Fuzz measure Room analysis & Anechoic Snare

So my journey begins with two parts to my project, first I had to get the room’s frequency response of two rooms using Fuzz Measure, with that knowledge add a reverb a equalizer to an anechoic snare to mimic that rooms acoustics. Analysing a room can be really helpful for knowing the room’s suitability for recording and mixing.

I analysed the large empty room next to the lecture theatre and the small drum booth in studio 1. To actually analyse the room Fuzz measure makes a sweep noise from 0Hz to 20Khz in a matter of seconds providing the sweep settings are set to long 10s sweep, 0dB, full range. I armed myself with a laptop to use FM, a speaker to play the signal from FM and a microphone to receive the FM’s signal and an Mbox. And so my journey begins in the large empty room.

Untitled Untitled1 Untitled3

So how this works is pretty straight forward, Fuzz measure plays the signal out of channel 1 on the interface, this signal gets sent to the speaker. I set the speaker at a volume that excites the room without actually distorting the signal. The reason you don’t want to distort the signal is simply to avoid playing any other waveform other than a sine wave which is just one frequency on it’s own without any harmonics. So now the mic picks up the signal and in turn the reverberations/acoustics of the room. I did this by making sure it was placed far away from the speaker to avoid just picking up the speaker. I did the same thing for the drum booth.

Photo taken by Liam F…happy now?

Untitled4

 

Finding the Standing Waves of a room

Large Empty room

Axis Distance Wavelength Fundamental Frequency (Hz)
Height 2.6m 5.1m 67.4Hz
Width 6m 12m 28.6Hz
Length 5.9m 11.8m 29Hz

 

The formula:

Calculating wavelength = Room Measurements x 2

Fundamental Frequency = Speed of sound (344 m/s) ÷ wavelength

Now to find the harmonic frequencies using the fundamental frequency, you will need to calculate the first eight harmonics.

The Formula:

Fundamental Frequency x the number of the harmonic

Now do this with the height, length and width to find the standing waves.

Height

Harmonic Modes Fundamental Frequency Multiply by Mode Freq Hz
Fundamental 1st harmonic 67.4Hz Fundamental x 1 67.4Hz
2nd harmonic (octave) 67.4Hz Fundamental x 2 134.4Hz
3rd harmonic 67.4Hz Fundamental x 3 202.2Hz
4th harmonic 67.4Hz Fundamental x 4 269.6Hz
5th harmonic 67.4Hz Fundamental x 5 337Hz
6th harmonic 67.4Hz Fundamental x 6 404.4Hz
7th harmonic 67.4Hz Fundamental x 7 471.8Hz
8th harmonic 67.4Hz Fundamental x 8 539.2Hz

 

 

Width

Harmonic Modes Fundamental Frequency Multiply by Mode Freq Hz
Fundamental 1st harmonic 28.6Hz Fundamental x 1 28.6Hz
2nd harmonic (octave) 28.6Hz Fundamental x 2 57.2Hz
3rd harmonic 28.6Hz Fundamental x 3 85.8Hz
4th harmonic 28.6Hz Fundamental x 4 114.4Hz
5th harmonic 28.6Hz Fundamental x 5 143Hz
6th harmonic 28.6Hz Fundamental x 6 171.6Hz
7th harmonic 28.6Hz Fundamental x 7 200.2Hz
8th harmonic 28.6Hz Fundamental x 8 228.8Hz

 

Length

Harmonic Modes Fundamental Frequency Multiply by Mode Freq Hz
Fundamental 1st harmonic 29Hz Fundamental x 1 29Hz
2nd harmonic (octave) 29Hz Fundamental x 2 58Hz
3rd harmonic 29Hz Fundamental x 3 87Hz
4th harmonic 29Hz Fundamental x 4 116Hz
5th harmonic 29Hz Fundamental x 5 145Hz
6th harmonic 29Hz Fundamental x 6 174Hz
7th harmonic 29Hz Fundamental x 7 203Hz
8th harmonic 29Hz Fundamental x 8 232Hz

 

Finally make a table of all the harmonic modes and look for frequencies that repeat or are very close to each other. The repeats will be the standing waves of the room.

Table of harmonic modes for room

Harmonic Modes Height Axis Width Axis Length Axis
Fundamental = half wave 67.4Hz 28.6Hz 29Hz
2nd harmonic = full wave 134.4Hz 57.2Hz 58Hz
3rd harmonic =3 x ½ λ 202.2Hz 85.8Hz 87Hz
4th harmonic =4 x ½ λ 269.6Hz 114.4Hz 116Hz
5th harmonic =5 x ½ λ 337Hz 143Hz 145Hz
6th harmonic =6 x ½ λ 404.4Hz 171.6Hz 174Hz
7th harmonic =7 x ½ λ 471.8Hz 200.2Hz 203Hz
8th harmonic =8 x ½ λ 539.2Hz 228.8Hz 232Hz

 

Now that you know the standing waves, you can use an EQ to rid of the problem frequencies from the diagram above. So from my calculations the problem frequencies will be 25Hz – 30Hz, 55Hz – 60Hz, 80Hz – 90Hz, 110Hz – 120Hz, 140Hz – 145Hz.

The downside is that due to the low resolution the lower frequencies are not going to be accurate.

Coefficient Absorption

This is where different materials of a room will determine the reverberance or absorption of sound of objects.

Full absorption is 1 whilst full reflection is 0

 

Floor materials 125Hz 250Hz 500Hz 1KHz 2KHz
carpet 0.01 0.02 0.06 0.15 0.25

 

So from the chart above we know that depending on the frequency will be the amount that the sound reverberates or absorbs from the material. Therefore, in the large empty room the sound will reverberate most on 125Hz due to the majority of the material on the floor being carpet.

In the drum booth the floor is made of wood, so the reverberations are around 0.03 on 125Hz less compared to carpet.

Re-creating the snare with fuzz measure results

The Sabine T60 method is a way of calculating the reverb time. Fuzz measure automatically follows this method to some extent. Basically Sabine calculated the time in which sound completely fades out.

Once I collected the information I started to work on recreating the snare as though it had been played in the large room and drum booth. I did this in P-T and began by importing an audio file of my pre recorded “anechoic snare”. I then made two stereo aux tracks used bus sends to route the anechoic snare to the aux tracks. I used “Dverb” (reverb) and EQ plugins on each track then used Bluecat frequency analyser, ensuring that I follow this exact order.

Untitled5

 

I had previously screenshot all my fuzz measure results as reference, looking at my results I then played my anechoic snare with the reverb and EQ, and made the frequency spectrum follow the Bluecat frequency analyser.

With my fuzz measure decay time results I set the D-Verb to follow it. With my Drum booth I set the room in D-verb to be Room1, as it sounded more like the size to the large empty room I recorded. Set the delay time to roughly the same as the results too. Once done I EQ’d to match the results from Fuzz measure. From my results frequencies 100Hz and below were around -50 – 60dB so I added a high pass filter to cut them.

Untitled6

EQ: Boosted frequencies around 200Hz +4.7dB, 818Hz +15.7dB, 1.7KHz +15.8dB and 6KHz +7.7dB. I also cut the OUTPUT on the EQ by 7.7dB as it clips in volume.

Untitled7

 

I matched the fuzz measure analysis quite accurately

Untitled8

For the Drum booth I went through the same procedure, added a high pass for anything below 100Hz, boosted frequencies 440Hz +11dB, 1KHz +9dB, 5Khz +18dB and 10Khz +12dB.

Untitled9

 

Again the results were simular

Untitled10

Untitled11

Conclusion

I found that Snare 5 is the large empty room and snare 2 to be the drum booth as it matched the frequency analysis quite well with the Fuzz measure analysis.

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s