Line array calculators indoors - some observations and questions

[TJ Cornish]

I've now got a number of shows on my JBL 4886/4883/Danley TH118 system, and would love to have my observations and theories vetted and/or corrected about show setup and how to compensate for the calculator. I'm using LAC 1.29, which is the latest version. Here's a picture of a typical gig for me:


I normally run the array so the bottom is 10'-12' from the floor. I'm usually throwing between 60' and 90' from the array, sometimes flat floors, sometimes with a rake.
Here's the LAC settings I ended up with for the above show. Note the nice, parallel lines and, theoretically, fairly consistent tonality.

In use, the system sounded pretty good in the front to the middle of the room, but noticeably dark on the FOH riser. I don't have Smaart or the like, but the system has been time-aligned by a friend, and I use the RTA on my GLD surface for frequency-domain measurement.

I suspect the issue is that LAC doesn't consider the room in its calculations, and that it may be fairly accurate for outdoors, but indoors some correction may be necessary. Subjectively it seems that 500Hz and down experiences fairly significant room gain. Here is another LAC plot, this time with a fourth box per side. I suspect this plot would work better in the room, and that the gray and cyan lines would actually look more like the magenta or green lines from 500Hz down, giving more even tonality over the coverage area.

Another anomaly - if I use the EQ filter in the calculator on the bottom box to tone down the HF boost in the front of the room, the whole array moves down - apparently there is a bug in the calculator, and/or I'm doing something dumb wrong.

Am I on the right track?
Thanks for any insight.

 

[Ivan Beaver]

Re: Line array calculators indoors - some observations and questions

Do you have any screenshots of your RTA?

The first step would be to be measure (even RTA is better than nothing) at the seat positions and compare that to the predicted response.

You have 6 seat positions-so 6 RTA measurements would give a good idea of what is really happening.

Test and verify is an art form that often overlooked these days.

When the prediction does not match the reality-it is time to look at the calculator (and what is happening) and see if you can figure out why the difference is there.

And it could be a bug.

But measurements are a required first step to help others to figure out what may be going on.

Verbal descriptions can often be very misunderstood.

 

[David Sturzenbecher]

Re: Line array calculators indoors - some observations and questions

I would say the majority (if not all) of the calculators out there from a speaker manufacture only calculate direct sound, therefor not accounting for the room at all.

Also, I would not recommend applying filters to individual speakers within an array, unless absolutely necessary. Remember that a along with a level change is a phase change which can really mess with how well everything combines.

Bennett had a nice write up to this topic here... https://soundforums.net/threads/3022-Array-Deployment-Question?p=21698#post21698

 

[Evan Kirkendall]

Re: Line array calculators indoors - some observations and questions

Forget the software. You've got 3 boxes per side. Hardly a "line array." You have little to no pattern control in the low Mids. What you're hearing is the HF spilling down from the array and no MF making it's way down front because the array is pointed at the middle of the room. You're hearing a nice imagine in the middle of the room since you're on axis of the whole thing and at FOH it gets dark because you're at the top of the array and those pesky horns aren't coupling as well as the Mids that far back. My suggestion? Max angle between each box and carry on with life. I can't tell you the last time I used software to fly a smaller PA.




Evan

 

[Helge A. Bentsen]

Re: Line array calculators indoors - some observations and questions

I did a lot of these arrays last year, mostly groundstacked, some flown. This was my basic setup for four 4886 a side.

2
4
8

I aimed the mid-point between box 1 and 2 at the far seats, and used some frontfill. IME this sounded pretty good and gave a consistent sound.

The JBL calc doesn't compute low buildup and HF loss correctly. Expect to loose more HF in the rear and a lot more below 200hz overall than the calc tells you indoors. But once you get familiar with it, you can archive some pretty decent results that translates well from the screen to the room

 

[Tom Manchester]

Re: Line array calculators indoors - some observations and questions

Forget the software. You've got 3 boxes per side. Hardly a "line array." You have little to no pattern control in the low Mids. What you're hearing is the HF spilling down from the array and no MF making it's way down front because the array is pointed at the middle of the room. You're hearing a nice imagine in the middle of the room since you're on axis of the whole thing and at FOH it gets dark because you're at the top of the array and those pesky horns aren't coupling as well as the Mids that far back. My suggestion? Max angle between each box and carry on with life. I can't tell you the last time I used software to fly a smaller PA.




Evan

+1. My first thought when you mentioned it sounding dark on the FOH riser is that you probably needed more boxes. Even a 1' riser can add the need for a lot more vertical coverage. You could also try trimming the boxes a little lower so they're pointing more straight out if your at max wrap and still not hitting the audience. Best solution is more boxes though.

 

[TJ Cornish]

Re: Line array calculators indoors - some observations and questions

Thanks for the responses.

Ivan - I don't have RTA pictures. Smaart is on the list eventually. The effect in question was obvious even without measurement, though it would surely be nice to have more detail. I doubt my board has screenshot capability, so I will probably have to resort to phone pictures next time.

David - I appreciate the caution about EQ changes on the bottom box. I'm curious as everything in live audio is a compromise if it would be better or worse than turning the whole bottom box down a bit - vs losing the LF output of the box and a bit more pattern control from the longer array.

Evan - I did not claim to have a "Line Array" in this post, and am well aware of the limitations of a "dash". It would be wonderful to have the budget to hang 8-12 boxes/side, but for the work I do, that's infeasible, and not necessary anyway. For my events, a flown "dash" is significantly preferable to a ground stack or any array of trap boxes that have the output to do the job due to labor, trim height, and rigging limitations. I don't have pattern control below 500Hz in the array, but that's not my area of concern here - my problem area lies in a range where I actually do have pattern control - 2K and up, or so.

Why would you max the angle between boxes? I understand the idea of keeping the angle between boxes constant, but it seems to me that as most boxes including the 4886 don't have the full vertical pattern at extreme high frequencies, that it would be better to use a lower angle - 8° per box or something like that would have fewer HF holes, and that if less than 45° vertical is required, would be better than max angle?

Helge - thanks for your datapoints and experiences with the calculator. One of the frustrating things to me about live audio is the challenge of knowing what to care about and what not to. Clearly the room makes a huge difference, and it's frustrating to have relatively small things in the calculator (tenths of a degree in the array hang, temperature and humidity) and there isn't even an "indoor/outdoor" switch, much less settings for wall materials, etc.

Tom - I actually did tip the array up - the original model I had suggested a 2° downtilt, and life got better when I adjusted to 0°. I suspect that if I did add a fourth box and set the angle between box 1 and 2 to 2° and subtracted that 2° from the existing angle between box 2 and 3 that would have helped tremendously. I could go a little higher with my stands than I have been, but that then requires (at least with the 3 boxes I have) steeper angles, and I get into HF holes in the pattern. Based on the calculator and a lot of playing around, it seemed where I had it was a good compromise, and also stays out of the way of my lighting.

I hope to add a fourth box eventually. I can cheat and run 4 on the yoke without too much additional expense, but going farther than that requires the extremely expensive array frames and Genie/Sumner lifts.

The events I have done with this system have been very successful, and I love the headroom of this system. Watching the limiters during very loud playback, I had 10dB left on the 4886 and 4883, and was at the edge of the TH118s. Every time I deploy it I learn something and get better at it.

Thanks for the thoughts - keep them coming.

 

[Ivan Beaver]

Re: Line array calculators indoors - some observations and questions

Thanks for the responses.


For my events, a flown "dash" is significantly preferable to a ground stack or any array of trap boxes that have the output to do the job due to labor, trim height, and rigging limitations. I don't have pattern control below 500Hz in the array, but that's not my area of concern here - my problem area lies in a range where I actually do have pattern control - 2K and up, or so.

.

Just as a matter of reference-here is a plot using the same audience area-same mic positions same trim height etc.

It is a single cabinet (easily flown), that has pattern control much lower-1 amp channel and costs about the same as just 2 of boxes in the model.

The scales are kinda close

 

[TJ Cornish]

Re: Line array calculators indoors - some observations and questions

Just as a matter of reference-here is a plot using the same audience area-same mic positions same trim height etc.

It is a single cabinet (easily flown), that has pattern control much lower-1 amp channel and costs about the same as just 2 of boxes in the model.

The scales are kinda close

Does your calculator show the SH96? If so, I'm familiar with the box (the SH96HO version anyway) - we demoed it for our church a few months ago. It's the size of a double-18 sub, weighs 200lbs, and would require Genie lifts or the like. Adding the cost of the lifts ($8000 or so for the pair), extra truck rental, and being impractical for all but my largest events is exactly why I own the system that I do.

My 4886 case holds 6 boxes plus yokes and is smaller than one of my TH118's. The system fits on $200 50lb Global Truss ST-132 tripod stands, I can transport it in my standard-length Chevy Express van, assemble it, and fly it by myself up to about 15' with continuously variable angles.

If I were doing an install I wouldn't be using the system that I have, and we very nearly chose the SH96HO (didn't due to a visual issue of the size of the box, and the SH60 which we also demoed didn't sound as good as the other box we evaluated). Because I do portable work in the room du-jour often by myself, large trap boxes simply aren't an option.

 

[Ivan Beaver]

Re: Line array calculators indoors - some observations and questions

Does your calculator show the SH96? If so, I'm familiar with the box (the SH96HO version anyway) - we demoed it for our church a few months ago. It's the size of a double-18 sub, weighs 200lbs, and would require Genie lifts or the like. Adding the cost of the lifts ($8000 or so for the pair), extra truck rental, and being impractical for all but my largest events is exactly why I own the system that I do.

My 4886 case holds 6 boxes plus yokes and is smaller than one of my TH118's. The system fits on $200 50lb Global Truss ST-132 tripod stands, I can transport it in my standard-length Chevy Express van, assemble it, and fly it by myself up to about 15' with continuously variable angles.

If I were doing an install I wouldn't be using the system that I have, and we very nearly chose the SH96HO (didn't due to a visual issue of the size of the box, and the SH60 which we also demoed didn't sound as good as the other box we evaluated). Because I do portable work in the room du-jour often by myself, large trap boxes simply aren't an option.

Yes that was as SH96.

Of course another option would be the SM80-MUCH smaller-cheaper-lighter-easily mounts on a pole etc.

Here is the same model with a single Sm80 (same height (14')) Just for reference.

The big thing is that point source options are a real option.

 

[TJ Cornish]

Re: Line array calculators indoors - some observations and questions

Yes that was as SH96.

Of course another option would be the SM80-MUCH smaller-cheaper-lighter-easily mounts on a pole etc.

Here is the same model with a single Sm80 (same height (14')) Just for reference.

The big thing is that point source options are a real option.

I'm curious what your output model indicates - peak, program, ??

I'm sure there are lots of good options for various scenarios. The fact that I already own the system that I own and that it is working well for me is its biggest advantage over other options.

 

[Ivan Beaver]

Re: Line array calculators indoors - some observations and questions

I'm curious what your output model indicates - peak, program, ??

I'm sure there are lots of good options for various scenarios. The fact that I already own the system that I own and that it is working well for me is its biggest advantage over other options.

It is based on the peak power rating. The continuous could be anywhere from 6 to 10dB or more lower. Depending on the particular music style. 10dB is a "standard" number".

 

[Phil Graham]

Re: Line array calculators indoors - some observations and questions

I've now got a number of shows on my JBL 4886/4883/Danley TH118 system, and would love to have my observations and theories vetted and/or corrected about show setup and how to compensate for the calculator. I'm using LAC 1.29, which is the latest version. Here's a picture of a typical gig for me:
I normally run the array so the bottom is 10'-12' from the floor. I'm usually throwing between 60' and 90' from the array, sometimes flat floors, sometimes with a rake.
Here's the LAC settings I ended up with for the above show. Note the nice, parallel lines and, theoretically, fairly consistent tonality.
View attachment 10441
In use, the system sounded pretty good in the front to the middle of the room, but noticeably dark on the FOH riser. I don't have Smaart or the like, but the system has been time-aligned by a friend, and I use the RTA on my GLD surface for frequency-domain measurement.

TJ,

I might describe it in different terms, but Evan is on the correct track here. You're going to need wider splay to get HF down to the front seats, and you need the array's overall HF coverage lobe and low mid coverage lobe to have more similar power responses.

Some thoughts:

1. Why the heck are you looking at LAC at 1kHz? This has no bearing on balancing the arrays coverage in the room, and is indeed in the octave where most line arrays have the strongest undesirable grazing side lobes to muddy the waters. Try using 250, 500, 4k, and 8k as your frequencies to balance and compare coverage.
2. Give up on trying to make the coverage even from front to back, but instead aim for consistent tonal balance in each section of the room. You don't have enough array, nor is it high enough, to get the thing even from front to back.
3. The array is aimed too low. The primary lobe finger at 8kHz should be above standing ear level at the back of the room.
4. You can use staggered high passes and/or delay on the low frequency bandpass of the top box in the hang to raise the primary axis of the low mid lobe
5. There's nothing wrong with attenuating and/or shelving the high end of the bottom most box to balance the HF behavior. Splay the boxes to get the audience in the coverage pattern, and deal with the VHF magnitude with processing.
6. Don't fret to process boxes and/or bandpasses separately, and remember that the lobing behavior adjacent to the array doesn't matter. All that matters is the result at the audience plane. The "special snowflake" processing approach to line arrays is something I strongly advise against.

Ping me offline, and I'll see if I can't get your LAC file cleaned up to behave more like you want.

 

[TJ Cornish]

Re: Line array calculators indoors - some observations and questions

Thanks Phil - replies inline...

TJ,

I might describe it in different terms, but Evan is on the correct track here. You're going to need wider splay to get HF down to the front seats, and you need the array's overall HF coverage lobe and low mid coverage lobe to have more similar power responses.

I had plenty of HF at the front of the room. It was my understanding that aiming in the calculator for parallel lines (first picture and the situation I used for this show) would have the effect of making the shape of the high/mid lobes as similar as possible. What am I missing?

Some thoughts:

1. Why the heck are you looking at LAC at 1kHz? This has no bearing on balancing the arrays coverage in the room, and is indeed in the octave where most line arrays have the strongest undesirable grazing side lobes to muddy the waters. Try using 250, 500, 4k, and 8k as your frequencies to balance and compare coverage.

I'm not looking at the picture as much as the frequency/SPL graphs. One of the pictures was 2K, the other was 8K.

2. Give up on trying to make the coverage even from front to back, but instead aim for consistent tonal balance in each section of the room. You don't have enough array, nor is it high enough, to get the thing even from front to back.

I'm not aiming for consistent level, but rather consistent (to the degree possible with the gear I have) tonality. My observation was that the front to perhaps 2/3 of the way back was pretty similar sounding, but dark at the back.

3. The array is aimed too low. The primary lobe finger at 8kHz should be above standing ear level at the back of the room.
4. You can use staggered high passes and/or delay on the low frequency bandpass of the top box in the hang to raise the primary axis of the low mid lobe
5. There's nothing wrong with attenuating and/or shelving the high end of the bottom most box to balance the HF behavior. Splay the boxes to get the audience in the coverage pattern, and deal with the VHF magnitude with processing.
6. Don't fret to process boxes and/or bandpasses separately, and remember that the lobing behavior adjacent to the array doesn't matter. All that matters is the result at the audience plane. The "special snowflake" processing approach to line arrays is something I strongly advise against.

I don't understand what you mean in number 6?

Ping me offline, and I'll see if I can't get your LAC file cleaned up to behave more like you want.

Thanks - I'll send you the LAC files.

 

[Phil Graham]

Re: Line array calculators indoors - some observations and questions

Thanks Phil - replies inline...

I had plenty of HF at the front of the room. It was my understanding that aiming in the calculator for parallel lines (first picture and the situation I used for this show) would have the effect of making the shape of the high/mid lobes as similar as possible. What am I missing?

You've got plenty of HF at the front because the array is aimed too low, coupled the inverse square law consequences of your low trim height/short array pairing.

I'm not looking at the picture as much as the frequency/SPL graphs. One of the pictures was 2K, the other was 8K.

Roger. The audience plane marker targeted frequency response in the bottom right look suspiciously smooth, and therefore probably not trustworthy, even though the principle you are aiming for is correct. Also, the low and mid frequency behavior of the room, as the array's power response opens up, is probably not accounted for in those graphs, and is having influence in the space. At high frequencies the direct field response of the array is what the audience experiences, but at low and mid frequencies it is the collective array + room power response that sets the tonality.

I suggest looking instead at the predicted directivity balloons at a number of frequencies, and ignore the pretty broadband predicted curves from LAC that clearly didn't correlate to the room behavior.

I don't understand what you mean in number 6?

What I mean is that all of the "line arrays are perfect, don't process any part of the array differently," is complete BS. Intra and/or inter box processing changes for a line array are often complex and unintuitive, but it doesn't mean that you shouldn't seek to achieve the desired array behavior through processing. Further, it doesn't matter how crazy LAC's polar balloons look in mid air above the audience, all that matters is consistency at the plane of the audience's ears.

 

[Max Warasila]

Re: Line array calculators indoors - some observations and questions

Further, it doesn't matter how crazy LAC's polar balloons look in mid air above the audience, all that matters is consistency at the plane of the audience's ears.

Unless of course they extend too far upward and hit the ceiling in this indoor case.

 

[TJ Cornish]

Re: Line array calculators indoors - some observations and questions

Unless of course they extend too far upward and hit the ceiling in this indoor case.

Unfortunately, sound doesn't work like a cartoon - some gets there whether the bubble shows it or not.

I think my scenario illustrates a lot of the challenges an average person faces in live audio. There are always compromises - budget, labor, time, the room acoustics, placement constraints, finite numbers of amp channels and boxes, comprehensiveness of the prediction tools, etc. I'm willing to be called dumb and/or be wrong about my approach or conclusions - that is frequently how learning is accomplished. Seeing how others choose to balance the constraints and what factors they pay attention to is also very instructive.

Before buying this rig, I spent probably 20 hours playing around in LAC, as well as several sessions with my friend/dealer who I bought the gear through. In the process, I also spent a bunch of time in the RCF, DBTechnologies, and Nexo calculators while looking at offerings from those vendors. All but LAC showed what I can only describe as very "optimistic" results - highly smoothed and/or simplified data that was magical. Even LAC clearly, though it offers much more data than other calculators, has limitations.

I know of several folks who consider the calculator's output to be gospel, though I think it's clear that there are more variables. Rules of thumb are helpful, and a calculator can offer a lot of insight, but it sure seems that there is a lot of room for the knowledge gained through experience to augment or correct the tools.

I'm grateful for a place with lots of people smarter than I am to ask questions of.

 

[Tim McCulloch]

Re: Line array calculators indoors - some observations and questions

Unfortunately, sound doesn't work like a cartoon - some gets there whether the bubble shows it or not.

I think my scenario illustrates a lot of the challenges an average person faces in live audio. There are always compromises - budget, labor, time, the room acoustics, placement constraints, finite numbers of amp channels and boxes, comprehensiveness of the prediction tools, etc. I'm willing to be called dumb and/or be wrong about my approach or conclusions - that is frequently how learning is accomplished. Seeing how others choose to balance the constraints and what factors they pay attention to is also very instructive.

Before buying this rig, I spent probably 20 hours playing around in LAC, as well as several sessions with my friend/dealer who I bought the gear through. In the process, I also spent a bunch of time in the RCF, DBTechnologies, and Nexo calculators while looking at offerings from those vendors. All but LAC showed what I can only describe as very "optimistic" results - highly smoothed and/or simplified data that was magical. Even LAC clearly, though it offers much more data than other calculators, has limitations.

I know of several folks who consider the calculator's output to be gospel, though I think it's clear that there are more variables. Rules of thumb are helpful, and a calculator can offer a lot of insight, but it sure seems that there is a lot of room for the knowledge gained through experience to augment or correct the tools.

I'm grateful for a place with lots of people smarter than I am to ask questions of.

Bingo. The prediction tool should be considered a starting point, especially if you use the "Get Me Started" button in LAC. Your observations about the DB Tech and RCF predictions mirror my own observations. After some time with a particular rig you get a feel for what is not being shown or is being obscured by other parts of the presentation and you can correct accordingly.

 

[Phil Graham]

Re: Line array calculators indoors - some observations and questions

Before buying this rig, I spent probably 20 hours playing around in LAC, as well as several sessions with my friend/dealer who I bought the gear through. In the process, I also spent a bunch of time in the RCF, DBTechnologies, and Nexo calculators while looking at offerings from those vendors. All but LAC showed what I can only describe as very "optimistic" results - highly smoothed and/or simplified data that was magical. Even LAC clearly, though it offers much more data than other calculators, has limitations.

Prediction tools are tough, it is hard to disentangle when they are inaccurate for technical reasons from when they are inaccurate due to the data display choices that the manufacturers have made to "pretty up" the results.

Of the prediction tools that I've used, and admittedly I've not used all of them, the only three that I've seen perform in a realistic fashion are Martin Display 2 (MLA), Meyer MAPP, and F-Chart (EAW internal tool). I strongly suspect that EASE Focus 2 could perform similarly if the manufacturers put in "real" enough data. It is not merely enough to sum data accounting for complex plane (frequency and phase) polar response, one also has to simulate and/or measure that data correctly, and account for radiation impedance effects within the array.

Assuming that you undertake all of the above technical bits correctly, then companies have to make the decision to display the results warts and all. Not everyone is comfortable doing this, obviously, especially at the lower end of the industry where users aren't advanced enough technically to tease out the key metrics from the ugliness typical of real data.

 

[Ivan Beaver]

Re: Line array calculators indoors - some observations and questions

Unless of course they extend too far upward and hit the ceiling in this indoor case.

I would not sacrifice coverage on the floor for what "might" happen when the sound hits the ceiling.

Yes we want to keep as much energy as possible from hitting the walls/ceiling-BUT NOT at the expense of direct coverage.

You will never make up the lack of direct coverage-and what goes into the ceiling may be diffuse and lower in level (due to longer path lengths).

I will take what WILL happen over what MIGHT happen any day.