Line array compromises - how to balance?

T

TJ Cornish

Graduate
Jan 13, 2011
1,263
1
0
St. Paul, MN
At the risk of exposing my ignorance, I have a really stupid question about the end goal of a line array, particularly when physical constraints prevent "textbook behavior".

It’s well understood – even by me – that the frequency range that the array has pattern control relates to the length of the line; short lines lose pattern control at a higher frequency than a longer array. This is proportional to the length of the array, so doubling the length of the array gives another octave of pattern control. The problem is that no matter how long the array is, eventually you lose pattern control, and correspondingly lose the ability to equalize volume over distance for some portion of the frequency spectrum on the low end. So, what are you really trying to accomplish?

An example room that I work in a lot would be a hotel ballroom with about 14’ ceilings and maybe 80’ throw from the array hang to the back wall, with a flat floor.

After spending quite a bit of time in JBL LAC (which is really the only simulator I’ve tried that I actually believe the results from – the RCF, DBTech, and even Nexo sims seem to do some “massaging” of the numbers), there seem to be two possible approaches:

1. Optimize the sound pressure over distance for the frequency range the array can control and let all frequencies lower than that be what they are – 6dB falloff over distance meaning the array is more bass and low-mid heavy at the front of the room than at the back, but frequencies in the range where the array does have control are very similar over distance
4886 even.png

2. Attempt to have the same frequency response at all places in the room, which means (for all of the room situations I’ve tried) you give up any attempt at equalizing the volume difference over distance. This seems, based on my noodling, to be best accomplished with a flat array slightly above ear height, which means you’re blowtorch loud at the front of the room.
4886 flat.png

The goal of even frequency response everywhere in the room is a noble one, as is the desire to have levels be as constant as possible. What gives first? Even with a 7’ long array, which seems to be “long enough” based on what I’ve read, and about as long as one can get away with in a room with a 14’ trim height, pattern control only extends down to about 250Hz. On paper, the identical frequency plots over distance for a flat array like in a ground stack configuration seem nice, but no one that I’ve ever heard of chooses a low ground stack when there’s a flying option available, and it’s not hard to think of the challenges of a wall of speakers 5’ off the ground.

Any clarity anyone can offer as to how this is done in “the real world” – meaning real rooms with normal gear, not Coachella with 30’ arrays for 150,000 people, would be appreciated.
Sorry again for the dumb question.
 
Re: Line array compromises - how to balance?

TJ,

In my opinion 14ft trim is too low to use a line array at all. Use a trap system. 18ft lifts are as low as you can go, many of the ballrooms we work in have at least 20ft ceilings.

Even if you only fly a seven ft long array, the extra four or five feet above the heads of the audience helps to keep it from ripping off their heads. And in that situation I feel it is utterly appropriate to do some fairly severe gain shading, having multiple zones is crucial.

Jason












TJ Cornish said:
At the risk of exposing my ignorance, I have a really stupid question about the end goal of a line array, particularly when physical constraints prevent "textbook behavior".

It’s well understood – even by me – that the frequency range that the array has pattern control relates to the length of the line; short lines lose pattern control at a higher frequency than a longer array. This is proportional to the length of the array, so doubling the length of the array gives another octave of pattern control. The problem is that no matter how long the array is, eventually you lose pattern control, and correspondingly lose the ability to equalize volume over distance for some portion of the frequency spectrum on the low end. So, what are you really trying to accomplish?

An example room that I work in a lot would be a hotel ballroom with about 14’ ceilings and maybe 80’ throw from the array hang to the back wall, with a flat floor.

After spending quite a bit of time in JBL LAC (which is really the only simulator I’ve tried that I actually believe the results from – the RCF, DBTech, and even Nexo sims seem to do some “massaging” of the numbers), there seem to be two possible approaches:

1. Optimize the sound pressure over distance for the frequency range the array can control and let all frequencies lower than that be what they are – 6dB falloff over distance meaning the array is more bass and low-mid heavy at the front of the room than at the back, but frequencies in the range where the array does have control are very similar over distance
View attachment 5702

2. Attempt to have the same frequency response at all places in the room, which means (for all of the room situations I’ve tried) you give up any attempt at equalizing the volume difference over distance. This seems, based on my noodling, to be best accomplished with a flat array slightly above ear height, which means you’re blowtorch loud at the front of the room.
View attachment 5703

The goal of even frequency response everywhere in the room is a noble one, as is the desire to have levels be as constant as possible. What gives first? Even with a 7’ long array, which seems to be “long enough” based on what I’ve read, and about as long as one can get away with in a room with a 14’ trim height, pattern control only extends down to about 250Hz. On paper, the identical frequency plots over distance for a flat array like in a ground stack configuration seem nice, but no one that I’ve ever heard of chooses a low ground stack when there’s a flying option available, and it’s not hard to think of the challenges of a wall of speakers 5’ off the ground.

Any clarity anyone can offer as to how this is done in “the real world” – meaning real rooms with normal gear, not Coachella with 30’ arrays for 150,000 people, would be appreciated.
Sorry again for the dumb question.
Click to expand...
 
I'm not sure if it's just jealousy, but I can't help but laugh when I see 2 or 3 boxes rigged way up high like some of our competitors do.

Are the really getting any further than our SRX rig, or is it just line array for the sake of having line array?

We haven't made the plunge because we want to be in a place where we can do it right the first time.
 
Re: Line array compromises - how to balance?

I did a 4886 groundstack last night that I thought sounded quite good. Audience and the talking heads were happy as well.
I've not tried this in LAC or made any measurements, but it sounded pretty good with 12 channels of wireless (5 DPA D-fine+B58), video feeds and a DJ.

I "guesstimated" all the angles as a test to see if I could improve it over the previous design I heard from someone else using 0 and 1-degrees.
From the top:
4x4883, degrees are between boxes.

2
4
8

From the 4886 to bottom 4883:
Used 0 degree on the 4883 flybar to the rear link.
Used the 0 degree hole in the 4886 in the last hole on the rear link, giving me a severe down tilt on the lower 4886 and the upper 4886 almost straight.
The 4883 was mounted on it's flybar on top of a S28 sub.
Amp racks were JBL own touring racks with Itech 12K
Preset were 3x4886 short trow, 160 hz HP
4883 60-160
VTX S28 60.
- 3dB gain on the bottom 4886.

You could definitely archive the same level of performance using a well-designed trap box or two and have a significantly shorter setup time, but that's the tools I was given on this particular job.
 
Re: Line array compromises - how to balance?

In the real world the low end will be a lot closer than it looks in the model.

I wouldn't stack line arrays if I could help it, in your second example I'd be really worried about slapback and consider that the software isn't showing you signal:noise.
 
Re: Line array compromises - how to balance?

I will try to make a practical comment. If you don't have enough length on the "line array" to use that as an advantage, don't bother to try, use the array as a "point source". Also if you have sub's as part of your system, well they are all ready falling off 6db pr double distance. The goals are still the same to try to produce even Spl (not the same number, that would be equal) over the audience area with minimum ripple variance. In this case a tilt of 6db pr. double of distance. If you wanted to produce equal Spl 1ft to 100ft away a ratio of 100 is needed and that equals 40 db, or splay 0 dgr,point all boxes at the back and still would not be enough. Look at your ground stack model you only have about 15ish db ratio front to back. You might consider to say; oki lets cover from 10ft to 100tf with the "array" then you at a ratio of 10 or equal to 20 d&b, and if a -6 db at the coverage edges are acceptable then your at least approaching possible. And properly also why some (and Helge) can achieve a "acceptable result" with a stacked system. We all know that the LAC don't tell us the full picture, as Brandon mentioned, indoor, the low end will be closer than the model, thats why in most cases we can get away with just getting the "Highs" in check. Also a lot of times, the room interaction will do much more damage over the audience area than the difference the model shows between stacked and flown. So you could quickly end up with a design that looks good in the model, but will behave bad anyway. As Jason mentioned above a crucial element, and a short coming of most LAC's is also that in real life we (might) have the possibility to apply processing. So as soon as you venture into using a box the way "it's not supposed to be used" you better have full control over: gain,polarity,frequency response and delay pr box, and use symmetric or unity splay angles in the array. Its a road full of dangers and possibilities to really screw things up, but in a lot of situations its also possible to make an array respond in a way that you are not able to predict, off the bat with the LAC. If its worth the trouble and hassle, and are better sounding, is up to you.
/R
 
Last edited:
Re: Line array compromises - how to balance?

Which is more of a concern I guess, coverage or volume? The short arrays should be fine in a talking head type gig, because what you gain in coverage outweighs the negatives. As far as music, an array that short in a ballroom will not work as well, as others have pointed out.

Now, on to your posted screenshots...I have found with line arrays in general and Vertec in particular, unless you're throwing 300 feet, try to keep the array from ever being flat. Zeroes are NOT your friend once you get above say, 4k. I think JBL even has a white paper out about this phenomenon, and I know other manufacturers have at least implied the same. I think it's the new Martin array that can only go as flat as half of a degree splay?
 
Re: Line array compromises - how to balance?

Jason Phair said:
Which is more of a concern I guess, coverage or volume? The short arrays should be fine in a talking head type gig, because what you gain in coverage outweighs the negatives. As far as music, an array that short in a ballroom will not work as well, as others have pointed out.

Now, on to your posted screenshots...I have found with line arrays in general and Vertec in particular, unless you're throwing 300 feet, try to keep the array from ever being flat. Zeroes are NOT your friend once you get above say, 4k. I think JBL even has a white paper out about this phenomenon, and I know other manufacturers have at least implied the same. I think it's the new Martin array that can only go as flat as half of a degree splay?
Click to expand...

There are several good white papers on line arrays at jblpro.com. VERTEC Tech lib
 
Re: Line array compromises - how to balance?

Jason Phair said:
Which is more of a concern I guess, coverage or volume? The short arrays should be fine in a talking head type gig, because what you gain in coverage outweighs the negatives. As far as music, an array that short in a ballroom will not work as well, as others have pointed out.

Now, on to your posted screenshots...I have found with line arrays in general and Vertec in particular, unless you're throwing 300 feet, try to keep the array from ever being flat. Zeroes are NOT your friend once you get above say, 4k. I think JBL even has a white paper out about this phenomenon, and I know other manufacturers have at least implied the same. I think it's the new Martin array that can only go as flat as half of a degree splay?
Click to expand...

In a ballroom corporate setting, I've used small distributed hangs of baby arrays on multiple occasions. In the typical large hotel ballroom with low ceilings, a distributed system is the best way to go in my opinion.

It's funny how many guys are flat packing Vertecs. It's the worst, but I see it way too often. IMHO, Vertec rigs can sound great, but plenty of people who use them really don't know how to.
 
Re: Line array compromises - how to balance?

David Karol said:
XYZ rigs can sound great, but plenty of people who use them really don't know how to.
Click to expand...
Isn't that true about MOST rigs being used?------------------------

Not using the gear properly (or having a clue about how to) is a real problem.

I saw a screen shot of a digital console recently in which all the eq bands on a mic had been pulled down all the way.

So the response was basically just acting as a 15dB (or greater) pad-with some ripple on the bottom.

At least with a digital console IT SHOWS YOU when you are cutting to much! PEOPLE-When the eq response dips below the 0 line ALL THE WAY across the response-you are cutting to much.

Such a simple thing-but soooooo often overlooked.

THE BASICS ARE IMPORTANT! Believe it or not!
 
Re: Line array compromises - how to balance?

Ivan Beaver said:
Isn't that true about MOST rigs being used?------------------------

Not using the gear properly (or having a clue about how to) is a real problem.
Click to expand...

I agree. I've heard some fantastic sounding Vertec rigs, but the majority have been lackluster.

I'm not blaming the boxes for this. When someone without a clue decides to buy a PA, JBL Vertec is commonly the first choice. Not because it's the best product for them, but because JBL is in it's name.

We all know 4888s can sound very good when well tuned and deployed. But how often do you see a d&b J, or Meyer MICA/MILO array poorly deployed? I think it's a lot more common to see that with Vertecs. I just think a lot of the Vertec rigs out there are in the wrong hands.
 
Last edited:
Re: Line array compromises - how to balance?

David Karol said:
I agree. I've heard some fantastic sounding Vertec rigs, but the majority have been lackluster.

I'm not blaming the boxes for this. When someone without a clue decides to buy a PA, JBL Vertec is commonly the first choice. Not because it's the best product for them, but because JBL is in it's name.

We all know 4888s can sound very good when well tuned and deployed. But how often do you see a d&b J, or Meyer MICA/MILO array poorly deployed? I think it's a lot more common to see that with Vertecs. I just think a lot of the Vertec rigs out there are in the wrong hands.
Click to expand...

Its also a geographical thing, over here we see more bad deployed D&b rigs than Vertec rigs. Putting up a PA is just not that easy, and thats hard to understand. Funny how easy a video wall and media server can be to setup when you consider the impact...36 mtr screen with the company logo 8O~8-O~:shock: WOW.. (while the CEO mic feedsback..8)~8-)~:cool:) how hard can that be you only rig those little black things out there... :twisted:


/R
 
Re: Line array compromises - how to balance?

Thanks all for the comments. I need to take some more time to digest what you've all said and finish the Vertec whitepapers. I've read some parts of them, but had gotten hung up on how the theory (which the whitepapers cover nicely) translates into less than ideal circumstances that I will run into with use - something less clear from the parts of the whitepapers I've read.

Anyway, here are a few thoughts/responses.

Regarding using trap boxes instead: I have a rule for the gear I own - if I can't deploy it myself, I can't own it. I'm not aware of any trap boxes that have sufficient output for what I'm trying to do that I can lift and rig myself. A "vertical array" gives me a number of things that are important to me: I can lift the modules myself, they're pretty much all flyable, I can adjust the amount of gear I bring for the show and have some scale options but still be using the same general tools. In addition to these general benefits, the 4886 boxes can be pole mounted, and with the U bracket and the ability to tilt the array, I have placement options with a portable system that I haven't found in any other product, and a lot of output per cu/ft.

RE Spencer: I'm very confident that there are people who stick a couple boxes up and call it a line array for marketing purposes. I have quite a bit of experience with SRX and similar trap speakers, and they don't do what I need them to do - they (at least JBL SRX) don't (IMO) sound that great, they're almost always deployed too low, and you have very few options for optimizing coverage or placement with them. They get loud, but that's not my most important goal.

RE Helge, Rasmus, Jason(s), Brandon: I appreciate your comments. The idea of treating a small array (3-4 4886) as a variable angle point source is sort of what I'm trying to accomplish.

RE David and Ivan: In audio, it's easy to discover all the challenges of good sound - room interaction, speaker interaction, the Tim Mc "3dB more money" principle, and get discouraged. The reaction of a lot of people is to give up and deploy something that sucks, which is fine, since "nobody cares what it sounds like anyway". I'm hoping that I care enough and have the mental capacity to do a little better than average at managing the circumstances presented, even if my current speaker of choice has an orange badge. :)


Please keep the thoughts coming.

TJ
 
Re: Line array compromises - how to balance?

You bring up a few good points about trap rigs all around, although we'll have to agree to disagree on the sound ;) I wasn't trying to start another Trap vs Line Array debate, we all know the answer to that is to use the one that is right for the job, I was mostly just weighing in on the companies that rig two or three boxes in the air and call it a line array, ignoring the possibility that it may not be the best tool for the job.

I feel like we are in the same boat, sitting on the edge of the line array diving board and trying to jump in without doing a belly flop per say... I'm going to be following this thread with interest, I hope you don't mind if I occasionally jump in with a question of my own.
 
Re: Line array compromises - how to balance?

Jason Phair said:
Now, on to your posted screenshots...I have found with line arrays in general and Vertec in particular, unless you're throwing 300 feet, try to keep the array from ever being flat. Zeroes are NOT your friend once you get above say, 4k. I think JBL even has a white paper out about this phenomenon, and I know other manufacturers have at least implied the same. I think it's the new Martin array that can only go as flat as half of a degree splay?
Click to expand...

I strongly agree! I have used the zero degree pins on a line array once, and that was when I was trying to throw about 400' and was willing to accept the corresponding giant ice pick lobe because of the output it represented.

TJ, when using a small number of line array boxes, I always think the "rules" are pretty simple. You have a LF pattern that you basically can't do anything about. So you must set the angles on the boxes to have the HF pattern approximate the LF pattern. The prediction software should make this pretty obvious. Yes, you're basically treating your line array like a trap array but, as you point out, you can rig a line array yourself and everyone wants one.

There's nothing magic about how a line array works, the advantages are all about having one box in inventory, having custom HF coverage for every venue, and having coherent summation across a wide frequency band. As you add more boxes you can do more tricks, but there's nothing invalid about a 4, 2, or even one box line array. It obviously still works, if the HF coverage will do what you want then life is good.
 
Re: Line array compromises - how to balance?

Brandon G Romanowski said:
In the real world the low end will be a lot closer than it looks in the model.

I wouldn't stack line arrays if I could help it, in your second example I'd be really worried about slapback and consider that the software isn't showing you signal:noise.
Click to expand...


Sometimes a stacked line array works well, especially in a venue with a bowl or even certain theaters with balconies. However in a ballroom like this I completely agree that slapback would become a big problem with a bunch of 0 degree angles. You keep all the HF off the ceiling and floor but put it all at one spot on the back wall. Years ago I made this mistake when I had to hang low in a ballroom due to the video screens taking priority positions. It is funny now but wasn't funny then - the band had IEM so they were fine but the keynote speaker was dizzy halfway through his speech :)




Rasmus Rosenberg said:
I will try to make a practical comment. If you don't have enough length on the "line array" to use that as an advantage, don't bother to try, use the array as a "point source".
Click to expand...

Yup. If I only have 14 foot trim height, I wouldn't hang 7 foot of speakers. You still want to keep the bottom box enough above peoples heads. It will still be loud at the front but hopefully with some gain shading you can keep it within reason. After all, it's only an 80 foot throw, right?




Rasmus Rosenberg said:
Also if you have sub's as part of your system, well they are all ready falling off 6db pr double distance. The goals are still the same to try to produce even Spl (not the same number, that would be equal) over the audience area with minimum ripple variance.
Click to expand...

Which is why it matters a lot if this is a musical event or a corporate talking head event where vocal intelligibility is paramount. Your system design will have compromises either way due to the trim height, but you may make very different rigging decisions.




Rasmus Rosenberg said:
As Jason mentioned above a crucial element, and a short coming of most LAC's is also that in real life we (might) have the possibility to apply processing. So as soon as you venture into using a box the way "it's not supposed to be used" you better have full control over: gain,polarity,frequency response and delay pr box, and use symmetric or unity splay angles in the array. Its a road full of dangers and possibilities to really screw things up, but in a lot of situations its also possible to make an array respond in a way that you are not able to predict, off the bat with the LAC.
Click to expand...

And that can mean needing the same size amp rack to run six boxes per side as you do to run 24 boxes just to have enough control. Which is expensive.




Rasmus Rosenberg said:
If its worth the trouble and hassle, and are better sounding, is up to you.
Click to expand...

It looks nicer :) and can make a difference in getting the show WRT your competition. But it is not easier at all. Which is why there may be some merit to the VRX and QSC KLA fixed angle arrays. I haven't used either (besides a ground supported demo of the KLA) though most reports of them in the wild are negative. I know people are out there making money with them, even if they aren't perfect they may be less unperfect then a butchered hybrid line array. What is your take on that?




TJ Cornish said:
Regarding using trap boxes instead: I have a rule for the gear I own - if I can't deploy it myself, I can't own it. I'm not aware of any trap boxes that have sufficient output for what I'm trying to do that I can lift and rig myself. A "vertical array" gives me a number of things that are important to me: I can lift the modules myself, they're pretty much all flyable, I can adjust the amount of gear I bring for the show and have some scale options but still be using the same general tools. In addition to these general benefits, the 4886 boxes can be pole mounted, and with the U bracket and the ability to tilt the array, I have placement options with a portable system that I haven't found in any other product, and a lot of output per cu/ft.
Click to expand...

These are all good points. I am not hands on familiar with the JBL line, my days as a stage hand unloading other peoples equipment were a while back and many of these boxes were not available then, so I have only deployed the two line arrays that I own (QSC Wideline and TurboSound Flex.) I've mixed on others but that doesn't count.

Are there really NO scalable, single-handedly depolyable point source options with easy rigging that can get loud enough? Seems like a job for Danley-Man :) I guess that was the idea with the TurboSound Flex as well. But they can't be pole mounted and are probably a little heavier than you would need.


Jason
 
Last edited:
Top Bottom