Has anyone been using the QU to “ring out a room”, “Ping a room”, “tune a room”…
Should be relatively straight forward, just wondering if anyone has done it and learned any tricks.
I don’t have my QU yet, hopefully next week!
Im planing to play white noise through a speaker direct into a condenser mic and use the RTA on the mic channel to determine the response of the particular mic I am using. Then I should be able to place the mic in the center of the room, studio, venue, whatever, and determine if there are nodes and standing frequencies that should be taken out on the master EQ.
Use pink. But, consider that broadband noise exciting a room can be deceiving. I have seen many situations where the RTA looked fine but there were still lots of problems. I use a custom CD with 31 individual pink noise frequencies. Works much better. I use an spl meter. Set reference level then write down db reading of each compared to ref. Adjust GEQ accordingly. Works much better.
hmm, wouldn’t this be a nice feature request: adding an (optional) steep bandpass filter to the noise generator with the frequencies of the GEQ along with next/prev selection when active in GEQ mode. …no need for the CD…
Anyway: Clever idea GCumbee, thanks!
Thanks. I can’t claim original credit. It was developed by a couple of Nashville engineer friends of mine. Dbx actually has a similar thing on a test disc they put out years ago. I like mine better. Longer recordings. Seems to be filtered better. I have ran both and had better luck with the custom one. I can tell you. It generates a much better sounding system result than full broadband pink noise.
Back in the 70’s UREI had a product called Sonopulse. It was basically the same idea but used pulsed tones. Idea was the tone was pulsed and then with a calibrated microphone feeding an spl meter the signal peaked and held. Pulsing tones is better especially in the LF range so as not to create standing waves. It was a great room tuning tool.
Nice, thanks guys. Makes sense about the pulses for getting really detailed.
Would definitely be a nice upgrade to the noise generator. Can you not send the Noise generator through an EQ or is it somewhere in the signal path that doesn’t have that capability? (Sorry, mine is in the mail so I cant try it out yet)
Have you guys found it crucial to use an omni pattern? I only have cardioid mics at the moment. Was hoping to not have to buy another mic, but gear shopping sure is fun…
I don’t exactly understand how I would do it. Yes, I would play pink noise through speakers, sound would go through microphone, I would hit PFL, because RTA follows PFL. But than I don’t understand quite clearly how things would work. Which frequencies would you cut/boost on GEQ? You would cut frequencies that appear most (have highest values) on RTA? Am I correct? Which frequencies would you boost, if for example one type of room ‘eats/absorbs’ some frequencies a lot. I also find RTA a bit hard to work with, those frequencies, and red colour change so fast, you must really look carefully to see those frequencies.
For best results, you want to use a measurement microphone (also called reference or RTA microphone). Your goal is to get a curve as flat as possible (a straight line) on the RTA when using pink noise. As you raise the volume, it will feed and it will show on the RTA which frequency it is (It will stay red on one band). Cut that frequency. I like to start with a 10db cut and 5db on the previous and following frequency band to get a smoother curve. General rule for the EQ is to cut and avoid boosting. It will most likely feed at more than on frequency band so rinse and repeat.
Isn’t a term ‘room tuning’ a bit incorrect? When you are making this whole proccess, you are not actually tuning a room, you are tuning a sound system in that room. Is that correct? So can I for example tune my system at home, and then use my GEQ setting in all types of room. I know system behaves differently in different rooms, but I don’t always have time to make ‘room or system tuning’ proccess. I never done tuning before, but I somehow managed to find some frequencies that are tending to cause feedback (frequencies in middle section, 1kHz, and some highest frequencies). And I almost never have problem with feedback.
You are correct. Tuning a room would actually be treating the room acoustically (bass traps, acoustic panels, etc…). Just avoiding feedback is ok when you don’t have time. But you also want the room to sound as flat as possible to help you shape your sound the way you really want it when mixing. This is where the pink noise is useful.
You can essentially do both. Tune your system to achieve as flat an output as possible, then in different rooms and venues tweak output further more to account for any frequencies that ring out louder than others.
I agree: “General rule for the EQ is to cut and avoid boosting”
If you’re dealing with your own system, best practice is to start with it tuned using DSP to get the response you want right out of the gate, then deal with system/room interaction with the mains L/R PEQ. With your speaker system optimized prior to setup and effective speaker positioning the PEQ should offer enough filters to deal with room modes.
I find the concept of using RTA and pink noise interesting, but prefer to do my tuning after ring-out by ear using a well-recorded ensemble representative of the music that will be on stage…making allowance for the typical changes once the audience arrives.
If for some reason the four PEQ filters on the main bus should not be enough I place a Sabine GraphiQ between the mixer and the amps or in a pinch route signal to the mains via Mix 9/10 and hard-patching back via ST1 or 2 so as to be able to use four or eight more PEQ filters.
I seldom find I need more than the four filters available on the mains bus, though.
When you “ring out” a room you are addressing room modes/standing waves which by their nature have very specific, fixed frequencies which are best dealt with using notch filters rather than GEQ filters which, although they are on 1/3 octave centers, pull down frequencies in a cut as wide as an octave thus removing or affecting program frequencies far in excess of those needed to address room/system interaction.
GEQ is then used for general tonal adjustments after the specific sonic surgery dealing with room modes has been properly addressed using the variable width PEQ filters. The narrowest filters available on the Qu desks are 1/9 octave. With my GraphiQ the filters are as narrow as 1/60 octave.
Side note: the greatest area of concern for room modes/standing waves will be found below 300hz…as a general rule. Above that there will be the possibility of anomalies between the measurement mic and the speaker system which will tend to disappear by simply moving the microphone. Room modes won’t disappear…
Are there any frequencies that more often tend to cause feedback than others? Which frequencies would that be? I noticed that maybe those: 1kHz, 6,3kHz, 1,25kHz, 250 Hz… ? Your experiences?
System interaction (mic/speakers) is situational and any feedback loops will depend primarily on physical distances (wavelengths/phase).
If you are working in rooms of the same dimensions you may find similar room modes. If your system setup always has the same physical relationship between the mics and speakers the same might be true…discounting reflections.
Thus any seeming consistent phenomenon would not be one of frequency itself but one of the physical similarities in setup and location manifesting in “hot spots”. I suspect that gross similarities in setup leads us to look to certain frequencies as “trouble spots”, but it is not solely the frequencies themselves that are the problem. Rather, they are a manifestation of problematic positioning and mic/speaker interaction.
There may be certain adjustments you’d need to make in your system DSP to address any consistent problems such as cross-over points/slopes and such. Speaker manufacturers often give “tunings” for their boxes which can either be implemented with digital system processing or by use of a proprietary “black box” processor.
Lastly, certain frequencies tend to stick in your mind from experience. I notice you mention 6.3K in your list. This is what I call the “ice pick” due to the pain produced when it goes off, so yes, I generally am careful about that particular one…not because it appears more often than others but because if it does appear IT HURTS!
Are there any frequencies that more often tend to cause feedback than others? Which frequencies would that be? I noticed that maybe those: 1kHz, 6,3kHz, 1,25kHz, 250 Hz… ? Your experiences?
There is some math with 250Hz and 1KHz: If 250 is a problem, check 500, then 1K. In this case 250 is the Fundamental Frequency and its upper harmonics may be causing problems too. 250 X 2 = 500 x 2 = 1K (Times 2). Also good to check 250 x 3 = 750 to see if theres any problem there.
DO NOT blindly follow this suggestion!! Using math is just a place to start, but it can help you get things rung out quickly with less trial and errors.
