Frequency Response of My Control Room

LDQ · Post by **LDQ** » Fri Oct 24, 2003 1:51 am

Hi all,

Check out the frequency response of my control room and let me know what you think. I measured the frequency response in various places in the control room. I used my Genelec 1032A's soffited in the wall. I also played around with using -6 db Bass Tilt no Bass tilt and +2 Treble Tilt. Barefoot I think I verified what you recommend me a while back on soffit mounting genelecs.

Anyhow checkout the graphs and let me know what you guys think.

Luis.

1. Sweet spot with -6dB Bass tilt
2. Sweet spot with no Bass Tilt
3. Sweet spot with -6dB Bass Tilt and +2 Treble tilt

LDQ · Post by **LDQ** » Fri Oct 24, 2003 2:07 am

1. Back of the room on Left side with -6dB Tilt. NASTY!!!
2. Back of the room on middle with -6 dB Tilt.

barefoot · Post by **barefoot** » Fri Oct 24, 2003 4:31 am

This is great information Luis!

It's always nice to have independent confirmation so folks don't get the idea that I'm talking out the wrong orifice.

Btw, your sweet spot measurements actually look very nice. ±4.5 dB isn't bad for a steady state measurement. Just remember that this is a convoluted measurement between the direct sound from the speakers and the ambient reflections. Your ear can distinguish between these two "sound fields" in the higher frequencies (wavelengths small compared to the room dimensions). And your brain gives the most weight to the early direct field response when making judgments about spectral balance. So, the fact that the +2dB treble tilt looks a little better should not lead you to believe that this is the correct setting. This adds +2dB directly to the speaker output, and therefore adds +2dB to the direct field response...... the response your brain gives the most weight to. Instead, you should conclude that your room is slightly too absorptive in the treble and that you might need to brighten it up a little.

In the low frequencies, where the wavelengths are large compared to the room dimensions, there isn't really any distinction between direct and reflected sounds. The direct and reverberant sound fields are the same. In this case it is appropriate to use your -6dB shelf to flatten out the response. Luckily Genelec has made specific EQ provisions for soffit mounting. Besides my own speakers, they are the only nearfield monitors I know of that do.

Thanks!

Thomas

LDQ · Post by **LDQ** » Fri Oct 24, 2003 5:28 am

John, Barefoot, Ethan...or anybody else is there anything I can do to improve the response a little on the highend 4K to 8K and the low mid 200 to 500 Hz. I wonder if my splay walls slot resonators are working correctly they were designed to absorve in the lowmid range.

For detail on my control room construction check Silva Productions studio at the Studio under contruction site.

Regards,

Luis.

barefoot · Post by **barefoot** » Fri Oct 24, 2003 6:42 am

Since you have a nice flat response region for at least 1.5 octaves below that problem area, my guess is the problem truly starts in that frequency range. These don't look like ripples from a lower frequency issue. This means the problem is fairly close the axis between the speaker and the listening position. You’ve have a dip at 4kHz, so it's reasonable to guess the problem is related to path length difference of a whole or half wavelength at that frequency (3.4" or 1.7"). These are pretty short distances. The only suspects I really see from the pictures are the soffit box edges, the ceiling, and the near side walls. You can try systematically covering the nearby edges and surfaces with 703 to see if it helps.

If not, then you should also investigate these questions:

Is the microphone flat?
Are there any obstructions near the microphone?
Is the monitor truly flat? (I can explain how to measure its reponse).

Thomas

Post by **John Sayers** » Fri Oct 24, 2003 7:48 am

That doesn't look too bad Luis - remember genelec measure the speaker from 1 meter probably in an anechoic chamber and they claim +/- 2.5 db.

Have you tried using the bass roll off instead of the Bass tilt?? The low mids look better with no tilt. I'd try the roll off instead of the tilt. I remember when setting up the Genelecs at Left Bank it was a matter of playing with both. Pity you didn't get the amp extension so you can have the amp mounted below the speaker with easy access to the controls.

BTW - I rechecked your pics - have you continued the speaker baffle below the speaker??

i agree with Thomas about checking the mic response etc.

cheers
john

LDQ · Post by **LDQ** » Sat Oct 25, 2003 3:01 am

Have you tried using the bass roll off instead of the Bass tilt??

I did last night and below are the results. Also I placed a couch in the back of the room and the -6 bass tilt response changed somewhat.

BTW - I rechecked your pics - have you continued the speaker baffle below the speaker??

Yes plywood on the back of the lower front frame insulation on the front and cloth Plus 1 foot opening at the bottom. I have new pics comming soon.

Below are the following graphs:

1. -6 dB Bass Tilt + couch on back of the room
2. -6 dB Bass Tilt and -2 Bass Rolloff
3. -4 dB Bass Tild and -8 Bass Rolloff

I think I pick #3 since most of the response is withing the +-2.5 dB. I am not sure how to get rid of the 8K dip as well as the 150Hz peak.
Any thoughts?

Luis.

1. -4 Bass Tilt and -8 rolloff

Post by **John Sayers** » Sat Oct 25, 2003 3:29 am

Luis - that 8k factor isn't consistent - it varies with where you are putting the mic. What does it sound like - that's what is important. Use your ears - they are more accurate

cheers
john

barefoot · Post by **barefoot** » Sat Oct 25, 2003 4:00 am

I used this very handy utility to pull the data from your pics. SPL Tools. I calculated the standard deviation from the mean for each curve.

1. Stdev = ±3.00 dB
2. Stdev = ±2.71 dB
3. Stdev = ±2.77 dB

So really, there's little difference in the flatness of the 3 curves. Curve #2 is marginally flatter.

John's right that the 8kHz dip is positionally dependent. Still, you do consistently have something going on in the octave centered at 6kHz. Like I said, this is being caused by something close to the listening axis (the imaginary line between the speaker and the listening position). Do you have the desk in place? If so, this is almost surely the culprit. Also, are you just measuring one speaker? Don't try to measure both speakers in mono. This will cause comb filtering. Only measure one.

Thomas

Post by **John Sayers** » Sat Oct 25, 2003 11:10 am

So it's the pole on your right Luis - that would also tend to move the image to the right as you mentioned. Have you wrapped it in insulation like we discussed as that will stop anything in the 6KHz region Thomas mentioned.

cheers
John

LDQ · Post by **LDQ** » Mon Oct 27, 2003 12:17 pm

Have you wrapped it in insulation like we discussed

I framed it, filled it up with 703 and covered with sheetrock. and put my RPG Foam in the front I haven't put insulation on the outer part.

Luis.