Since it appears that no amount of theory is going to satisfy you, you have two options. Rely on people's opinions or take some responsibility and do some testing.
Take your stereo outside. Place the subs and mains where you are planning to put this system - subs 60 feet apart and mains 30 feet apart. Do measurements EVERYWHERE that an audience member could conceivably be standing. Full frequency sweeps in each location.
Then leave the mains where they are and cluster both subs together in the middle. Repeat the same measurements in the same spots.
This should take all of about 15 minutes (you only have to measure on one side, the other side will be mirror imaged). You can run 50 - 100 measurement sweeps in that time. (Of course setting this all up will take longer but the measurements are quick).
You can even test a variety of mains locations with both sub configurations as well if you like, but that part can come later after you decide on a central sub or stereo separated subs.
A fully calibrated measurement mic is about $100. The measurement software is free. You will need a very long cable for the mic.
Like I keep telling you, you can answer most of your questions yourself by taking your current system outside and evaluating your choices. Once you measure the effects of two subs placed 60 feet apart I'm pretty sure you won't want anything to do with that configuration. It will be ok on axis (perpendicular to the centerpoint of the horns, aka the hot tub location) but even as little as a couple degrees off axis your frequency response is going to start becoming very problematic.
Just do it already.
Hi 'just a guy',
Transients that follow up do not overlap or sum and is something else than lubing.
Regards,
Djim
Hi Entropy,
Like 'just a guy' already mentioned perhaps hiring a small PA and do some testing at the actual location can give the info you are looking for.
If you ask me, subs for left and right. Keep the distance between left and right small enough to create a power lane instead of a tiny 'punch' sweet spot... in the garden your 'on-axis' neighbor ;-)
Regards,
Djim
Last edited:
Already did this like 2 weeks ago, but sure I'll do it again and give a few more snapshots.
Speakers are 18 meters apart (9 and -9 meters in relation to virtual mic axis) and mic location is 24 meters back.
First pic - 0 degrees, directly on axis
Second - 2.83 degrees off axis
Third - 11.31 degrees off axis
Fourth - 45 degrees off axis
An externally hosted image should be here but it was not working when we last tested it.
The really narrow notches at higher frequencies don't matter, that's what we hear with stereo speakers all the time and it's not a problem. It's the big wide notches at lower frequencies that matter - these are dead spots with no bass at that location at that frequency.
Like I said, even a couple degrees off axis things start to look bad when the speakers are that far apart.
And I will repeat - you NEED to do some testing and measurements on your own to determine how much this issue is going to affect your audience positions. And as the sims show, you have to do a LOT of measurements in a LOT of areas because the notches move around in frequency and location.
I have no idea what you mean by "transients that follow up do not overlap or sum". And you said yourself the issue is center to center distances and arrival times - the only difference between perfect summation and a huge comb filter notch is distances and arrival times so the issues are caused by the same thing whether you regard them as the same issue or not.
He has a small PA inside his house.
When you are planning horns with a 17 foot diameter mouth it's impossible to put them close enough together to avoid issues even with a crossover as low as 80 hz and a relatively small audience position.
Just a guy, we already talked about this - i.e. me taking my wife's stereo outside this weekend. It's only Thursday. And it's currently dark, raining, and my two kids are a handful.
(Djim, below is the system I'm taking outside)
bentoronto, I have had regular audiograms starting in 1995. They do not measure the very low frequencies, or the very high frequencies. I have measured my ear's high frequency cutoff directly (signal generator & tweeter) The first time I did it, I could hear about 19.5 kHz (age 17). And about two years ago (the last time I checked), it was about 17 kHz. It is a bit unnerving - hearing the sound simply cutout. . . My hearing loss over the past 20 years has been gradual, and not necessarily abnormal - I'm simply getting old. My eyes don't work as well either (I've got some floaters, and I need glasses). . . . But you know what? no one gets out of this Earth alive (except those jackasses going to Mars - they'll be dying on mars. . . )
Here's an old-school picture of me on the boat. I don't have many of these - I wish I'd taken more. . ..
(Djim, below is the system I'm taking outside)
bentoronto, I have had regular audiograms starting in 1995. They do not measure the very low frequencies, or the very high frequencies. I have measured my ear's high frequency cutoff directly (signal generator & tweeter) The first time I did it, I could hear about 19.5 kHz (age 17). And about two years ago (the last time I checked), it was about 17 kHz. It is a bit unnerving - hearing the sound simply cutout. . . My hearing loss over the past 20 years has been gradual, and not necessarily abnormal - I'm simply getting old. My eyes don't work as well either (I've got some floaters, and I need glasses). . . . But you know what? no one gets out of this Earth alive (except those jackasses going to Mars - they'll be dying on mars. . . )
Here's an old-school picture of me on the boat. I don't have many of these - I wish I'd taken more. . ..
Hi 'just a guy',
My last attempt for today...
How can a transient sum or cause a "huge comb filter" if you like, if the second transient (from another source) arrives behind the first one???
Regards,
Djim
Hi Entropy455,
Sorry I didn't read the whole thread, but that will do
Regards,
Djim
This is getting more complicated than I think it needs to be.
This is all a lot easier outside.
A single LF source where most recorded material is mono anyway mitigates the power alley / Lf destructive interference issue.
The interference patterns shown in simulation is valid when both L and R mains are playing the exact same thing.
Is you put correlated (single source ) pink noise through both left and right channels and walk around the sound field it will verify the interference simulation. If you put non correlated (two noise sources) through it this interference pattern all but dissapears.
So, unless you listen to monophonic recordings and move around a lot while listening critically, this is nearly academic only.
I am a measurement guy with a lot of professional audio measurment gear that I use often but I have a very busy life. Sorry, no post after post of screen shots from me.
Just build it!
Barry.
This is all a lot easier outside.
A single LF source where most recorded material is mono anyway mitigates the power alley / Lf destructive interference issue.
The interference patterns shown in simulation is valid when both L and R mains are playing the exact same thing.
Is you put correlated (single source ) pink noise through both left and right channels and walk around the sound field it will verify the interference simulation. If you put non correlated (two noise sources) through it this interference pattern all but dissapears.
So, unless you listen to monophonic recordings and move around a lot while listening critically, this is nearly academic only.
I am a measurement guy with a lot of professional audio measurment gear that I use often but I have a very busy life. Sorry, no post after post of screen shots from me.
Just build it!
Barry.
I apologize in advance if this is a dumb question - but what happens at a large outdoor venue? I've seen the Eagles, the Grateful Dead, Pink Floyd, and Steve Miller. All of these systems had a fantastic bottom end - clean mids, and crisp highs. I though it sounded great (although Steve Miller needs to spend more time playing music, and less time blabbing about himself between songs. . . .) But I digress
Shouldn't there have been fantastic bass lobing? Was there fantastic bass lobing, and I simply didn't notice it? I find it hard to believe that they had essentially a point-source sub - especially for the Pink Floyd concert - where 103 thousand people were in attendance. . .
- - -unless you listen to monophonic recordings and move around a lot while listening critically, this is nearly academic only - - - Just build it!
So. . . . . You're recommending a pair of stereo bass horns? I'm trying to determine what "it" is that I should build.
I'm a bit taken back that the largest unresolved design issue is whether to build a mono bass horn, or stereo bass horns. I did not anticipate this as being an issue.
Every single sim I showed except for the 0 degree on axis ones have the second sound source arriving behind the first one, this time delay is a phase difference and it causes a null if the time delay is large enough.
I can pretty much guarantee there were many issues with the fantastic sounding concerts. As far as recognizing lobing, if you are in a null area it's only a null at a certain frequency range so it's not a complete lack of bass. If you didn't measure the frequency response (which would require access to the mixing board) than you would never know what the frequency response in your location was. It could be very bad but still sound very good. And if you didn't walk around in the crowd you wouldn't experience the different frequency response in different locations.
I can guarantee you with 100 percent certainty that the frequency response varied wildly at different places in the concert venue. If that's what you want for your system then there's no problem at all placing the subs and speakers 60 feet apart. But if you want a controlled and even frequency response over a wide area you need to plan it properly.
Just a guy - I took a close look at your lobing graphs for my proposed "stereo" system (post 524 - thanks for the sim by the way).
On-axis, things look great. No point in saying more about on-axis.
At 2.83 degrees (about 4 feet to the left or right from centerline at 24 meters (tangent function), which I admit is not a lot) there’s destructive interference around 200 Hz. I will have this problem regardless if I build stereo bass horns, or a mono bass horn, as 200 Hz will likely come from my mid-bass "stereo" horns anyway. At 11.31 degrees (16 feet to the left or right from centerline at 24 meters – tangent function) I get nasty destructive interference around about 50 Hz, and again at about 150 Hz. I see your point. However I also see a crap-load of lobing between 200 Hz and 1 kHz also. Are you saying that lobing from 150 Hz and down is a deal breaker, whereas lobing above 150 Hz is ok?
To summarize your findings - if I build a pair of stereo bass horns and space them 60 feet apart, I’ll get zero destructive interference on-axis - and then at 4 foot off-axis, I’ll introduce a lobe at 200 Hz, which will then shift down to 50 Hz - as I continue moving off-axis to a distance of 16 feet. Is this really a deal breaker? The only thing a mono bass horn seems to buy me, is preventing lobing of the bass-horn frequencies only. If I cross my bass horn around 160 Hz, then I start to sacrifice stereo separation in the kick frequencies - where they should be very directional. If I cross my bass horn at 80 Hz, I'll maintain my stereo separation, but I'll still have lobing issues from the mains starting at 80 Hz. (I am trying to weigh the advantages and disadvantages of each.)
At 45 degrees, I consider the listener as outside of the listening area (non-applicable).
On-axis, things look great. No point in saying more about on-axis.
At 2.83 degrees (about 4 feet to the left or right from centerline at 24 meters (tangent function), which I admit is not a lot) there’s destructive interference around 200 Hz. I will have this problem regardless if I build stereo bass horns, or a mono bass horn, as 200 Hz will likely come from my mid-bass "stereo" horns anyway. At 11.31 degrees (16 feet to the left or right from centerline at 24 meters – tangent function) I get nasty destructive interference around about 50 Hz, and again at about 150 Hz. I see your point. However I also see a crap-load of lobing between 200 Hz and 1 kHz also. Are you saying that lobing from 150 Hz and down is a deal breaker, whereas lobing above 150 Hz is ok?
To summarize your findings - if I build a pair of stereo bass horns and space them 60 feet apart, I’ll get zero destructive interference on-axis - and then at 4 foot off-axis, I’ll introduce a lobe at 200 Hz, which will then shift down to 50 Hz - as I continue moving off-axis to a distance of 16 feet. Is this really a deal breaker? The only thing a mono bass horn seems to buy me, is preventing lobing of the bass-horn frequencies only. If I cross my bass horn around 160 Hz, then I start to sacrifice stereo separation in the kick frequencies - where they should be very directional. If I cross my bass horn at 80 Hz, I'll maintain my stereo separation, but I'll still have lobing issues from the mains starting at 80 Hz. (I am trying to weigh the advantages and disadvantages of each.)
At 45 degrees, I consider the listener as outside of the listening area (non-applicable).
Last edited:
On the piano, from about 31 to 62 Hz, there's twelve notes (i.e. a full octave of music). I agree that these destructive interference nodes look wide at low frequencies, but they are not killing a whole octave of music - maybe one or two notes at best. However if you look between 1000 Hz and 2000 Hz (on the 11.31 degree chart) - you've got ten nodes alone within a single octave. How can this be less impactful?
I'm not trying to argue with you, but rather to understand.
You will have this problem IF the mains are 60 feet apart. If you have them closer together, like maybe 15 feet apart this won't be so much of an issue, the lobes will be pushed up higher in frequency.
That is absolutely not what I'm saying. I'm saying if you only have one bass horn you eliminate the lobing issues at the low subwoofer frequencies. And then if you place the mains properly you eliminate a lot of the issues through the crossover frequency overlap zone.
Is it a deal breaker? I don't know, you tell me. The only thing dual horn subs gains you is twice as much cost, twice as much labor, twice as much space and lobing issues deep into the subwoofer frequencies. If that's what you want it's not a deal breaker.
Ok, look. I'm trying to explain sound source separation, frequencies, angles and time delay as a concept. The spreadsheet I'm using is definitely not the best tool for this, don't take the sims as gospel. Directivity isn't even considered in these sims and it's going to make a difference.
The point here is that dual subs will create triangulation timing issues off axis even into subwoofer frequencies if they are far enough apart. And mains can be a lot closer (like 15 feet apart) and push the triangulation timing issues way up in frequency. As far as the mains positioning, this doesn't have to be decided yet anyway, you can experiment.
What I'm trying to do is encourage you to measure and figure out for yourself if these issues are deal breakers or not.
At the same time, it's probably well past time that I learned to use proper tools to simulate total system coverage. So I'm looking into Danley Direct software. It should be simple enough to use and you don't have to input any actual speaker data - you just pick Danley products from a menu and the frequency response and hopefully directivity info is already preloaded. Then it's as simple as placing the speakers in the Sketchup-like GUI (vertical, horizontal position as well as toe in and rotate are available), defining the audience position, and checking the coverage map and/or frequency response at any defined virtual mic location.
With this tool it should be quick, simple and painless to sim coverage patterns of a variety of different configurations, speaker locations, number of speakers, etc.
The only problem is that DSL doesn't have any products even remotely similar to your proposed sub horn so directivity for the sub might be an issue. I could try stacking several large DSL subs to approximate your sub horn but I'm not sure if I'll be able to achieve directivity characteristics of a full size horn. We'll see.
Once I figure out how to use this software then I can look for something a bit more complex, which can accept actual speaker data, although if something like that exists it would probably be a nightmare to load all the data at all angles, unless it could accept something like a polar map which has all the data in one picture. Otherwise I'd have to make and load several .frd type files at varying angles.
By the way, if you are familiar with the basic features of Sketchup, Direct will be no problem at all for you to use, so I'd recommend that you get it and try it too. Then you can see visually what the power alley and lobing issues look like. It shows you coverage in a way similar to the polars I showed earlier - over your defined audience area it shows sound intensity as different colors so you can see where it's strong and where it's weak. And you can check at any frequency you like and see the coverage pattern. If I'm not working tomorrow I'll learn to use it, but it would save me a lot of time if you were also using it, then I wouldn't have to capture, save and upload pictures of everything.
If the goal is concert-like sound and "wow" factor, and specifically NOT realism, 'fixing' something that happens at concerts (and that hasn't previously bothered the OP) seems like a form of mission creep.
Something that doesn't seem to have come up, yet:
A rock band's CDs are usually a lot more compressed than their concerts, aren't they? I've seen that some people process their CD sound to try to restore the dynamics.
That might be worth looking into... e.g. this guy likes to set his system to exaggerate the dynamics. His approach seems to fit what Entropy is hoping for, with his talk of "kick" and "wow" factor.
www.Basspig.com The Bass Pig's Lair
EDIT:
Another (free!) way to process / restore CD sound for playback on a big system:
https://community.klipsch.com/index...udacity-remastering-to-restore-tracks/&page=1
"Most of these recordings cut the bass below 100 Hz"
Last edited:
That's what I was trying to say in post #422. ( "One other thing to consider is that the recordings you are listening to are probably not mixed properly for live level playback - a lot of rock music sounds bass thin on CD so if you do set up a system to play it 'how it should sound' then other types of music may sound bass heavy. " )
Maybe try an expander ?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.