Ben,
My intentions are simply to help you achieve a smoother system response at your listening location.
If the 6 dB drop in the 110 Hz crossover region when the mids are added does not indicate they are reverse polarity from the lows, please explain what in your many decades of exploration it does indicate?
Art
Art -
I thought we were going to pick at your bumps and dips today?
BTW, I guess you didn't know that sound travels at about 1000 f/s. Think about it. I am sure if you do, you'll be able to answer your own question.😀
Kidding, of course, but how does it feel to you to be treated in a patronizing manner by someone posing (as a joke) as a supercilious intelligence?*
Yes, good advice is welcome in its time and place. But, may I suggest, it is rather insensitive of you to suggest that after many decades of effort, I (or anybody posting here) is ignorant about phase!!!
Can we please get back to the thread.
Ben
*But I am not kidding that I just gave you an obvious reason for the least-of-evils dip that you seem to have "overlooked." Perhaps unexpected for a guy who works as an alignment contractor.
Last edited:
Ben,
If you want to pick at any of the frequency responses I have posted, go ahead, ask questions or make observations.
I did not suggest you are ignorant about phase, but you do seem perfectly willing to ignore the aural consequences of using loudspeakers that can not be phase aligned 😉.
I am well aware that sound travels at 1130 FPS.
It looks to me that the 6 dB drop in the 110 Hz crossover region when the mids are added indicates a reverse polarity from the lows.
However, the phase difference between the lows and mids may be such that either polarity will result in a dip in the crossover region, which could be easily measured.
If you have measured, and found neither polarity combination sums positively in the crossover region, short of using means to correct the phase difference, you are stuck with it.
We do agree the difficulties in creating coherent summation from multiple sources each having different radiation patterns in a small room is not trivial.
One observation: an open baffle placed closely against a wall is no longer an open baffle, as the baffle/wall is an enclosed back volume.
Your large OB is functionally a BLH, a back loaded horn.
Art
I thought we already beat that misunderstanding to the ground? Only owners of hotels surrounding Loch Ness have ever seen the monster.
Thanks to my footnote hint in the last post, I think you've got it. Great!
That's about as profound (and helpful) as me saying your front loaded direct drivers are really playing into horn, also known as your room. The gap between my highly irregularly shaped giant baffle board and the wall bears little comparison to a horn.
Please stop.
Ben
Last edited:
Hi Art
I thought that we were restricting our discussions here to the modal region in rooms. Radiation patterns have no meaning in that regime.
Ben
Art is one of the more knowledgeable, respectful and polite people at DIY. If you are having trouble with him then I suggest that you look elsewhere for the problem.
Art is one of the more knowledgeable, respectful and polite people at DIY. If you are having trouble with him then I suggest that you look elsewhere for the problem.
Earl,
Are you saying that there is no difference in room modal excitation whether the radiator is a dipole or a monopole?
^
A dipole is two separated monopoles with opposite polarity. So a dipole will interact differently than a monopole at the same room location.
I think your idea of "creating coherent summation from multiple sources" is based on false conclusions. You're probably thinking about how source summation works in a free field but we're talking about source summation in acoustically small spaces. The direct sound is swamped by the room response. So the radiation pattern isn't exactly meaningless but just one parameter in getting smooth response in frequency and time.
A dipole is two separated monopoles with opposite polarity. So a dipole will interact differently than a monopole at the same room location.
I think your idea of "creating coherent summation from multiple sources" is based on false conclusions. You're probably thinking about how source summation works in a free field but we're talking about source summation in acoustically small spaces. The direct sound is swamped by the room response. So the radiation pattern isn't exactly meaningless but just one parameter in getting smooth response in frequency and time.
Of course there is a difference, but you cannot view it as a directivity response problem. In a room that has discrete modes the wave is restricted to move in only one direction per mode (well actually two orthogonal directions). The idea of directivity requires the response of a source along a variable direction, but the sound cannot travel away from the source in an arbitrary direction if there are discreet modes. As Markus says it is best to view LF sources of all types as sums of discrete sources separated in space and phase. This is always safe, but looking at it from a directivity standpoint is dangerous.
It is this kind of thinking that leads to false concepts like "Dipoles excite fewer modes." which is false. You can see this if you view a dipole as two sources out of phase and separated by a distance. The dipole will actually tend to excite more modes, which of course is desirable, not the other way around.
Agreed.
As Markus also said:
"The direct sound is swamped by the room response. So the radiation pattern isn't exactly meaningless but just one parameter in getting smooth response in frequency and time."
After further review of the tests posted in #1730, appears that time alignment has less to do with smooth response at my listening position, and "luck" at choosing the right place to put my couch in relation to the room modes more .
One interesting finding was that opening the rear room doors increased LF +8 at 33 Hz at the listening position.
My traces in #1713 illustrate nicely the complex interaction of enduring modes arising from room shape and the influence of seat location. The room modes are clearly enduring, but the two traces are distinctly different as well. (If you wanted to name the major artifacts, you'd name the room modes.)
Without even whispering a word about human hearing that might give a major twist to this discussion, it is easy to fall in fallacious thinking by confusing the simplifying assumptions and inspirational diagrams used to propel analysis with the sound of music introduced into a room. We all simplify to analyze, I'm just saying (pending better and better 3-D waterfall charts) there's a big step between that and the final audition.
Although thinking about the complex sound of dipoles might give the more analystic among unipole fans the heaves, it sure seems to fill-the-bill for dipole fans. Ditto for the topic of this thread: multiple subs.
Ben
Last edited:
Does multi sub approach also work in smaller rooms?
Hi Earl,
Will the same multi sub approach also work in smaller rooms, something like 15x13x10?
The reason why I ask is that in OptimalPlaybackInSmallRooms.ppt you have defined a small room as 20x14x10. Some rooms may be smaller like above.
Thanks,
Goldy
Hi Earl,
Will the same multi sub approach also work in smaller rooms, something like 15x13x10?
The reason why I ask is that in OptimalPlaybackInSmallRooms.ppt you have defined a small room as 20x14x10. Some rooms may be smaller like above.
Thanks,
Goldy
The smaller the room, the more pressing are bass issues. For example, room modes are in more audibly detectable ranges. So having another bit of control (by working with multiple subs) is better. Of course, the smaller the room, the harder it is to decorate your room with many subs.
Aim for heterogeneity of box types too.
B.
Multiple subs always works regardless of size. The specifics of each room are so widely varied that you really need to simulate the situation using something like the software on my site.
I'm sorry to have neglected any comment from Earl, even if just "huh."
My favourite example of relying on your eyeballs instead of your ears-on-music is comb filtering. Few diagrams are as dispiritingly horrible as comb filtering. But there's reason to doubt comb filtering spoils music much or at all.
Ben
I have built a number of 4 element column speakers, improving each design over time.
My latest has 4 Dayton PA-130's and a small horn.
I cross them to woofers and then subs at 153 Hz BW4.
I think they sound great.
They are so efficient I can not tolerate full volume with a 40 watt amp.
.
My Daughter comes over and says "How can you stand all that swish, swish"😱
I guess everyone has their own expectations and hears what they want to.
Subjectivity seems to swamp reality.
P.S. I fully subscribe to the Multiple Sub approach.
Dave
Last edited:
Practical inputs please - sub x 2 purchase soon
Making my way through this mammoth thread, am up to page 80 and have paused just now to make this post, which is longer than intended - so apologies in advance!
I have some time off next week, where i'd like to purchase a couple of additional subs and put Earls method into practice. So i'd like to ask for some practical inputs please - recognising that the answers might well be in the next 70 odd pages, but i might not get there in time...
My current single sub is an SVS SB12-NSD - Measurements (Here)
As you can see from the measurements, the sub though sealed has some quite high Group Delay.
1. From the first 80 pages or so I'm not clear if GD in the subs is removed as a potential issue with Earls approach?
2. If it is a problem, should I try to address it by siting this sub close to the listening position (as it is today)?
3. If GD is a problem then the other 2 purchases should presumable be low GD subs?
My interpretation of whats been said around the choice of sub so far, i.e. being fine at the cheaper end of the spectrum, probably implies that GD isn't significant - but i'd really like some input before buying.
Currently my intent is to buy a couple of 10 inch subs from the same manufacturer. I was going to go for ported, the ones where you get plugs to change the configuration between 2, 1 and 0 ports and run them configured differently. That way I get the required 3 differently behaving subs.
4. Or should I get a 12 and 10 inch ported instead, to maximise the differences? I'd prefer 10 x 2 simply because they're smaller!
I'd put the 12 inch SB12 in the corner next to the mains (hence the GD concern) and the other 2 as per Earls suggestion and Markus's page (Here)
Some hopefully useful context re my current setup.
Room is as follows;
I have a Umik 1 and have 'played' with REW before, but tbh I didn't have access to a hardware parametric equaliser and so 'made do' with the inbuilt parametric equaliser in VLC - less than ideal.
Likely if I still need EQ, that i'd go the miniDSP (possibly Dirac) route, though i'd like to try the subs on their own first.
Making my way through this mammoth thread, am up to page 80 and have paused just now to make this post, which is longer than intended - so apologies in advance!
I have some time off next week, where i'd like to purchase a couple of additional subs and put Earls method into practice. So i'd like to ask for some practical inputs please - recognising that the answers might well be in the next 70 odd pages, but i might not get there in time...
My current single sub is an SVS SB12-NSD - Measurements (Here)
As you can see from the measurements, the sub though sealed has some quite high Group Delay.
1. From the first 80 pages or so I'm not clear if GD in the subs is removed as a potential issue with Earls approach?
2. If it is a problem, should I try to address it by siting this sub close to the listening position (as it is today)?
3. If GD is a problem then the other 2 purchases should presumable be low GD subs?
My interpretation of whats been said around the choice of sub so far, i.e. being fine at the cheaper end of the spectrum, probably implies that GD isn't significant - but i'd really like some input before buying.
Currently my intent is to buy a couple of 10 inch subs from the same manufacturer. I was going to go for ported, the ones where you get plugs to change the configuration between 2, 1 and 0 ports and run them configured differently. That way I get the required 3 differently behaving subs.
4. Or should I get a 12 and 10 inch ported instead, to maximise the differences? I'd prefer 10 x 2 simply because they're smaller!
I'd put the 12 inch SB12 in the corner next to the mains (hence the GD concern) and the other 2 as per Earls suggestion and Markus's page (Here)
Some hopefully useful context re my current setup.
Room is as follows;
- 2.65m x 3.6m x 4.5m (4.8m in 2 alcoves)
- 3 solid walls (brick), 1 partition wall, wooden floor with room underneath and plasterboard ceiling into loft space above.
- Heavily furnished and messy enough that I've never been tempted to share any pictures!
I have a Umik 1 and have 'played' with REW before, but tbh I didn't have access to a hardware parametric equaliser and so 'made do' with the inbuilt parametric equaliser in VLC - less than ideal.
Likely if I still need EQ, that i'd go the miniDSP (possibly Dirac) route, though i'd like to try the subs on their own first.