Room Acoustics: The Dreaded Chair.

[ozzy9832001]

Hi guys,

For some background on this post.

When I first started doing room measurements, I always placed my UMIK-1 on top of the back of my desk chair, because with the stand, it was exactly at the height of my ears and the tip was right at my ear from the back of the chair. It was perfect....or so I thought.

Recently, I bought a mic stand for another reason, but decided to do some measurements from it and get the chair out of the way.

I always knew the chair had an effect on the measurements, though according to Floyd Toole, we can't (or shouldn't) correct for the comb filtering it creates.

I assumed (probably incorrectly), that only the back of the chair would cause an issue and that technically the direct sound arrives before it would interact with the chair.

What I didn't realize is that, in my case, the chair has a huge effect on the frequency response.

So the attached graph is as follows:

Purple line is me holding the mic without a chair, simulating the effect of just a body.
Green line is on top of the chair.
Orange line is mic on stand, chair out of room.

The position of the mic is the exact same for all 3 measurements. What I was shocked about was the differences in the lower mid frequencies, even down to 200hz had an effect. I didn't save the measurement, but I noticed if I flipped the chair around and the seat was facing the back wall, the graph changed yet again. Which means the seat as well as possibly the armrests are also interacting with the sound.

What really, really surprised me was the effect the chair had on the spectrogram.

Chair:

Notice the spikes at 2k, 3.5k and how low the 5k+ region is.

No chair:

Everything has smoothed out quite a bit.

If I create 2 EQs, one based off the Chair and the Other without. I find the one without the chair yields a much cleaner, flatter sound and the one with the chair is much more vibrant in the mids.

The chair appears to provide some absorption in the low mids and reflections in the higher mids. I suppose that makes some sense, since the chair is probably close to 4" thick for the back and seat.

You guys can ignore the bass region on the graphs. I had disabled all filters while I ran these tests to ensure they were accurate.

My main question is:

Does the chair actually make a difference...and should be account for its presence, or as Mr. Toole suggests, not correct for it.

If we shouldn't correct for it...why is that?


Thanks everyone for your input.

 

[LTig]

With my Gravity chair the sound was worse with my head on the head rest. Now I sit on a sofa where the influence is neglilible. Anyway I don't think one should correct it.

 

[DWPress]

If you correct for the space without the chair and listen to music you will be standing around a lot, just be sure you don't move away from where that mic stand was. We correct for the chair because we like to sit, Floyd likes to sit too.

Try a multipoint measurement and create an average or use the MMM method to get an alternative idea of what's happening beyond the single point measurement - especially the latter.