Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Hope209]

How do you measure the frequency reponse of your room from your mix position?

[Brendan Thompson]

Wouldn't this question be better suited to the acoustics forum?

[franman]

We primarily use the TEF package from Goldline. It's a hardware/software package that does Time/Energy/Frequency testing. We use a calibrated test microphone (from Earthworks or B&K) and do sine wave sweeps which the TEF uses to gather impulse and frequency response data. There is, as with all systems, a trade off of frequency resolution vs: how much of the "room" you include in your measurements...

This subject could take an entire forum on it's own.. Come on, team, let's have some replies and get it rolling...

[Ethan Winer]

Fran,

> Come on, team, let's have some replies and get it rolling... <

ROFL. Okay, here's how I do it:

http://www.realtraps.com/art_etf.htm

--Ethan

[jfrigo]

franman wrote on Tue, 24 October 2006 16:52

We primarily use the TEF package from Goldline. It's a hardware/software package that does Time/Energy/Frequency testing. We use a calibrated test microphone (from Earthworks or B&K) (SNIP) This subject could take an entire forum on it's own.. Come on, team, let's have some replies and get it rolling...

ACO Pacific is the other company to look to for calibrated test mics. If you're really serious about it, the mic goes back to them once a year for calibration. They'll send it back with a nice printout of the response, and a sticker on it that says "due again xx/xx/xx," kind of like an oil change for the ol' Family Truckster! The important thing that TEF brings to the party over just a simple RTA is the time element. Knowing how things decay (the ringing Ethan rightly warns against) is certainly as important as the peaks and valleys in the response. If one of those peaks rings 5 times as long as the average decay over the rest of the spectrum, it's gonna cause trouble.

[franman]

To elaborate a litte more on the time issue, most frequency anomolies in the mid and high frequencies that are not actual loudspeaker response issues, are all "time" issues... results of early reflections causing comb filtering. This can easily be verified by overlaying your tests and moving the microphone by a few inches.. Reflection caused comb filter will shift in frequency (the dips and peaks) as you are changing the time relationship between the reflected (delayed) sound and the direct sound thus changing the comb filter... make sense?

[jimmyjazz]

True, and the reason why is because the modal distribution is so dense in the middle frequencies and above.  Acoustic anomalies aren't typically modal except in the bass regime, unless we're talking about a small closet!

[Hope209]

Well, I've moved EVERYTHING around my room and changed positions of my realtraps a dozen times...and I finally moved everything back to where it was before. I don't know what it was, but on the other side of the room there was this null we just couldn't get rid of and I couldn't hear much below 200hz.

So I ran a quick test right now, could you guys tell me if it's a valid way of testing my monitoring position?

1. I tossed up an SM81 and put it on a boom so it's capsule is right where my left ear is(the flat response was selected on the mic). It then ran through my  Vintech pre

2. I ran the pink noise .wav file Ethan suggested on one track in Pro Tools and put Waves' frequency analyzer as one of its inserts

3. I recorded the pink noise playing through my NS-10s and sub and recorded it to a new track in Pro Tools. After recording I put another Waves' frequency analyzer on its track.

4. I visually looked at the responses and compared them.

Picture 1 is the pink noise:
http://img.photobucket.com/albums/v347/Wrx07/Picture1.jpg

Picture 2 is what the SM81 captured:
http://img.photobucket.com/albums/v347/Wrx07/Picture2.jpg

Thanks for all the continous help everyone and Happy Halloween!

[Ethan Winer]

Hope209 wrote on Tue, 31 October 2006 14:04

could you guys tell me if it's a valid way of testing my monitoring position?

It depends how you define valid. I'm suspicious that the displayed response at the lowest frequencies is not at a high enough resolution to give a true picture.

I use the ETF software which resolves to 0.7 Hz, and there are tones on the RealTraps site at 1 Hz intervals you can burn to a CD to measure at a similar resolution. That's the sort of resolution you really need. But more important - how does it sound?

You'll never get it perfectly flat, and at some point you should avoid getting too hung up in single point (in the room) response graphs. Not that confirmation of your room's response isn't useful! But the only way to improve things further is via placements and/or adding more traps. If you've already exhausted every possible listener and speaker and bass trap placement, then either 1) get more bass traps, or 2) accept the results. Which brings us back to, "How does it sound?"

--Ethan

[ruffrecords]

Well I am no expert but my first reaction is that the second graph clearly shows the action of the sub and it seems its level is set too high. It might be worth repeating the test without the sub.

Ian

[franman]

Yeah the analyzer resolution is just not there down below 250Hz.. you're probably seeing 1/2  - 1 octave resolution... not really helpful in evaluating below 250Hz..

[L_Tofastrud]

I suspect the SM81 mic is a unipolar or cardioid mic.  I would recommend trying with a REAL omni mic (like a B&K or Eartworks) - switchable mics aren't really omni.  Cardioids will have a "flat" frequency response only at one given distance from the source - if you go further away the bess will roll-off and if you get closer you get an increasing low-frequency response.  This together with the fact that an cardioid mic is listening forward and not to the whole room will influence your measurements.

Regards
Lars T

[thestudio]

What equipment to get good results in the 300hz and below range?

[franman]

... see the beginning of this thread for equipment we use as well as Ethan's suggestions...