Room measurements, howto?

Get your "what mic?" frustration or "have you heard" out here. The language could get real okka in here mate.

Moderators: Aaronw, kendale, John Sayers

electroaudio
Posts: 8
Joined: Sun Nov 23, 2014 9:45 am
Location: Stockholm/Sweden
Contact:

Room measurements, howto?

Post by electroaudio »

New to the forum here, i have a question later about how to treat a small livestage. But for now i would like to know how do i measure the frequencyresponse "in" the stage, and how to determine what frequencies i should threat.

Also, all the roomnodecalculators ive seen are for rooms with walls all around. this is a stage, so it is like a room where one wall is missing, how does that affect the calculations?
Soundman2020
Site Admin
Posts: 11938
Joined: Thu Aug 21, 2008 10:17 am
Location: Santiago, Chile
Contact:

Re: Room measurements, howto?

Post by Soundman2020 »

Hi there "electroaudio", and welcome to the forum! :)

You don't need to bother figuring out anything about room modes for the stage of a large hall: modes won't be a problem there (unless it is a very small stage).

You could use the REW acoustic application to do some testing of your stage area, if you really wanted to, but most likely you'll find that general broad-band absorption on the walls and above the stage will do the job. To help you decide on how many panels / how big / how thick / where, it would be good if you could post a few photos of the stage and the rest of the hall, and also describe your impressions of what you are hearing there right now.

The shape of the stage can have a large effect on the acoustics, for example. I recently had to treat the stage of a church, where the architect had designed it in a curved shape, with the back wall of the stage basically being constant radius (cylindrical section). The musicians found it impossible to play there, as they were getting flutter echo and multiple reflections of each instrument at all points, so their timing was off, and it just sounded terrible. A half dozen 10cm thick mid-range broadband absorbers solved the problem very well. so a curved stage, or a very wide but shallow stage. or very deep but narrow stage, or one with a low ceiling above, etc., can have a big influence on the acoustics.

Photos, dimensions, and a description of your impressions would be a good start, and perhaps a REW test as well, if you wanted to get a more detailed analysis of the space.


- Stuart -
electroaudio
Posts: 8
Joined: Sun Nov 23, 2014 9:45 am
Location: Stockholm/Sweden
Contact:

Re: Room measurements, howto?

Post by electroaudio »

Hi, it was the REW thing i wondered about how to do ;) i am a total newbie on this.
I do however have a win-XP computer with a measurementmicrophone, which i only have used as a help for finding feedbackfrequencies with a simple FFT program which i guess isnt capable of anything like REW.

The stage is very small, we do have a large stage downstairs with its own sets of problems, but them i have solved by ear with some very narrow notchfilters. but that is not possible to do on this stage.
I will measure the size and do a real drawing and take some photos when i get back, but i would like to do the REW thingie first ;)

There are three walls around the stagearea, all made of painted concrete, they are carrying the weight of a big building so they are really thick. (I think the with is about 3.9m and i would guess the depth is about 2-2.5m, and then the roofheight of 2.12m on that..)
The floor for the stage and the publicarea were made long after the building was made, and it is made out of ironbeams which holds a concrete slab ontop of it, it usually vibrates a lot at 38Hz when the big stage is playing so i have cut that frequency out.
The stage has a height of 2.12m while the room outside is approx 40-50cm higher because the whole roof inside the stagearea is part of the ventilationshaft for the big stage, with a 50cm hole in it in the middle of the stage, and the ventilationshaft is made of concrete.
The concreteroof above it all, must also be really thick to carry the weight of a whole restaurant without any loadbearing columns.

The primary problem is a huge bass buildup in the stage from both the PA and the bassguitar, making it impossible for the musicians both to hear anything in the mess of ringing bassfrequecies on the stage or for the soundguy to make the kickdrum loud enough to hear without feedback.
The sub (2x15" @ 3000w) is now in a position where it makes the most impact for DJbased electronic dancemusic, (which from the beginning was what that area only was used for) But a movement of the subs of 5-10 cm in relation to the stageroom also decides how much bass there is in different parts of the audiencearea, so the resonances in the stageroom is really strong.
Secondary problem but much easier to solve is a messy midrange.

Best regards, Ted
electroaudio
Posts: 8
Joined: Sun Nov 23, 2014 9:45 am
Location: Stockholm/Sweden
Contact:

Re: Room measurements, howto?

Post by electroaudio »

I found a program called roomeqwizard http://www.roomeqwizard.com/ with google, will install it and play around with it at home today and bring it to the venue tomorrow.
All the help to understand the measurements seems to be aviable there too :yahoo:

-I think a REW is a very important first step, for example, it wouldnt be useful to calculate on a roofheight of 2.12m if the ventilationshaft is invisible to those frequencies ;)
electroaudio
Posts: 8
Joined: Sun Nov 23, 2014 9:45 am
Location: Stockholm/Sweden
Contact:

Re: Room measurements, howto?

Post by electroaudio »

Here are a drawing of the stage and a old picture of it.

Mdat file wasnt allowed to upload, but from what i can see there are peaks at approx
37,42,49,57,62,75,85,95,113,128,147,166,188,211,227,235,243,273

half a wavelength at 4.13m (width) is 41Hz
half a wavelength at 2.12m (height) is 80Hz
quarter of a wavelength at 2.35m (depth) is 36Hz

There is also a steep leveldrop in the decay between 55 and 72hz (The ventilationshaft acting as a basstrap?)

Now that we can confirm that there is a huge problem here, then what would be the best way to solve this??
Soundman2020
Site Admin
Posts: 11938
Joined: Thu Aug 21, 2008 10:17 am
Location: Santiago, Chile
Contact:

Re: Room measurements, howto?

Post by Soundman2020 »

Mdat file wasnt allowed to upload,
Upload it to a file sharing services, such as Dropbox, then post the link here on the forum.
but from what i can see there are peaks at approx
37,42,49,57,62,75,85,95,113,128,147,166,188,211,227,235,243,273
The peaks in the SPL graphs are basically irrelevant to determining what the acoustic problems are in the room. They show only the frequency response as seen at the measurement mic, but they don't tell you anything about the phase or the time-domain response, nor the way the energy decays over time, nor reflections. All of that information is in the MDAT file, but it's not so easy to interpret. The frequencies you mentioned might or might not be associated with the actual problems you are experiencing. Also, you only mentioned the peaks, but peaks are only part of the issue: A far bigger problem is the nulls...
half a wavelength at 4.13m (width) is 41Hz
half a wavelength at 2.12m (height) is 80Hz
quarter of a wavelength at 2.35m (depth) is 36Hz
That might or might not be relevant. Modal response is a lot more complex than just axial modes, which is what you are looking at. There are also tangential modes and oblique modes, and it is quite possible that only a small set of modes is actually being triggered.
There is also a steep leveldrop in the decay between 55 and 72hz (The ventilationshaft acting as a basstrap?)
Possibly, but unlikely. It is far more probably that it is acting as a resonant cavity, and is resonating at one or more of its natural frequencies.

a large stage downstairs with its own sets of problems, but them i have solved by ear with some very narrow notchfilters.
Notch filters do not fix acoustic problems. No EQ can fix acoustic problems. EQ can disguise some problems to a certain extent, but it cannot fix acoustic problems properly, because acoustic problems are not in the frequency domain: they are also in the time domain, and EQ can do nothing about things that happen after the sound leaves the speakers...
I found a program called roomeqwizard
Yup, thats REW! Room Eq Wizard"...
All the help to understand the measurements seems to be aviable there too
There's a lot of info there, but analyzing REW data completely isn't as easy as it sounds. The guy who wrote REW (another "John"!) also posts over there, and you can ask him, but he's very busy as you can imagine, and he can't answer everything.
-I think a REW is a very important first step, for example, it wouldnt be useful to calculate on a roofheight of 2.12m if the ventilationshaft is invisible to those frequencies
REW won't help you with that, and in fact your ventilation shaft won't be "invisible" to any frequencies. But since you mentioned it, what is the problem with the shaft? Is it causing problems? What problems?
the with is about 3.9m and i would guess the depth is about 2-2.5m, and then the roofheight of 2.12m on that..
That is indeed a very small stage! And with a very low ceiling as well... That's about the size of a bedroom, not a stage! How many musicians are on there, and what instruments are you playing?

The primary problem is a huge bass buildup in the stage from both the PA and the bassguitar, making it impossible for the musicians both to hear anything in the mess of ringing bassfrequecies on the stage or for the soundguy to make the kickdrum loud enough to hear without feedback.
It sounds like you have three different problems there, and you probably won't like what I'm going to say, but I'll say it anyway.
Now that we can confirm that there is a huge problem here, then what would be the best way to solve this??
OK, so here goes: the wqay I see it from what you have said so far is that you have these three basic problems:

1) Lousy acoustics on the stage. This can be treated to a certain extent with acoustic absorption panels, but the stage is too small for treatment to be the only solution. There is not enough room on the stage to install the amount of treatment that you would need. You could put up some thick OC-703 panels on the walls (at least 10cm thick, and preferably 15cm or 20 cm, spaced at least 20cm away from the walls), and also on the ceiling, and they will help to a certain extent, but that's not the real solution here.

2) Stage monitoring issues. Pretty much all musicians that I have every worked with are partially deaf (not belittling: just stating fact), which means that they want the stage monitors LOUD. With your stage, it is not an option to have loud stage monitors. They MUST be kept down very, very quiet, or they will just be triggering all of the stage build up (modes). In fact, with your stage the best idea is not to have any stage monitors at all! This can be solved easily from the technical point of view, but not the human point of view, or the budget point of view. Technically, the solution is to eliminate the stage monitors completely and replace them with an in-ear monitoring system, where each musician has control over his own return mix. However, many musicians find it hard to adapt to in-ear monitoring, and you'll get a lot of resistance at first, but it is the only real solution. If you stick with it, then it will solve a large chunk of your problems. However, a good in-ear system is not cheap. But it is the best solution.

3) Microphone setup / system setup / mix setup. If you are getting feedback on the kick drum, then there signal chain is very badly set up! :shock: :? :!: Very likely you are not using a drum mic setup that is good for that specific stage, and most certainly the signal chain after the mic is not configured or adjusted correctly. If you had correct mic setup, correct gain control, and good gating, then it would be impossible to have feedback related to the kick drum. I would suggest that you try the "Recorderman" mic setup, using just four mics, and put your kick mic deep inside the kick drum, through the hole in the front head, and positioned just an inch or two away from the spot where the beater hits the rear head. Then put a gate on that channel on the console, using the "insert" jack, and adjust the gain and gate to get only the initial attack of the beater striking the head, with just a tiny amount of actual shell resonance. So your kick gate will only be open for about half a second, and there won't be any feedback on that mic at all. You'll also need to heavily damp the kick inside (blankets, pillows, foam, etc.). Also EQ that heavily: pull out the entire mid range form that mic, leaving only the low-end for the "whooommmm" sound, and some mid-high for the actual beater "slap" sound. You will need a second gate for the snare mic, which should be close-mic'd very low to the snare head. EQ out the kick and cymbals. With this system, the overheads are actually the main mics for the drum kit, and the kick and snare mics just add the "bite and snap" that the overheads don't get. If you set it up properly, and also EQ the kick and snare mics properly, then there should never be any feedback. Of course, there should also never be any of the drums mics in the monitor mixes either. It simply isn't necessary. If this is done right, there will be more than enough kick volume in the FOH speakers, and no feedback at all.
The sub (2x15" @ 3000w) is now in a position where it makes the most impact for DJbased electronic dancemusic, (which from the beginning was what that area only was used for) But a movement of the subs of 5-10 cm in relation to the stageroom also decides how much bass there is in different parts of the audience area, so the resonances in the stageroom is really strong.
It sounds like the entire sound system was never installed or configured properly! It sounds like it was just done by a DJ or sound tech to "sound good" for his music, which often happens in clubs. It would be good to get a professional sound engineering company to come in and re-install the system properly. You need someone who understands both sound systems and also room acoustics, who will first test the room in many ways to understand it, then re-install and re-orient all of the speakers where they are supposed to be for optimum results, followed by checking over and re-calibrating all of the equipment in the sound system as well, to optimize it for the speakers and room acoustics.

This is a problem I deal with all the time: when I do consulting for situations like yours with customers like you, invariably the first thing I find is that the speakers are positioned incorrectly, aimed incorrectly, adjusted incorrectly, and are often the wrong type of speaker anyway! Fixing the speaker setup is probably about 25% of fixing the problems I'm hired to fix. The next 25% is the sound system gain structure, equalization and dynamics, all the way through from the mics to the speakers. Just setting up those two items correctly (speaker aiming, gain structure) normally gets a very big smile on the face of the customer. I usually gets comments like "Wow! What did you DO? Is that the same equipment we have always been using? Did you install something new?" Nope. All I did was adjust it correctly. The next 25% is acoustics. Sometimes acoustics accounts for more than that (often a lot more), and good use of acoustic treatment can make a really big difference. In your case, putting in great acoustic panels will make things maybe 20% to 30% better. That's all.

The final 25% is the hardest of all, because it's not about technical things that can be adjusted by turning some knobs or aiming a speaker differently or hanging an acoustic panel. This is the "people" issue, and it has to do with the musicians and the sound operators. They do not like someone coming in from outside to tell them that they are doing things wrong and have to change. The speaker does not care if I move it, the console does not care if I change the gain or the EQ, the gates and compressors don't care if adjust their settings,and the panels don't care where I hang them... but the musicians do NOT like being adjusted! And that's the biggest issue I face in fixing the problems that my customers want me to fix to make their halls sound good. You will face the same issue, I'm sure: the musicians will NOT want to have the floor monitors placed differently, mixed differently, or run at lower levels. They will also NOT want to have in-ear monitors. They will NOT like the sound of the dry, gated, EQ'd drums. And they will want it all to go back the way it was before.... If so, then it will not be possible to get the results you are looking for! :)

Sad to say, but if your band is like the vast majority of the bands I deal with, you'll only ever get the first 25% to 75% of your issues fixed... unless you can convince the musicians to be a bit more humble, think of the audience (instead of themselves) and learn a better way of doing things. In the few places where the musicians really do want to learn, and are willing to make some uncomfortable changes, the results have been quite spectacular....

So that's the basic plan here: Your problems are only partly acoustic, and that part can be improved with acoustic treatment, but no amount of acoustic treatment is going to solve all of your issues, because 75% of them are not even acoustic.


- Stuart -
electroaudio
Posts: 8
Joined: Sun Nov 23, 2014 9:45 am
Location: Stockholm/Sweden
Contact:

Re: Room measurements, howto?

Post by electroaudio »

Upload it to a file sharing services, such as Dropbox, then post the link here on the forum.
I think i need to redo my measurements with a longer timewindow first.
Notch filters do not fix acoustic problems. No EQ can fix acoustic problems. EQ can disguise some problems to a certain extent, but it cannot fix acoustic problems properly, because acoustic problems are not in the frequency domain: they are also in the time domain, and EQ can do nothing about things that happen after the sound leaves the speakers...
Nope, but it will keep the bottles on the shelves in the bar :)
REW won't help you with that, and in fact your ventilation shaft won't be "invisible" to any frequencies. But since you mentioned it, what is the problem with the shaft? Is it causing problems? What problems?
I dont know, i just want to give as much useful info as possible. And i cant say that something is rigid if it isnt ;)
That is indeed a very small stage! And with a very low ceiling as well... That's about the size of a bedroom, not a stage! How many musicians are on there, and what instruments are you playing?
True, "small and cosy" ... as a marketingguy would put it.
It is just about everything possible on it that doesnt attract a big enough audience to be on the big stage.
Mostly amateurs, but we are sometimes trying to make bluesevenings with some good musicians, however last time we had an audience of ONE! during the whole evening...
It sounds like you have three different problems there, and you probably won't like what I'm going to say, but I'll say it anyway.


.. i am here to listen ;)

1) Lousy acoustics on the stage. This can be treated to a certain extent with acoustic absorption panels, but the stage is too small for treatment to be the only solution. There is not enough room on the stage to install the amount of treatment that you would need. You could put up some thick OC-703 panels on the walls (at least 10cm thick, and preferably 15cm or 20 cm, spaced at least 20cm away from the walls), and also on the ceiling, and they will help to a certain extent, but that's not the real solution here.
Hard to cover the ceiling as it is too low already, but not any problems with the walls.
However, it also needs to be sturdy enough for drunken musicians to fall into.
I also need to make storage on the walls for micstands and cables and such stuff.
-First time i did sound there myself there were still sofas on the walls and back then it wasnt any problems that i noticed.
2) Stage monitoring issues. Pretty much all musicians that I have every worked with are partially deaf (not belittling: just stating fact), which means that they want the stage monitors LOUD. With your stage, it is not an option to have loud stage monitors. They MUST be kept down very, very quiet, or they will just be triggering all of the stage build up (modes). In fact, with your stage the best idea is not to have any stage monitors at all! This can be solved easily from the technical point of view, but not the human point of view, or the budget point of view. Technically, the solution is to eliminate the stage monitors completely and replace them with an in-ear monitoring system, where each musician has control over his own return mix. However, many musicians find it hard to adapt to in-ear monitoring, and you'll get a lot of resistance at first, but it is the only real solution. If you stick with it, then it will solve a large chunk of your problems. However, a good in-ear system is not cheap. But it is the best solution.
It is mostly young musicians so supplying a headphoneamp wouldnt be a problem.
3) Microphone setup / system setup / mix setup.
...Then put a gate on that channel on the console...
...With this system, the overheads are actually the main mics for the drum kit...
Gates would have solved/hidden the problem but there are no gates on a mixwiz console.
The usual setup is just one mic inside the kick and one overhead as a backup, the rest of the drums like the snare will be loud enough.
However, i did solder a phasereversal cable this weekend, and that phasereversal made all the problems with the kickfeedback to dissapear.
But before i did that i also noticed that if i stood on a certain point it would also dissapear, however, inte same point a (not connected) basspeaker would aggregate the problem vastly.
invariably the first thing I find is that the speakers are positioned incorrectly
I have been thinking about moving the sub to the other side, would that be good?
Soundman2020
Site Admin
Posts: 11938
Joined: Thu Aug 21, 2008 10:17 am
Location: Santiago, Chile
Contact:

Re: Room measurements, howto?

Post by Soundman2020 »

I think i need to redo my measurements with a longer timewindow first.
Time window is not about measuring: it is about visualizing. Adjusting the time window makes no difference at all to how the measurement is made. It only changes the way you look at the data. The time window adjustment is for taking a closer look at the way sound behaves as it moves through the room, but it uses the data that was already recorded during the measurement procedure. The only adjustments you can make to the measurement procedure are the levels, the frequency range (which should be about 15Hz to 21,000 Hz), the length of the sweep, and the number of sweeps. There is no adjustment for time window in the measurement, since it is meaningless to adjust it there.
However, it also needs to be sturdy enough for drunken musicians to fall into.
:!: :shock: :D Then you'll probably need to put some widely-spaced wood strips across the front of the panel, in front of the fabric.
I also need to make storage on the walls for micstands and cables and such stuff.
Be careful not to make an enclosed closet or box for that! It will be a resonant cavity that will probably mess up the sound even more. I'd suggest that you build a sort of "cage" for that, and hide it behind the acoustic treatment panels where it can't be seen. You could even make the panel to be part of the front door of the mic locker.
Gates would have solved/hidden the problem but there are no gates on a mixwiz console.
Then buy one yourself! You can get very good used gates on e-bay for under US$ 100. Take it with you when you leave, of course... :)
I have been thinking about moving the sub to the other side, would that be good?
It's hard to see from the picture, but it looks like your subs are stacked on top of each other, next to the stage. That's not a good idea. Both subs should be on the floor, and to get even coverage they should be on opposite sides of the audience area. But it's impossible to tell just from looking at a picture or a diagram where they should be. It would need a complete acoustic analysis of the room to determine that.

- Stuart -
electroaudio
Posts: 8
Joined: Sun Nov 23, 2014 9:45 am
Location: Stockholm/Sweden
Contact:

Re: Room measurements, howto?

Post by electroaudio »

The only adjustments you can make to the measurement procedure are the levels, the frequency range (which should be about 15Hz to 21,000 Hz), the length of the sweep, and the number of sweeps. There is no adjustment for time window in the measurement, since it is meaningless to adjust it there.
Well, the measurement i did was not done properly but atleast here it is. i will redo them properly when i get the time next week.
ftp://neotek.electroaudio.se/Public/lillascenen2.mdat
However, it also needs to be sturdy enough for drunken musicians to fall into.
:!: :shock: :D Then you'll probably need to put some widely-spaced wood strips across the front of the panel, in front of the fabric.
Sounds like a good idea :)
I have been thinking about moving the sub to the other side, would that be good?
It's hard to see from the picture, but it looks like your subs are stacked on top of each other, next to the stage. That's not a good idea. Both subs should be on the floor, and to get even coverage they should be on opposite sides of the audience area. But it's impossible to tell just from looking at a picture or a diagram where they should be. It would need a complete acoustic analysis of the room to determine that.
It is only one sub, i understand the advantages of one on each side, the bassalley in good use for once ;) ... and i guess fewer excited nodes(?)
By the way, here is a video of our latest concert on that stage, next event will be in february so i have plenty of time to rebuild everything :)
ftp://neotek.electroaudio.se/Public/20141125_002.mp4
Soundman2020
Site Admin
Posts: 11938
Joined: Thu Aug 21, 2008 10:17 am
Location: Santiago, Chile
Contact:

Re: Room measurements, howto?

Post by Soundman2020 »

the measurement i did was not done properly but atleast here it is.
Right! That only goes up to 1 kHz, and seems to be just for the sub. You should measure the entire range as I mentioned above( 15 Hz to 21 kHz.)

Also, there are four different measurements in that file, but they are not labelled so I don't have any idea what I'm looking at for each one.

- Stuart -
electroaudio
Posts: 8
Joined: Sun Nov 23, 2014 9:45 am
Location: Stockholm/Sweden
Contact:

Re: Room measurements, howto?

Post by electroaudio »

Merry chrismas :D
Yes, that was the old file ;)
-Not so well done...
But it has been a busy month so havent had the time to redo the measurements yet.
However on monday is the last gig , and after that i have some time to redo it.
electroaudio
Posts: 8
Joined: Sun Nov 23, 2014 9:45 am
Location: Stockholm/Sweden
Contact:

Re: Room measurements, howto?

Post by electroaudio »

Hi again, havent had the time to do any new measurements yet.
But i have to fix some storagespace there in a hurry, so what do you think of the wall of insulation with builtin storage i have drawn?
Post Reply