iPhone for the test. I took the measurement from the centered listening position:

Distances:

Front wall: 6 feet
Side wall: 4.5 feet
Speaker separation: 6.5 feet
Speaker To listener: 6.5 feet
Height to mic: 43”

Drum roll please….

View attachment 343224

This seems pretty consistent with my “ears” test,
specifically the high end bump, noticing a little brightness in the upper treble, most noticeable on songs that have a lot of high volume level content in that frequency range. On most program material, it was not very noticeable, not impacting the overall tonal balance all that much.

So, is this response curve more or less ragged than you guys were expecting from a Omni? It should be interesting to see what the response looks like for the other speakers in the mix.

Anyone want a phase or group delay plot next?

Here is the AMT dipole response in same location for comparison:

1705786360773.png


This is consistent with my “ears” test—AMT seemed bright with exaggerated high end.
 

jim1274

Active Member
Forum Donor
Joined
Dec 28, 2023
Messages
280
Likes
184
did more response sweeps from in the kitchen standing and living room very close to the right speaker seated. This is a very rough (and lousy) diagram of the measurement positions:

1705726742489.jpeg



Note that my distance from front wall of 6’ in prior test was actually from the plane of a whole wall of tall gear and large TV—actual front wall was about 2’ behind that.

The off axis living room position was seated on a wide ledge surrounding my large fireplace jut-out, about 3.5’ to the right speaker. Response from there:

1705726780058.png

Here is the AMT response for the off axis seated position compared to Omni graph above:

1705786614783.png


Pretty smooth!

The kitchen test position was where I barely had a line of sight to the right speaker 11’ away, at standing height since one would be in that position there. Response:

1705726889918.png

Finally, the sweep from kitchen position compared to Omni graph above:

1705786750073.png
 

jim1274

Active Member
Forum Donor
Joined
Dec 28, 2023
Messages
280
Likes
184
Here is the AMT dipole response in same location for comparison:

View attachment 343376

This is consistent with my “ears” test—AMT seemed bright with exaggerated high end.
I moved the AMT dipoles a little closer to front and side walls for this sweep:

Front wall: 1M
Side walls: 48”

(Yeah, yeah..I mixed metric and English..)

1705787495625.png


Note how much this cleaned up the upper end peak—didn’t expect that big of a difference. That may be a game changer in my main criticism of the AMT being way too bright. I’ll run some test songs in this smoother top end response position for a quick “ears” test.

Well, I now see that both the Omnis and the dipoles are going to need a whole grid of locations testing to find the smoothest response curve. Looks like it’s time to draw some scale drawings of the room since we now will have multiple position tests of at least 4 different speakers.
 

jim1274

Active Member
Forum Donor
Joined
Dec 28, 2023
Messages
280
Likes
184
I’ll get back on topic “dipole vs box speakers” soon. I plan to use my BMR monitors for my “box” speaker. Next phase is to instantaneous A/B the Omnis vs box, dipole, and bipole, mainly for soundstage comparison. I’ll be sure to do a dipole vs box and report in this thread.

I’ll restrict further posts here to just “dipole vs box” going forward—I’m duplicating too much that is on the Omni thread. If anyone is interested on the Omni/dipole/bipole/unipole “shootout”, they can follow the onmi thread:

 

jim1274

Active Member
Forum Donor
Joined
Dec 28, 2023
Messages
280
Likes
184
I can’t wait much longer to try a “box” speaker for in room response—maybe yet tonight. I’m enjoying the AMT so much now, maybe I’ll wait a bit. The soundstage part of the equation that started me down this path is really pretty impressive on these, still subject to direct A/B with the “box” (and Omni, the horse to beat). We shall have a box comparison soon. Fleetwood Mac “Gold Dust Woman” just played on the AMTs and soundstage was very wide and deep, seeming a little better on the spatial accuracy of instrument placement within the soundstage than some other demo test songs.

By chance I checked the frequency response stated accuracy of the AMT. it’s +/- 3db 30-23k.


If not 10 degrees outside and deck covered with snow, I’d run a frequency response outside well away from the house wall.
 

jim1274

Active Member
Forum Donor
Joined
Dec 28, 2023
Messages
280
Likes
184
I moved the AMT dipoles a little closer to front and side walls for this sweep:

Front wall: 1M
Side walls: 48”

(Yeah, yeah..I mixed metric and English..)

View attachment 343382

Note how much this cleaned up the upper end peak—didn’t expect that big of a difference. That may be a game changer in my main criticism of the AMT being way too bright. I’ll run some test songs in this smoother top end response position for a quick “ears” test.

Well, I now see that both the Omnis and the dipoles are going to need a whole grid of locations testing to find the smoothest response curve. Looks like it’s time to draw some scale drawings of the room since we now will have multiple position tests of at least 4 different speakers.

For “box” speaker comparison I used my Energy 22 Reference as they were close by (will try BMR monitors too later). These have recently rebuilt tweeters and matched pair to original OEM spec. These are accurate speakers, I believe used as studio monitors by the CBC in that era. Since Floyd Toole is quoted so often here, thought I’d add this quote in:

“Unbeknownst to us, at the same time that we were finalizing our Energy 22 design, Floyd was conducting an international test of studio monitors for the CBC (Canadian Broadcast Company, I think.) In the elaborate process of double blind speaker comparisons of the best speakers in the world, Floyd sneaked in our Energy 22 prototype. It was our speaker that won the contest. We promised Floyd that we would make no changes to the sound quality in subsequent production.

Subsequently the energy 22 received many rave views and great commercial success in Canada.”

Source:

For the first Energy 22 response test I used the same 1M to front wall and 48” to side wall as in the above AMT dipole positioning:

1705869330971.png


Then I moved them much closer for a near field measurement, about 7.5’ from front wall and 48” from side walls, close to an equilateral triangle to the listening position:

1705869523952.png


Soundstage: the near field position didn’t have much of a soundstage (especially depth) but the 1M/48” positioning was another story, actually surprising me, much deeper and wider than expected. It did not seem as 3-D or immersive as the dipole AMT (or even less so than the Omni), but seemed to have better spatial accuracy and positioning of instruments. I particularly noticed it on Bob Marley “Jamming”. There is a lower level percussion that sounded like it was 5’ or so in front of the speakers and panned all the way to the wall.

When I did the much earlier box vs Omni soundstage evaluation, the box speakers were a long distance from the front wall in a more near field position, and the Omnis walked all over them on soundstage. (I tried both the Energy 22 and BMR for those tests). That was a mistake to draw any conclusions without trying more box speaker placements for a better soundstage. This soundstage showdown just became a tighter race.

I now need to set-up direct A/B comparison of the dipole and box soundstage.
 

jim1274

Active Member
Forum Donor
Joined
Dec 28, 2023
Messages
280
Likes
184
I came across this and my AMT dipole positioning was pretty close to their recommended “golden ratio” method calculator.

I noticed that the “golden cuboid” room is literally almost identical to my room except for the 10’ ceiling part. My ceiling is gently sloped from 8.5’ to 9.5’, so average of 9’ vs “golden cuboid” 10’.


1705898497650.png

“Golden” sounds like a good thing…I can see eyeballs rolling…again…
 

StigErik

Active Member
Forum Donor
Joined
Feb 22, 2023
Messages
274
Likes
389
Location
Norway
Yes…. Eyeballs are rolling

Seriously, room dimensions like that will at least give a good spread of room modes. But that speaker placement recommendation is debatable at best IMHO.
 

jim1274

Active Member
Forum Donor
Joined
Dec 28, 2023
Messages
280
Likes
184
Yes…. Eyeballs are rolling

Seriously, room dimensions like that will at least give a good spread of room modes. But that speaker placement recommendation is debatable at best IMHO.

I knew that would get someone’s attention…

I dug around a bit and that is a scientifically based mathematical calculation but came up with almost nothing on a forum search for “golden cuboid”. I hope I didn’t really stir up a hornet’s nest and get beat to a pulp now…

It’s also based on a “golden” listening position, closer than I’ve been sitting on most of my placement experiments. Hey—I like my couch even if back too far from optimal. Better get my folding chair out and get “golden” for a test listen…
 

princeTemi

New Member
Joined
Aug 7, 2023
Messages
1
Likes
1
Location
Poland
And bass is only one problem, it is quite difficult to have a dipole in HF.
Hi. it's not a problem, just use proper dipole tweeter. Forget standard domes, go for Mundorf AMT, Radian LT, Beyma tpl or BG Neo's. It seems that You can shave/cut rear cup of Compression Driver to create dipole - check out La Dolce Audio.
 

StigErik

Active Member
Forum Donor
Joined
Feb 22, 2023
Messages
274
Likes
389
Location
Norway
It's just a cube where the dimensions are calculated based on the so-called Golden Ratio of ~1.618

a) 1
b) 1.618
c) 1.618^2 = 2.617

Nothing fancy at all really.
 

StigErik

Active Member
Forum Donor
Joined
Feb 22, 2023
Messages
274
Likes
389
Location
Norway
Below are a few measurements of my own system, which is dipole from 80 Hz and up, as well as being line-sources.

The main advantage of a well-designed dipole is that a more uniform radiation pattern can be achieved in the horizonal plane, so that the reflected energy from the room has a frequency response that is more similar to the on-axis response of the speaker than would be possible with a typical closed box speaker. A line-source will also (in theory) create a plane wave in the vertical plane, effectively eliminating floor and ceiling reflections. My subwoofer system (below 80 Hz) is a Double Bass Array, that also creates a plane wave in both the vertical and horizontal plane. Combined with the active cancellation of the rear array, it will very effectively cancel out all room modes. at least in theory....

An other advantage of the open baffle dipole is the lack of internal and structural box resonances, or internal early reflections radiating back through the speaker drivers.

And of course there are disadvantages, the main problem being bass extension and SPL. In an open baffle speaker, the requirement for cone excursion increases 8 times for each octave below the dipole peak - where the front and back waves sum in phase, typically several hundred Hz depending on the baffle size.

An example : Let's say we need 2 mm cone excursion at 100 Hz to reach 90 dB. If we want to have the same output one octave down, we must multiply by 8 and get 16 mm. At 25 Hz... 8x16 = 128 mm. As you see, dipoles do struggle at low frequencies! Linkwitz created a nice spreadsheet for calculating max SPL from dipoles and monopoles. It's still online : https://www.linkwitzlab.com/spl_max1.xls

My own system is based on Magnepan 3.7, a speaker that in my opinion did not sound good at all. The frequency response and off-axis response is very uneven, there is virtually no bass, and massive amounts of distortion at virtually any SPL level below 500 Hz. However - convert this thing to an active speaker with DSP, choose crossover points that makes more sense, add an open baffle low-mid system to make it a four-way, and add a proper subwoofer system.

Measurements below.. all done in the main listening position at 3 meter distance. The sloping frequency response is intentional - aka "House Curve".



StigErik-FR.png
StigErik-WF.png


StigErik-DIST.png
 

architect

New Member
Joined
Apr 6, 2023
Messages
3
Likes
0
veru nice StigErik. can you also post a picture of your listening system / listening room?
 

architect

New Member
Joined
Apr 6, 2023
Messages
3
Likes
0
very nice measurement : )
an inspiring system you have
but your solutions are a big bit to much for my domestic living room ; )
 
Top Bottom