[tony359]

Hello everybody

I have converted one of my bedrooms in a very small home theatre. The room is only 8sqm unfortunately. About 2.7x2.9m. Unfortunately this is what I can get!

My previous home theater was large and set-up with proper cinema equipment; this one is too small and it's really not worth a big investment. However, I am still trying to get the most out of it. This time in a more scientific way. On my previous attempt I blindly covered the walls with very dense and thick rockwool, causing the sound to become very dull.

In the meantime I've gained some extra knowledge - but lost the big space unfortunately! What I am aiming to do in this room is to control the sound without killing it. I have taken a first measurement and because this is my first attempt with REW and because the room is very small, I would like some advice to get started, please.

I have had a look at the step-by-step guide and come up with a good reading. What I am trying to identify at the moment is the obvious ringing caused by the drywall room. It is not too severe but definitely audible. It's an high frequency ringing, I am trying to identify it on the graphs so I can identify the frequency and dampen it with small frames filled up of rockwool or with sound diffusers.

I understand the waterfall graph is meaningless over 300Hz. What should I look for to try and identify such issue?

Also the room seems to have an RT of about 0.3s which I do not want to lower unless it is necessary. How would you define what the proper RT is for such a small room?

I appreciate some of you may say "why wasting time on this". It's my hobby. I am also a cinema installer - the big ones - and cinema is my big hobby as well!

I'm posting a few graphs, I hope I have set the parameters correctly. Given the small size of the room I am not sure. Please note this is the centre channel only.

I am open to any advice. Thank you for your help on this!

 

[JohnM]

The high frequency ringing is probably slap echo. Doesn't matter too much what the exact frequency is, diffusion or light absorption on the opposing surfaces should get rid of it. Might show up on the wavelet spectrogram.

 

[tony359]

Thanks John

I can build some small diffusers and I had already in mind to make small 'picture frames' covered with fabric and filled up with high density rockwool. I was hoping to be able to identify the issue on a graph so I knew when the diffusion/absorption was enough. I guess I can keep an eye on the reverberation time and make sure it does not drop too much.

Attached the spectrogram, is there anything you see that may help?

 

[JohnM]

Not really, no

Your ears will tell you though. If you want to avoid things being too dry then favour diffusion over absorption.

 

[AudiocRaver]

Your small room presents interesting challenges. The 0.3 sec RT60 is probably as low as you want it to go, so - as John suggested - a mix of diffusion and absorption that tames annoying reflections without drying the sound too much is a good approach.

I would spend more analysis time looking at impulse response than frequency response, just as you seem to be doing. Let us know what you come up with.

 

[tony359]

Hi AudiocRaver,

Sorry I had missed your reply. Yes, I did the mistake of blindly dampening a bigger room in the past making it dead (0.1-0.2ms RT) so that's why this time I'm using a different approach.

Before putting anything up on the walls I wanted to be able to see the ringing on a graph or something similar. As the readings did not really show much, I created some short bursts of sound and played them at high level. The ringing is obvious and I have also recorded it for you so I can share the actual situation with you.

I still would like to be able to identify this ringing using some sound tool but failing that I guess I can play with diffusers/absorbers till the ringing is gone/less obvious - keeping an eye to the reverberation time while I do that.

Here you can hear the recordings (it's my domain, it's a safe place!)
500Hz
1Khz
3Khz
Pink Noise

Probably more noticeable here on the below attachment.

Your opinions are all welcome!

 

[AudiocRaver]

Interesting, but you can tell more from the impulse response diagram that REW generates. It is calculated from the swept analysis signal and is highly accurate.

 

[tony359]

thanks. I believe the graph is just a few posts up the thread, could you give me a hint on how to read it? I understand that I should look for spikes which are coming up over -20dB.

I am not on my home computer now, so I had to take a screenshot of the graph above - apologies for the low resolution. Am I highlighting the spikes correctly?

Thanks
Tony

 

[Joe Bloggs]

I think you may look at where the echoes are happening in your *room* rather than "where" in these graphs.

If you do not want to reduce the reverb time of the room, you just need to treat the direct reflection areas, even total absorption just at those some areas will not change total reverb much. Direct reflection means treating your sound waves as though they were light beams--imagine holding a mirror against the wall, the floor or the ceiling and moving it around until you see a loudspeaker reflected in the mirror when you look from your listening position. Those are the direct reflection points that should be covered with treatment.

Then, too, our ears are very sensitive to frequency response while room treatment does not directly address this. For example you will hear the effects of even those direct reflections not as fistinct echoes but as coloration of the sound due to comb filtering of different frequencies (ie given an echo at a fixed delay, some frequency wavelengths are reinforced (boosted) by this delayed copy while others are cancelled (attenuated). Treatment will reduce the sharpness of the comb filtering but not eliminate their overall effect on frequency response. You didn't post any frequency response plots but I noted from your spectrogram hotspots at around 80Hz and the 150-200Hz range. In my room I actually applied zero treatment but fully EQed out every null and peak especially in the bass response and the result has been a tremendous reduction in boominess, increase in clarity while retaining the lively reverb of an ordinary room. I know this goes totally counter to all room treatment wisdom but think it's an indication that frequency response at least shouldn't be totally ignored and there really ought to be at least some form of parametric EQ in the system.

 

[tony359]

Sure, I appreciate that stopping first reflections is the way to go. And it is indeed what I am planning to do. However I was looking to try and identify the issues scientifically before placing panels around the room. I would like to have a tool to identify what the issues are so I can keep an eye at it while I am installing absorbers/diffusers. Now I have the reverberation time, I'll keep an eye on it so I know I am not lowering it too much.

Frequency response is attached but I have two DCX2496 on the line and there is a rough EQ on the speaker already. When the acoustic is more controlled I'll be happy to spend some more time on it. In fact I am planning to use Dirac in the future. The attached is not supposed to be any good, I'm aware of that I just smoothed the raw response.

Don't worry, I understand the importance of sound EQ. I just want to correct the acoustic a little first as - as you may have heard - so far the room sounds like a bathroom would!

 

[AudiocRaver]

That's the spirit!

Those higher points in the diagram represent reflections that can be disruptive to soundstage and imaging (SS&I).

Here is one formula that will help:


(speed of sound) x (amount of delay) = length of longer sound path


1.126 ft/msec x 3 ms = 3.378 ft

A 3 ms delay for a reflection in your diagram equates to a 3.4 ft longer path, or reflection path, longer than the direct from speaker to LP. The trick is finding that reflected path, and a mirror and laser (laser distance measurer) can help you do it. Early Reflections will be the result of a single reflection, longer reflection paths can involve multiple reflection points and are less of a problem by their nature. In your case, the first 3 or 4 of those higher points are the ones to pay attention to.

 

[Joe Bloggs]

I just want to correct the acoustic a little first as - as you may have heard - so far the room sounds like a bathroom would!

Hmm...

A bathroom is a small, reverberant space.

Your room is a small, reverberant space.

If you want to make them sound different... you can't make your room larger but you can make it not so reverberant.

Frankly, an over-treated room is a first world luxury I haven't had the luxury of experiencing... :laugh2:

What I heard in your recordings was highly reverberant treble and it didn't seem to matter whether it was early or late reflections (if anything I would say it is the long tail that is contributing most to the impression)

You would be treating the first reflection points anyway, you don't need a graph to tell them where they are, a helper moving a mirror around in the room would serve much better.

Now, if you've treated those points and the sound impression doesn't change much, it may be time to treat the rest of the room as well :R

That said, can you upload an mdat for us to look at? I would like to look more closely at the spectrogram with different settings to see whether my theory of reverberant treble holds up in measurements... :smile: