Rational Acoustics



PrestonSoper
June 2nd, 2010, 08:04 PM
I was wondering about other peoples approaches to the request of subs on an aux. Does one put subs on an aux at the request of the person using them or for ones self so as not to put inputs into the subs that don't belong there and the high pass wont totally eliminate from your mix ?
We should all know the relationship between phase and magnitude and how they interact with each other when you raise or lower the magnitude past the point of your initial optimization. I feel like there are alot of installers and engineers that request this are throwing the baby out with the bathwater.
Sometimes, I wonder if we make it back to the same sub level and phase as we would with a L/R mix as it sounds pleasing to the ear.
I would love to know some of your feedback in your approach and optimization of subs on an aux.

Harry Brill Jr.
June 2nd, 2010, 10:20 PM
When I was doing what you do on a regular basis, I had my console prewired with multipins. I drove the sub from the mono bus. When someone would ask for the sub on the Aux, I would explain my reasons for driving the sub from the mono, but I was prepared to move that XLR if needed. Using the mono output is the best of both worlds and it's rather convenient having the mono fader next to the main left and right faders for those occassions we need to turn the entire mix down a bit. By assigning only sub instruments to the mono output we can eliminate some instruments from the subs while we are able to maintain the balance between the subs and the mains. I had a few folks that wanted everything driven from the stereo bus (easily done on the processor). I only ever had 1 guy insist on having it moved to an aux, because he indeed used it as an EQ. These days I would likely just drive all the various sub systems from the matrix. This allows switching between stereo fed, aux fed, or mono fed, with no need to repatch.

Nexo has a nice simple explanation of some of the issues with aux fed subs in their operation manuals.

When it comes to making BEs happy, I always figured if the worst engineer shows up and works on my system, my job is to make sure this is the best day of his tour.

PrestonSoper
June 3rd, 2010, 01:47 PM
When zero is zero and you have optimized the system within the constraints of sub on an aux for your client, Where does where does the correct phase or altered phase end up after being changed in the venue during playback?
I would ponder that we both end up close to the same place, I have not measured this and with load in schedules what they are it's not really been an option usually. Anyone out there done a measurement on this to see how far out of whack we actually are or are not throwing our sub alignment out?

James Woods
June 4th, 2010, 06:16 AM
This is one of my favorite subjects!

I am one of those guys who loves driving subs from aux sends. However I have had to come to terms with the smaller mid and small size line arrays that just donīt produce enough LF to get the job done, and so I have to run the stereo signal into the L & R subs to get a proper spectral balance on my system.

I have also encountered a lot of guys who do not understand the benefits of driving subs from the aux, and who do not want to use this method. So as a System Engineer what I usually do is this:

I set up my system processor to be able to drive the system and be able to align it. Usually this means splitting up the Main PA drive into two or more sections for each side (PAL or PAR). I generally use a processor that has a matrix or mix capability to send the different inputs where I need them. For example:

Supposing I have a four input system processor I might route things like this:
Input A = Mix L Console Output
Input B = Mix R Console Output
Input C = Aux Console Output (for Subs)
Input D = Matrix Console (or Mono) Output for FF, Lateral Fills, etc.
If I have more inputs I will split things up even more

Supposing a medium sized self powered line array:
Output 1 = Main L PA Upper
Output 2 = Main L PA Lower
Output 3 = DF L
Output 4 = Main R PA Upper
Output 5 = Main R PA Lower
Output 6 = DF R
Output 7 = Subs (all driven in mono)
Output 8 = FF

So, what I do is route or mix my A and B Inputs (stereo console Output) to the approriate places:
Input A to Output 1, 2, 3 and 7 and 8
Input B to Output 4, 5, 6 and 7 and 8
Input C to Output 7 (just the subs)
Input D to Output 8 to whatever (FF?)

OK so now I have done my routing. If I set up my system correctly by using the internal mix or routing of the system processor I will have a mono output of the stereo signal to the subs that will track whatever the main mix is doing. This will keep happy all those guys who prefer no subs on auxes. It also keeps the subs alighned to the main PA (especially for those mid and small size line arrays). I can also balance the system where the subs are just adding extra energy to the low end to make the system do what I think it ought to do, but not muddy up anything.

By doing the above, the non aux sub engineer will have total control, just as he expects. And, the guy who wants the subs on an aux still has them to add extra low end to bass, kick, keys, and sound effects and not add unwanted extra low energy from the stage.

The only problem is that you have to use a processor that can do a mix or matrix between their inputs (not all of the popular processores can do this).

I have been using this set up in festival situations a lot. It is transparent and intuitive to use. I have also found out that some guys who want the subs on an aux end up not using it.... I can only think that it is because the system I provide and align gives them the sound and headroom necesary fro what they want to do.

lundbergsound
June 5th, 2010, 02:54 AM
If the mixer guy wants subs on an aux, as he often does, I will give it to him, but I will calibrate the system so that he'll get a flat frequency response all the way down if he sets the aux send level to -10 dB (borrowed from cinema standards - a little arbitrary, but it works). This way, if there's prerecorded video, for instance, he can set the aux send at -10 and accurately reproduce video, but he can use the aux to taste for all live inputs.

PrestonSoper
June 7th, 2010, 10:45 AM
So, what I do is route or mix my A and B Inputs (stereo console Output) to the approriate places:
Input A to Output 1, 2, 3 and 7 and 8
Input B to Output 4, 5, 6 and 7 and 8
Input C to Output 7 (just the subs)

By doing the above, the non aux sub engineer will have total control, just as he expects. And, the guy who wants the subs on an aux still has them to add extra low end to bass, kick, keys, and sound effects and not add unwanted extra low energy from the stage.

James,
This appears to be an interesting approach to gaining better tracking between subs and tops. Do you have any issues with double routing of the sub signal?

With respect to assigning subwoofer levels on each channel auxiliary bus level, there exist two distinct schools of thought. The first, more technically correct and purist, aims to reflect the same frequency balance that would be achieved without driving the subwoofers separately. We must therefore calibrate the channels' aux levels as well as the main aux controls so that the subwoofer level is the same as if we had driver them in the conventional way.

The second school of thought would be more liberal, and would assign subwoofer level on each channel subjectively as required by each instrument.


James,
With your approach it would seems like as it would be a sweeten to taste type approach, but doesn't allow for the BE that wants control over what goes into the subs, all l/r inputs drive the sub out.
l look forward to giving this application a go as well as Harry's suggestion of using the mono output to drive the subs.

Harry Brill Jr.
June 8th, 2010, 10:25 PM
So, what I do is route or mix my A and B Inputs (stereo console Output) to the approriate places:
Input A to Output 1, 2, 3 and 7 and 8
Input B to Output 4, 5, 6 and 7 and 8
Input C to Output 7 (just the subs)
Input D to Output 8 to whatever (FF?)

By doing the above, the non aux sub engineer will have total control, just as he expects. And, the guy who wants the subs on an aux still has them to add extra low end to bass, kick, keys, and sound effects and not add unwanted extra low energy from the stage.

Hey Gym,

The issue I have with your method is it does NOT remove energy from the subs from instruments that should not be there. I come from the festival side of things and we would often have blue grass and country bands with lots of open mics. The subs would start howling even with the 75Hz HPF engaged on every channel (what that console was capable of). I've experienced this with symphonic gigs too using real desks. I also recomend extreme caution when mixing outputs with different paths. Often times Aux outputs do not share the same phase response with main outputs. Mixing them back together could cause a lot of problems. Obviously you haven't had to deal with this, but some folks reading may run into this issue. Nexo has a good write up on this in their manuals.

PrestonSoper
June 17th, 2010, 03:18 PM
(I also recomend extreme caution when mixing outputs with different paths. Often times Aux outputs do not share the same phase response with main outputs. Mixing them back together could cause a lot of problems.)

This is where I have an issue with the above mentioned method proposed by James. It does not take into account for selective sub placement on inputs and may have phase issues from multiple sources summed back to one source as Harry elaborated on in the previous post. Subs on an aux , a necessary dilemma we are faced with living with in our installs.
I like the mono output idea yet some BE like that control over the ratio of sub content that a mono output bus routing will not achieve. When time permits I will be doing some experimentation with all three methods that have been discussed in this thread and posting screen shots. In most playback excluding the heavy bass driven material I'm fairly certain most engineers arrive back to the point they would with a l/r or mono routed sub drive, it's a matter of convincing them to live outside of a comfort zone of the aux. I will give the engineer what they like but it's a matter of the effort we put in to obtaining proper phase response thru out our lf/sub region that I would like to keep valid.

PrestonSoper
June 17th, 2010, 06:19 PM
Supposing I have a four input system processor I might route things like this:
Input A = Mix L Console Output
Input B = Mix R Console Output
Input C = Aux Console Output (for Subs)
Input D = Matrix Console (or Mono) Output for FF, Lateral Fills, etc.
If I have more inputs I will split things up even more

Supposing a medium sized self powered line array:
Output 1 = Main L PA Upper
Output 2 = Main L PA Lower
Output 3 = DF L
Output 4 = Main R PA Upper
Output 5 = Main R PA Lower
Output 6 = DF R
Output 7 = Subs (all driven in mono)
Output 8 = FF

So, what I do is route or mix my A and B Inputs (stereo console Output) to the approriate places:
Input A to Output 1, 2, 3 and 7 and 8
Input B to Output 4, 5, 6 and 7 and 8
Input C to Output 7 (just the subs)
Input D to Output 8 to whatever (FF?)

OK so now I have done my routing. If I set up my system correctly by using the internal mix or routing of the system processor I will have a mono output of the stereo signal to the subs that will track whatever the main mix is doing. This will keep happy all those guys who prefer no subs on auxes. It also keeps the subs alighned to the main PA (especially for those mid and small size line arrays). I can also balance the system where the subs are just adding extra energy to the low end to make the system do what I think it ought to do, but not muddy up anything.

By doing the above, the non aux sub engineer will have total control, just as he expects. And, the guy who wants the subs on an aux still has them to add extra low end to bass, kick, keys, and sound effects and not add unwanted extra low energy from the stage.

The only problem is that you have to use a processor that can do a mix or matrix between their inputs (not all of the popular processores can do this).

I have been using this set up in festival situations a lot. It is transparent and intuitive to use. I have also found out that some guys who want the subs on an aux end up not using it.... I can only think that it is because the system I provide and align gives them the sound and headroom necesary fro what they want to do.[/QUOTE]




This is the method that James was proposing mixing running the subs to left / right as well as an aux on the subs.
I think that two paths summed back together at the speaker may create phasing issues and doesn't achieve exclusion of
subs on discrete channels.

James Woods
June 24th, 2010, 01:17 PM
Preston,

I started writing this and got waylaid. So here goes:

The sub signal comes from the system processor (usually a Meyer Sound Galileo), in my example I am using Output 7. This is a mono signal to all of the subs (usually in a horizontal line in front of the stage).

I have not had timing problems with mixing the signal from the main stereo mix and the auxiliary send in the system processor. This may be because the consoles I have been using (Yamaha PM-5D and Midas Pro6) probably have the same latency on their outputs. Not all consoles are like this. Yesterday I was at a theatre having an interesting conversation with the tech guy from the manufacturer, the FOH mixer and the guys from the sound company supplying the gear. Turns out that the (a very popular Big Name brand) console can have different latency on its outputs. How interesting. But this is another subject. If this is the problem you are having with subs, then the problem is with the mixing console and not the subs system.... (Reminds me of the hum on stage. It is everywhere and someone says to the sound guys where is that humm comming from? More often than not I have found the humm problem comming from the fans on the moving lights. Not a sound problem at all, but we get pinned with the problem anyway (and have to waste a lot of our time "Finding the problem").

So what we need to do is to use Smaart to measure the console! and depending on what we measure, we can then decide what we want to do.

Think of wavelengths, distance, speed of sound, and phase differences. Small timing differences (phase displacement) in the low end is always there, even in the pure analog world that I know so well. What I am getting at is that even if there is a few samples of delay between outputs on the low end (below 100Hz) you will still get a sum, as the comb filter will be happening way above the crossover point.

Next item, How do you want to set up your system for your application? Do you always do the same thing? Do you always use the same PA system set up the same way for every gig? I guess you get what I mean, every gig is different and needs its own particular solution, although we all tend to generally work along the same lines of thinking, the method of getting there in a particular application may be slightly different for a variety of reasons.

If I am using a smaller PA system that does not have the low end needed to sound balanced, I will usually use the Matrix approach described above. With this method I usually just add in enough subs to flatten out the frecuency response (or a bit more to taste, but tipically not more that +6dB).

It I have a large format PA with doble 15īs that can go down to 50 or 60Hz, I usually have the main PA do all of the work for the low end. For me I prefer to have the low end punch come from the some source as the rest of the signal, and augment the pant leg and gut punch with very low end subs on an Aux (keeping the vocals and other things out of the low end as is what Harry (and I) prefer.

In the end, it is just as Harry says, Our job is to make the BEīs as happy as possible.

I would love to here what you try out there and see the measurements!

Rasmus Rosenberg
July 6th, 2010, 04:55 PM
When zero is zero and you have optimized the system within the constraints of sub on an aux for your client, Where does where does the correct phase or altered phase end up after being changed in the venue during playback?
I would ponder that we both end up close to the same place, I have not measured this and with load in schedules what they are it's not really been an option usually. Anyone out there done a measurement on this to see how far out of whack we actually are or are not throwing our sub alignment out?

Well speaking rock'n'roll ( well Hiphop, metal etc ) and not theater og symfonic work. The most BE that requests sub's on an aux seems to be the ones that "can't get enough sub". By that I mean that even if you doubled the amount of sub's they could still need "a little more". So I would say that we end up with more sub level.. So do we screw up the alignment? well depends.. Where are we starting? Where are we ending? What are we comparing to? Normally we align the PA and subs where they interact the most, at equal level...right?? So how do we measure it, when that point changes as the BE pleases?
And also if we view the auxs sub as part of our system, level and align them to work in conjunction with the rest of the system, how can we keep it as a part of the system, if the BE makes an "artistic" decision that one second the aux sub's are an part of a system and the next its "it's own" system or the kickdrum system etc??

Sorry if im confusing the thread or messing up concepts, but as sad as it is, the most people that I encounter should spend more time on mixing than to worry about getting there sub's on and aux and "play" system tech...

I have been using Josh's metode of aligning the aux sub's to the PA at the lowest expected drive level and aligning from that crossover frequence and up. it does seems to work quite good,but that's basically just guessing.
Mvh
Rasmus

Chris Johnson
November 23rd, 2010, 12:29 PM
Hey Preston,

Long time no see - we met a few times doing Joyce Meyer stuff with Hillsong London. As a sidenote, i enjoyed reading your views on MLA - i'm thinking of putting it into London's O2 arena next year for an event im handling Audio for.

My approach to 'Aux Fed' subs is as follows:

I like to derive a seperate send from the console for subs for control reasons. It allows me to define what sources get that extra low frequency extension and use less of the HPF which often is either too steep or too shallow in terms of dB/Octave rolloff to give the desired effect.

Now, i appreciate the need to keep the system crossover point the same and not undo the work of my System Tech so i make sure of the following things:
1. I usually use a fixed point send to the subs. This way there is no variability, its either in or out
2. I make sure that the feed to the subs is phase aligned with the feed to the mains.
3. I make sure that when an input is fed to the subs, the level and phase matching mean its the same as if the mains had been routed to the tops and subs.
This way, i'm not so much adding sub, but removing inputs from them.

Partly i can use this method because i'm mainly mixing on larger format systems where the main arrays have a pretty full-range response, so the sub is more to reinforce the low-end, and provide some floor coupling, as opposed to doing the job of a low-mid driver.

By doing all of the above, i gain control over the feed to the subs while keeping phase-alignment and crossover points unchanged.

The 'can't get enough sub' comment is valid, and i have seen engineers work like this. However, sometimes this has been because the system has been sub-light to begin with. Generally when working in pop/rock and hip-hop live events, we BE's seek to have what l'acoustics call a 'reinforced-contour' response and some lesser System Tech's don't understand that and then thats when the aux-fed sub becomes a get out of jail card.

Thats my 0.02

Kip Conner
December 14th, 2010, 11:36 PM
I'll take the aux on sub anytime I can get it... the reason being is that the majority of the systems that I work on haven't had anyone to properly align the subs in the acoustic crossover. I find that in the quest for more more low end (in a world that most times doesn't require it) we have so much overlap. The end result is cancellations and lost headroom.

As a system engineer I'm fine with setting up the system all inclusive and am ready to go to auxes if requested. The delay doesn't change as long as the aux out is set to zero. There are benefits in both worlds as long as the engineer understands what they are doing with what they have asked for in the system.

Kip Conner
December 16th, 2010, 04:20 AM
I had aux fed subs tonight at the Knitting Factory in Boise. From a mix perspective I found myself using the send on the aux as an effect during some songs. Oddly, I do it all the time but never really considered that way. It's not correction, I was driving the subs with acoustic guitars to make the mix a little more interesting.

Rasmus Rosenberg
December 22nd, 2010, 04:50 PM
Hey Chris,
Very nice post. I know that I come off strong with regards to sub's on aux and would like to say i love mixing with it my self. And if gear/time permits it will always set it up so I can switch to it fast. The problems I have encounted have been, with your points 2 and 3:

2. I make sure that the feed to the subs is phase aligned with the feed to the mains.
3. I make sure that when an input is fed to the subs, the level and phase matching mean its the same as if the mains had been routed to the tops and subs.
This way, i'm not so much adding sub, but removing inputs from them.

The way I understand you is that you, your self or in cooperation with your system tech do that. Do you carry a measurement rig to make sure that you can do that? or do you expect that your sys tech has one and will do it? Also I assume that you would at least have an idea where to align the sub/mains to etc. All I have encounted (on the bad side of life) have had no Idea and was mot helpful in any way to what they wanted other than just make it work (like last night, where we had twice as much gear and 3 days rehearsal etc.) or before hearing the system at all, have boosted 50hz 15 db.. in a room where you would proberly be better off cutting 50 hz 3 db. But there will always be bad examples.

I like the classification that L'acoustics have done. Very good way of looking at things when talking about the final system response. Here is a link http://www.l-acoustics.com/products-systems-overview-63.html

Tuning with only ears i end up with something looking alot like "reinforced contour". With ears and Smaart it's more like "medium contour". That's if I don't know whats going to happen for festival work with a lot of different band etc. I how ever had my first problems with "reinforced contour" in the Venue the other day. The BE expected "flat contour", setting that up with out a measurement system was a pain. Lucky I wasn't at work as System tech ( I just had to fix an insert cable.) But got woven into very heated debate with the BE, as the SE working had his hands full on stage. I did what I could to make him happy, but he didn't want to use the "array EQ" in the amps ( d&b CPL) or a low shelving filter on the array. He wanted the passive boxes to magicly be bi-amped with seperate High/low control and wanted to change the X-ovr frequences (that are locked in the amps, by the manifacture and can't be acessed).Didn't wan't the subs on an aux.. many Italian bad words to my face later he bugged down and just EQ'ed a few filters on his EQ and made proberly the best sounding show this year.
Leasons learned: If your not hired/paid don't get involved. If involved bring Smaart :)