ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 [2] 3   Go Down

Author Topic: Arraying horn loaded subs,options?  (Read 1659 times)

Peter Morris

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1028
Re: Arraying horn loaded subs,options?
« Reply #10 on: August 08, 2017, 11:43:17 pm »

Eugen,

1) Other than the upper range of the pass band the subs operate in, the dispersion will be dictated by the mouth size, "power", as in SPL, will not be kicked left or right.
2) Horn subs can have a huge rising response, like 12 dB per octave, so your 68hz, 24dB/octave may only result in a 12 dB slope with another +6 dB peak here and there in the upper range.
That can still let a boatload of "junk" through in the range between the subs and tops. This "junk" will vary in phase between the two depending on your listening position and the directivity index of the cabinets- the tops will be omni at their bottom, the horn subs may have less than 90 degrees at the top of their bandwidth.
3) No "hole", just multiple phase problems- the subs need to be phase aligned to the tops through the crossover region, and when that region is wide due to the problems described in #2, all sorts of crap response happens, resulting in sound only being good along the axis which is equalized for flat response.

Bottom line- the array is not the problem, it is the interaction between the subs and tops that is messing with your "bass" perception, which may be most critical in the overlap in the octave you crossed at, 68-130 Hz.

Cheers,
Art

I suspect any or all of the above and possibly a polarity issue with one pair of subs.

Logged

Eugen Jeličić

  • Full Member
  • ***
  • Offline Offline
  • Posts: 140
Re: Arraying horn loaded subs,options?
« Reply #11 on: August 09, 2017, 06:23:55 am »

Eugen,

1) Other than the upper range of the pass band the subs operate in, the dispersion will be dictated by the mouth size, "power", as in SPL, will not be kicked left or right.
2) Horn subs can have a huge rising response, like 12 dB per octave, so your 68hz, 24dB/octave may only result in a 12 dB slope with another +6 dB peak here and there in the upper range.
That can still let a boatload of "junk" through in the range between the subs and tops. This "junk" will vary in phase between the two depending on your listening position and the directivity index of the cabinets- the tops will be omni at their bottom, the horn subs may have less than 90 degrees at the top of their bandwidth.
3) No "hole", just multiple phase problems- the subs need to be phase aligned to the tops through the crossover region, and when that region is wide due to the problems described in #2, all sorts of crap response happens, resulting in sound only being good along the axis which is equalized for flat response.

Bottom line- the array is not the problem, it is the interaction between the subs and tops that is messing with your "bass" perception, which may be most critical in the overlap in the octave you crossed at, 68-130 Hz.

Cheers,
Art


DON:

I misunderstood you. I tought you asked how did i check if all of the four subs on each side were in polarity with each other. Not how i checked if left subs were in polarity with the right subs.

To check if the left and the right side were in polarity, i stood in the center, at FOH. And tried flipping the polarity on one side of the subs. The power loss was huge. When i got them back to the same polarity. The summation at FOH (center) was fine again. Since they should summate at a triangular center in front of the left and right subs. I concluded that L and R subs are in polarity.


ART:

I think we are getting somewhere. Both i and the guy from the PA company suspected that my low crossover setting was part of the weird situation. Because he told me that usually they don't work that way when they crossover the subs at 100hz, or even 110.

Now let's go back a few steps. You wrote that horn loaded subs have a rising response when going from sub, to lowmid frequencies. This is exactly what i noticed when i removed the crossover and did a sinewave sweep from 50 to 150hz with the subs running only. Their power rises vastly the higher you go. I noticed similar things with my own EV T18 horn loaded boxes. This is what gives them a "punchy, boxy" sound. From my perception this can be potentialy usefull for trance and techno music (exactly what is usually the main use of this PA that i worked with). But for mechanical music, it makes your kickdrum boxy as hell. "thump" instead of "boom".

The other thing that made me think this problem is even more prominent with these subs is the fact that the "invader bin" is a design with a sealed chamber. So the driver is sealed from one side, and pushing air into the horn on the other side. The sealed chamber chokes the subs even more down at lower frequencies then it does in the lowmids. Because the driver needs to travel more at lower frequencies so the compressed air in the sealed chamber chokes it's movement more.

The result of this is that i had to go for a 68, and even 65hz 24dB/octave crossover. Which just as you wrote, sloped them down from 65 do 100hz just enough to flatten the response up to 100hz. The sinewave sweep confirmed this.
As soon as i would try opening the crossover up to 75-80hz, i would find myself eq-ing that out of the kickdrum and the bassguitar on the channel EQ.


Now, one thing, that we did not do, and that may be the cause for some of the problems. Is that we didn't delay the tops to the subs. I even brought a measurement MIC and SMAART on my laptop, and was planing to do phase alignment. But we ran out of testing time, and on top of that i saw the situation from a potentialy wrong perspective.

I tought, that considering the top boxes are crossed over at 140hz, 24dB/octave BW. And the subs are crossed over at 65hz, 24dB/octave BW. The top boxes won't reach much into the lowmid 100hz spectrum where the subs start going quiet, so there is no need to phase align them. I tought that i am going to have a hole in the overlapping region that is going to be so wide that it won't make much sense doing any phase alignment.

This was kind of my mistake of listening with my "pre-opinion" instead of with my ears.

However, even tought we suspected the same thing that you wrote, one thing is still not clear to me. These weird sub distribution problems, were very present down at 50hz. That means there was something weird going on with phase problems even down there where the tops don't play any significant amount of material.

Could it be, that the low crossover setting, made a phase anomaly at very low frequencies, that made the Left and right side of subs summate in weird ways? Even tought that the same crossover affected left and right subs and that left and right subs were in polarity with each other?
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8027
  • Atlanta GA
Re: Arraying horn loaded subs,options?
« Reply #12 on: August 09, 2017, 07:47:30 am »




I tought, that considering the top boxes are crossed over at 140hz, 24dB/octave BW. And the subs are crossed over at 65hz, 24dB/octave BW. The top boxes won't reach much into the lowmid 100hz spectrum where the subs start going quiet, so there is no need to phase align them. I tought that i am going to have a hole in the overlapping region that is going to be so wide that it won't make much sense doing any phase alignment.


There is often a BIG difference between the electrical xover and the ACOUSTICAL xover.

The acoustical xover is really all you need to be concerned with (as long the the electrical xover is not stressing the drivers).

A proper phase alignment can make the difference between a hole and a flat response-even if the electrical xover freq are not changed.

It can be a huge hole-and a little bit of alignment will fix it.  At least in one listening position.

How well does this alignment hold up over the audience?  It depends on the actual freq (size of the wavelength) and the distance between the devices being aligned.

You generally want the distance to the listener to be 1/4 wavelength (of the acoustical xover) or less.

A couple of ms can make a huge difference in the overall sound/response/punch etc.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Don T. Williams

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 425
  • Midas Pro 1 & 2, M32, dbTech T12, T8, S30, DM12
    • Q Systems Music & Sound
Re: Arraying horn loaded subs,options?
« Reply #13 on: August 09, 2017, 10:26:59 am »

Eugen, I agree with you and think the left and right side subs don't have the polarity inverted.  You have an interesting and complex problem.
Logged

Art Welter

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1102
Re: Arraying horn loaded subs,options?
« Reply #14 on: August 09, 2017, 11:28:41 am »

ART:
1)The other thing that made me think this problem is even more prominent with these subs is the fact that the "invader bin" is a design with a sealed chamber. So the driver is sealed from one side, and pushing air into the horn on the other side. The sealed chamber chokes the subs even more down at lower frequencies then it does in the lowmids. Because the driver needs to travel more at lower frequencies so the compressed air in the sealed chamber chokes it's movement more.

2)Could it be, that the low crossover setting, made a phase anomaly at very low frequencies, that made the Left and right side of subs summate in weird ways? Even tought that the same crossover affected left and right subs and that left and right subs were in polarity with each other?
Eugen,

1) A front loaded horn (FLH) driver's excursion actually reduces to a minimum at low frequencies down around Fc (cutoff frequency) below which the horn "unloads", and frequency response and excursion is basically the same as the small sealed chamber would produce on a flat baffle the size of the horn mouth. Obviously, with high power, a steep HP filter at Fc must be used unless you like the sound of flapping cones ;^).
As you can see in the FLH simulation examples below, excursion is greatest a bit under an octave above Fc, about the upper cutoff frequency you chose on the gig in question. IIRC, the "Invader" has a higher Fc than the examples shown, so the upper excursion peak would also be higher.

2) The LF "weird summation" could be the result as Peter showed of a polarity reverse on one or a pair of subs, or likely the building and truck on one side of the PA. If the sub boundary conditions are not the same either side, the pattern won't be either.

Art
« Last Edit: August 09, 2017, 11:40:52 am by Art Welter »
Logged

Eugen Jeličić

  • Full Member
  • ***
  • Offline Offline
  • Posts: 140
Re: Arraying horn loaded subs,options?
« Reply #15 on: August 09, 2017, 03:52:41 pm »

Ivan.

Yes i realize that. But i didn't think some boxes have such prominent natural boosts that i would need to start cutting almost an entire octave lower (65hz), to flatten the response up to the higher octave limit (130hz). So this situation was a bit weird to me.

ART:

How is that possible? When the horn loaded sub starts going very low, the wavelenght starts exceeding the horn lenght by a lot so the horn stops being effective in loading the driver. Because most of the air just moves away at the horn exit. The horn isn't long enough to hold the air in one place under pressure.
That would mean that the driver starts moving under lesser air impedance (smaller load), which would mean it moves more free and moves further?

That's why the lowcut filter is applied right? To stop the driver from exceeding xmax and doing damage to itself.
Just like with reflex boxes. Once you pass the port resonance, the box stops loading the driver well. Therefore the driver starts moving too much and flapping, potentialy damaging itself. So you apply a LC filter?

Or my logics failed me? Can you explain what you wrote up there a bit better?
You claim that driver moves the most almost an octave above the LC.
We set the LC for the invaders at 35hz.
« Last Edit: August 09, 2017, 03:55:48 pm by Eugen Jeličić »
Logged

Roland Clarke

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 517
Re: Arraying horn loaded subs,options?
« Reply #16 on: August 09, 2017, 04:28:00 pm »

I think that there maybe some misunderstanding due to English usage.  From my reading of the original post he was describing power alley as the line directly in front of one sub stack, not directly between the speakers as we would describe it.  That would make all the other issues just usual stuff to be expected from some pretty low standard bass horns.

Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8027
  • Atlanta GA
Re: Arraying horn loaded subs,options?
« Reply #17 on: August 09, 2017, 05:10:49 pm »

Ivan.

Yes i realize that. But i didn't think some boxes have such prominent natural boosts that i would need to start cutting almost an entire octave lower (65hz), to flatten the response up to the higher octave limit (130hz). So this situation was a bit weird to me.


If you look at the measured response of many cabinets, and compare the output at 40Hz vs 100Hz, you will see that the 100Hz is quite a bit louder, so it has to be reduced even more to be equal to 40Hz.

Not all are this way-but most are.  they get much louder up higher in the "bass" range.  In fact, it is above 100Hz that a lot of cabinets get their claimed sensitivity-NOT down around 50Hz-where it should be stated.

Some cabinets have had sensitivities for sub cabinets measured above 1KHz.

That is not exactly "sub range" to me.

YES-it is important to look at the freq response-the unprocessed response-to get an idea of what is going on.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Eugen Jeličić

  • Full Member
  • ***
  • Offline Offline
  • Posts: 140
Re: Arraying horn loaded subs,options?
« Reply #18 on: August 09, 2017, 05:35:00 pm »

Roland:

Exactly, that's what i wrote. The lack of power was not in the FOH (center) summation location.
It was in the line in front each of the sub stacks. English is not my primary language but i think i wrote that correct.
Can you explain what do you mean by "standard" issues caused by low standard horns? I'm curios what's happening here.

Ivan:

Is this also the case with reflex boxes or just with horns? It seems to me that horn loaded subs tend to get these 100 plus hertz power rises, non existant 200-500hz harmonics causing honk,boxy sound and similar things.
It seems to me that boxes like G-SUB and similar things don't suffer from these problems. Yes they have some other problems but it seems to me they tend to stay quiet clean above 100hz.

And yes of course. You look at the sensitivity and power ratings and think "wow this is going to be a powerfull box" and then relise it's just a bunch of 100-150hz junk that you don't need. And you are lacking 40-50hz power.
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8027
  • Atlanta GA
Re: Arraying horn loaded subs,options?
« Reply #19 on: August 09, 2017, 06:51:51 pm »


Ivan:

Is this also the case with reflex boxes or just with horns?
It is best to look at the measurements of the particular boxes you are looking at.  They vary.

But yes, many reflex boxes are the same way, with a sloping response, or the output above 100Hz is much louder than the response down at 40-50Hz

Not all horns are created the same, and all reflex boxes are not the same

It is best to NOT to generalize, but to look at SPECIFIC boxes to get actual answers.

Are all gasoline engines the same?  What about diesel?  The answer is no.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!
Pages: 1 [2] 3   Go Up
 


Page created in 0.095 seconds with 21 queries.