Best way to calibrate monitors with Breakaway RTA

Hi Kreso!

I will make a demonstration video within a couple of weeks, along with screen shots, showing how to do it.

If you don’t want to wait that long, here’s how I do it:

Let’s assume you have 2 different computers — the Breakaway Live computer, and the RTA computer. (It can be the same, but you have to work out the sound card routing yourself, perhaps different sound cards.)

Set Breakaway RTA to capture audio from the Microphone.

Hold the microphone in your hand! Don’t use a stand. You will need to move it around constantly to avoid being fooled by room reflections.

In Breakaway Live, turn on Pink Noise, for just one speaker. Unless the speakers are very close together, you will need to RTA each speaker separately, one at a time!

Set RTA to 100 averages, 1/6 octave. You will need to decide on a "baseline" to aim for, where you will get flat frequency response using just a couple of eq boosts (+ gain), but mostly eq cuts (- gain). It’s generally better to cut than to boost, to avoid eating up the available headroom.

Start by identifying the tallest peaks. You want to do the job using as few bands of EQ as you can! Apply bands of EQ one at a time, flattening the curve as you go. To make sure you line the EQ up with the peak you see on the RTA, don’t be afraid to set Gain to +18.0 dB, so that you see the peak shoot way way up. After that, turn the gain down and adjust until it flattens out. Wide bandwidth EQs sound better than Narrow bandwidth EQs (cleaner impulse response), so use as wide as you can to do the job.

When you have one speaker as flat as you can, do the same with the other speaker. It will take practice — you will probably have to start over a few times until you get the hang of it.

When you’re all done, and have the response as flat as you can, the result should sound like a pair of new speakers — they should sound really open and transparent, without annoying resonances, but they will also lack bass and sound thin. That’s when you go into the Loudness Settings and apply EQ for taste, like bass boost. You can also make these boosts volume dependent, so that you boost bass only at lower volumes, etc. Remember, we’re not actually trying to make the frequency response completely flat — we’re trying to get rid of resonances flatten the response to get rid of tinniness, coloration, and make the speakers sound natural and transparent. Once you’ve achieved that goal, it’s perfectly fine to boost bass and treble to taste!

When it’s all done, pressing the Bypass Speaker EQ button should be like night-and-day.

I will have much better instructions in the future. I just typed this up impromptu. If anyone has comments / suggestions / corrections / additions, feel free! That’s why it’s a forum 😉.

///Leif

[quote author=”celar”]WOW- fantastic, I never knew you had to move the mic around while calibrating.[/quote]
There are those who disagree with me on this. My take is that if you calibrate with a stationary mic, you calibrate for ONE ear in exactly that position, and if you move two inches you need to recalibrate, because it won’t be valid. 🙂

[quote author=”celar”]
One question about your instructions: I always assumed one should use white noise to calibrate, since white noise frequencies are of equal power. Can you elaborate on the use of pink noise for this purpose?[/quote]

Certainly!

White noise is equal power PER FREQUENCY.

Pink noise is equal power PER OCTAVE.

To go up an octave, you double the frequency. Thus, each higher octave contains twice as many frequencies as the previous one.

So, White Noise measures flat on a spectrum analyzer.

Pink noise measures flat on an RTA.

White noise measures top-heavy on an RTA. Power rises 3dB per octave.

Pink noise measures bottom-heavy on a spectrum analyzer.

As you can see, a spectrum analyzer IS NOT an RTA, and vice versa.

There’s also brown noise. It measures bottom heavy, power falls 3dB per octave:

To me personally, white noise sounds like the rain, pink noise sounds like the ocean, and brown noise sounds like the niagara falls. 🙂

Best,
///Leif

Would this be a good replacement to avoid buying the Behringer mic?

http://www.vikash.info/audio/measurement_mic/

Good points you make. This has actually crossed my mind, but being an unemployed student spending so many hours on the internet does make it a bit hard to measure (no pun intended) 🙂

[quote author=”Leif”][quote author=”celar”]WOW- fantastic, I never knew you had to move the mic around while calibrating.[/quote]There are those who disagree with me on this. My take is that if you calibrate with a stationary mic, you calibrate for ONE ear in exactly that position, and if you move two inches you need to recalibrate, because it won’t be valid. 🙂[/quote]
Since I haven’t posted in this topic yet (lol) I will say that I don’t disagree with you. You’re right, but… (there’s always a butt)

Some speakers are more pronounced in this effect in the room they are in than others. For instance fully horn loaded, time aligned, single point source speakers have nearly zero boundary effect or any other bad effect caused by the outside of the box itself. Boundary effects are mostly the problem with a more conventional speaker in a space. The more boundaries you are near, the more this effect can be bad. Sometimes it can actually be a good thing too, like for subwoofers, but it has to be very carefully and intelligently setup to increase the output while only making the bass extend without screwing up the frequency response (or only making the frequency response flatter)… and yes, you will need to check for room modes etc, especially the smaller the room gets from say… the size of a stadium. (and even those can have room modes sometimes, been there, done that)