Newbie need confirmation

[Inigo]

Hi. everyone.
This is my first atemp using this software and want to know if I am heading in the right direction.
I am trying to measure my hifi system and interactions with my room.
The left speaker is a bit boomy and here is the first graph measured with my miniDSP UMIK-1
The file can be downloaded here
https://drive.google.com/file/d/0B9FhaHX3WvvcSmZqQjNvQnF0T2s/view?usp=sharing
I guess that my problem is the peak at 50Hz and 90Hz.
Comparing to optimal acoustics is it a huge problem? or something that can be controled easily?
Any one has any coments on other issues? I am a bit lost with room acoustics but want to understand whats going on.
Thanks for your coments.

 

[Matthew J Poes]

Hi Inigo and welcome. I won’t be able to download your file to check it further but I can comment on what I see.

It would be good if you could post how you took that measurement. What is the room layout, where was the mic, etc.

Since you say the left speaker sounds boomy I am wondering if it’s placed in a location too close to boundaries and is thus causing bass buildup. It could also be a coincidence. The left corner may have a bass buildup problem and so it sounds like it’s the left speaker but in fact any bass reproduction would cause that from either speaker.

As for the measurement I suggest applying a little smoothing to make it readable. 1/6 or 1/12 is fine.

To identify modes I use a different approach. First I would generate a waterfall and plot it from 20hz to 500hz. I would give it a 600ms time frame. I would use the same db scale you have here. This will help identify ringing.

Next I would go into the room simulator and input your rooms dimensions. Apply a very small amount of absorption to each surface to help smooth the graph a little. At low frequencies all walls are at least a little lossy. Something like .1 is fine. Place your speakers where you have them and your head/person where the mic was in the room. Look at the modes that show up and see how they align with any ringing. From there we can start to identify the sources and suggest the best solutions.

 

[John Mulcahy]

Hi Inigo, looks like a successful measurement. Although you have those peaks at the low end there isn't much ringing, so a little experimentation with speaker and listening position placement may help even things out. You can use the RTA and the Pink PN test signal to get a live view of how the response changes as you move the speaker or listening position, use the RTA settings below for that.

 

[Inigo]

I finally gather all the data.
Here I enclose the waterfall from both speakers at the same time
Also the drawing of the livingroom
And the graph from both speakers, all smooth at 1/6
I hope it helps

 

[Matthew J Poes]

I finally gather all the data.
Here I enclose the waterfall from both speakers at the same time
Also the drawing of the livingroom
And the graph from both speakers, all smooth at 1/6
I hope it helps

This is great thanks. The response is a little rough and shows that some damping added to the room would help a lot. A DSP correction device would also help quite a bit and having both would be ideal.

We see some modal ringing at 37hz and around 55hz. EQ would do wonders down that low and good bass trapping would help as well.

You don't have sidewall reflections so the response seems to be the result of other early reflections and modes. I need to check your numbers for speaker location and modal response to see if I can identify the cause of the problems. I'll try to get to this later tonight.

 

[Inigo]

thanks a lot for your conmintment

 

[Matthew J Poes]

What's your ceiling height. I'm trying to model your room to see if I can make sense of the modes, and see if any locations make sense.

Also would you be able to share the construction type of the room. Wall materials, ceiling, floor? This helps to know how much damping is in the room.

My initial model of your room (I guessed at a 3m ceiling) suggest that you may have some modes close together and some that lie on top of each other. This may be because the values I put into the model were multiples of each other. I couldn't replicate in the model the modal behavior above 150hz and my guess is because your room is not a perfect rectangle. I've also heard that traditional modeling approaches may get it wrong above about 100hz.

In any case, I think that you might consider moving the speakers as John suggested and his suggested approach is a good one. I played with locations in the model but nothing stood out. If you bring them closer to the wall it will raise the front wall boundary interference frequency and potentially collapse the soundstage. If you move them farther out it will lower that frequency and may or may not improve the soundstage. I would move them closer together if possible by about .3 meters. If you don't like it you can always move them back.

Adding multiple subwoofers will help with the bass smoothness a lot. I modeled with and without subs and found a substantial improvement doing that. The room layout prevents optimal placement of the subs, so I suggest playing with different locations. REW has a room modeling program. Because your room is not a perfect rectangle you will need to fudge the values to get the major modes to align with your measurements. As John said, you don’t have severe ringing so just smoothing the bass will make a substantial difference in removing the boominess.

Finally, adding eq for the subs and/or mains would help. You can use eq to reduce the bass. It is possible that your receiver already has eq. That may be an option to play with. If not, then buying a Minidsp device would be good.