Home Studio Acoustics (DIY Pics & Asking for Feedback)

[rhuobhe]

Hello,

After reading as much as I can in this forum, I have done some basic panel installation in my home studio. So what I'd like to do is, first share my panel build method in case it's useful for somebody else and then give some specific info about my room and ask for feedback from some of the experienced users here. This should be a bit long so I'm gonna split up my posts and try to keep it less cluttered.

Panel builds first. Materials used: 40 kg/m3 rockwool, some wood strips, standard breathable fabric, nails, staple, hooks & chains (only needed for ceiling clouds)

It is pretty simple (and cost effective) for the most part so just gonna get to the pics. Sidewall panels first:

00.jpg

0.jpg

1.jpg

2.jpg

It doesn't have to be too sturdy, the cloth will hold it together nicely unless you go too hard on them.

3.jpg

5.jpg

6.jpg

7.jpg

8.jpg

9.jpg

10.jpg

11.jpg

The installation is pretty simple: Two horizontal nails on your wall spaced like 30-40 cm apart and hang the panel up using the wooden frame behind like you would hang a painting.

On to the next post.

[rhuobhe]

Same design for the ceiling clouds except you attach hooks on the wooden frame and on your ceiling (used 4 each):

12.JPG

13.jpg

14.jpg

Nevermind the weird lines on the wall, just glue residual

15.jpg

So as you can see the ropes looked terrible so I've bought some chains. Haven't installed them yet though.

For corner traps, you install wooden frames on the wall:

c1.jpg

Again nevermind the glue residue

c2.jpg

Carve the rockwool a bit with a sharp knife so that it's a bit more compact

c3.jpg

Hooks on the wooden frame, rockwool stacked inside and some rope to hold things together

c4.jpg

c5.jpg

Then stretch the cloth as much as you can and use paper pins (English?) to seal it up.

c6.jpg

c7.jpg

The final look is something like this:

c8.jpg

[rhuobhe]

And now some info about the project and my questions.

I'm an electronic music producer and this is for my home studio. I have Mackie HR 824 MKII speakers mounted on DIY speaker stands (PVC filled with sand) and they are connected to a Focusrite Scarlett 2i2 audio interface. My measurement mic is Behringer ECM8000. As I've said before I've used 40 kg/m3 rockwool and the corner traps are NOT superchunks.

Sketchup pics (Excuse the clumsy models, tried it first time today):

room1.jpg

room2.jpg

room3.jpg

I hope 2nd and 3rd images make sense because I have some weird shapes in the room like a horizontal beam on the left wall and some sort of decorative plaster molding thing all around the ceiling (Carton Pierre? No idea how it's called in English but it looks terrible. Can't remove it at the moment though). Both the window and the door are strangely asymmetrical in the original room.

After treatment:

room4.jpg

REW readings. Untreated measurements first:

untreated_left.jpg

untreated_right.jpg

And final graphs:

final_left.jpg

final_right.jpg

I'm not sure if these waterfall graphs are sufficient so please let me know and I can post up more graphs or anything really. Note: There was a sofa in front of the rightwall in the untreated room and it had to go out. The dip in the left speaker got pretty big after that and stayed that way :S

Overall, it seems to me that I've reduced the decay times significantly (perhaps too much?) but I wasn't able to change the low end response too much. Perhaps it's my room modes suck but I still haven't figured out how ratios work so I wouldn't know. I think I'm gonna hang the next panel on the rearwall door.

Questions:

1-How did I do? What can I do to improve things?
2-It looks like I will need more bass trapping?
3-Would superchunks have done a big difference?
4-Should I cover my corner traps with plastic to compensate for the mid/high dampening?

Sorry for the mega post. Would appreciate some feedback, thanks in advance.

[Soundman2020]

Hi "rhuobhe", and welcome!

You have some pretty major modal issues in that room! But it's hard to say exactly what is going on from just those waterfall graphs. Maybe you could post the actual MDAT file somewhere that we can download it, to analyze it more completely?

Also, please post the actual SKP file, so we can take a closer look.

How was the mic set up in the room for those measurements? In other words, what was the distance of the mic form the front wall, side walls, and height above the floor?. Even more important, was the ECM8000 in EXACTLY the same place for all measurements? If not, then the two sets of graphs cannot be compared. It is important the the tip of the mic must be in the exact same position in the room, in all three dimensions, for all readings. It needs to be accurate to within a few millimeters. You have to get it to the same location every time you take readings.

Also, removing the sofa clearly changed the acoustic response of the room.

The graphs have obviously changed between the two sets of readings, but we don't know WHY they changed: it could be due to the sofa being removed, or it could be due to the mic being in a different location, or it could be due to the treatment you installed. There's no way to know, unless you repeat the tests with and without the treatment.

But before you do that, I would suggest that you should first get your speakers, chair and desk set up correctly. Right now it looks like there are several problems with the way it is set up, but it is hard to be sure dorm the photos.

Anyway if you can upload the MDAT and SKP files, then we can take a closer look and try to see what might be wrong.


- Stuart -

[rhuobhe]

Hello Stuart, thanks a lot for your response. Here are the 2 files you've asked:

https://dl.dropbox.com/u/59897139/room2.skp
https://dl.dropbox.com/u/59897139/rewstuff.mdat

The mic was in the mixing position (100 cm to frontwall, 143 cm to sidewalls, 146 cm to ceiling) I marked it's location & tried to keep it in the same spot for all measurements, it shouldn't have moved more than a few millimeters. In the mdat file, I've labeled all the steps I've measured (like sideclouds first, ceilings after that etc.) so perhaps that should be of some help. But now I see the problem was taking the sofa out after doing the sidewalls & clouds :S I can try take off all the panels and start over a fresh measurement if it's needed.

I'm not sure what could be wrong about my speaker etc. setup (trying to keep it symmetric in the front side of the room, avoiding to sit in the 50% spot etc.) but I'm posting another picture (pre-installment) to give a better view hopefully. Table and speaker stands are in the same spot right now.

frontwall2.jpg

[Soundman2020]

I'm not sure what could be wrong about my speaker etc. setup (trying to keep it symmetric in the front side of the room, avoiding to sit in the 50% spot etc.) but I'm posting another picture (pre-installment) to give a better view hopefully. Table and speaker stands are in the same spot right now.

I would push your speakers right up tight against the front wall, leaving a space of just 10cm for treatment. Also ensure that they are both at the same distance from the side walls (in other words, distance from center of left speaker to left wall is the same as distance of center of right speaker from right wall). Right now your speakers are about 54 cm from the side walls (to center). I would move them a bit further away (closer to each other). Try setting them about 70 cm from the side walls (about 145 apart from each other). They are angled t0o tightly right now: Angle them inwards at exactly 30° (they seem to be about 45° right now). Set up your chair so that your ears are about 132 cm from the front wall, and your nose is on the center line of the room. Also, they are too low: raise them until the acoustic axis is 1.2 m above the floor. If you do all that correctly, then you will have the correct speaker geometry, and the acoustic axes will intersect about 25 cm behind your head. Now also adjust the "Acoustic Space" switch on the rear of each speaker to position "B" (Half Space), "Low Freq" to 37Hz, and "High Freq" to 0 dB.

143 cm to sidewalls, 146 cm to ceiling)

It looks like you are going to need to convert your ceiling cloud into a hard backed, angled cloud. Since those two dimensions are practically the same, your head is not in a good position in the room. But a hard-backed cloud can help fix that. So modify your cloud for a hard-backed design, move it a bit further back, and angle it (lower over the speakers, higher over your head).

On your MDAT file there is some evidence of earl early sound, so you need to make the speaker stands a lot more massive (maybe fill them with sand), and also decouple the speakers from the stand with Sorbathane pads.

The MDAT is excellent, by the way! Very nicely done. It looks like you really did get the mic in the exact same position every time!

The good news is that the treatment you have installed so far is working! You are clearly damping the modes and smoothing the response in general. The bad news is that you don't have enough bass trapping yet: you will need more. LOTS more...

The other good news is that you have a pretty good ITDG, although there are some reflections in the first couple of ms, so you might need more or deeper absorption on your first reflection points. There are also reflections which I suspect might be coming from your desk surface, but raising the speakers to the correct height, correcting the angle, and pushing them up to the front wall (as outlined above) should all help.


By the way, you seem to have some type of electrical problem in your room! There's a very clear 100 Hz hum issue... Check your grounding. The ECM8000 is very sensitive to hum.

I made the changes to the way the speakers should be setup in your SKP file: you can download it here:

http://digistar.cl/sayers-forum/room-thing-S01.skp

Your MDAT file also shows some major, big-time modal issues: the worst one is at 53 Hz, and that corresponds to your 1,0,0 axial mode, which is the one between the front and rear walls. So you need more trapping on those walls. I would put more insulation in the corners, behind your existing diagonal panels, and also do superchunks across the front of the room, above the window, and across the rear of the room, above the door.

Next, you have a humongous dip at around 90 Hz, which is harder to pin down: It might be a tangential mode, but there's nothing obvious on the modal response prediction.

And finally, you have another big peak at 120 Hz, and that is clearly your 0,2,0 mode, which is one running sideways across the width of your room. So you need superchunks up at the tops of your side walls as well, in the wall/ceiling corners. Big ones.

And one more: there's a reflection coming off your door that needs treating. It comes in at about 14 ms, which is too soon. Some thick absorption on the door should help with that.

So try out all the above, then measure again and see how that works out.


- Stuart -

[rhuobhe]

Hello again Stuart,

That's quite a homework Really appreciate it though. Got a little update and maybe some questions.

So it turns out I've been using a wrong geometry for years. After reading your post and checking out the sketchup you've posted, I finally "got it" so I had to change my setup quite a bit (the desk had to go) but eventually I think I almost got it right. I've only failed at getting the speakers 145 cm apart, they are around 155 cm at the moment. From center points of the speakers of course. I might push them a bit more but then the screens start getting in front of them. It looks like it would be ok but here's a quick pic:

IMG_0774.jpg

Couldn't find much info about hard-backed clouds but I believe one of your posts said 20mm-ish thick plywood etc. would be ok. So I've got 22mm's of OSB atm and will start modifying the clouds tomorrow.

About speaker stands, they are already filled with sand so I'm not sure how I can make them more massive. Because of my top notch construction skills they kind of sway slightly when pulled or something so that might be the cause? I'm not sure, but perhaps I can build them from scratch after being done with the treatment work. About their height though, you've said 1.2m for the acoustic axis and the ear height but would it be ok if both would be at 1.1m? I intentionally made them a bit shorter back in the day because I like to sit somewhat lower (probably a bad idea but )

Also it looks like it won't be easy to find Sorbathane pads in Turkey so could there be any alternatives to that?

After some bookkeeping, I think doing superchunks in the ceiling corners would be stretching my budget a bit too much so I'm gonna have to stick with diagonal traps across the ceiling but do the superchunks in the vertical corners. For starters at least. Besides it's less construction work so that's a bonus too. I might still decide to convert them into superchunks at the last moment but will measure first and see, I guess. I hope it's not a bad idea.

Anyways, this is pretty much it so far. Will post more when I get more work done.

[rhuobhe]

Hi,

Back with more stuff. Here's the brand new REW file. Made from scratch with new geometry & panels.

https://dl.dropbox.com/u/59897139/r2.mdat

I'm gonna update the sketchup file once I get more time but until then here are the current panels in the room:

-Superchunks in 4 vertical corners (They made a HUGE difference)
-One 60x120 10 cm thick hardbacked angled cloud (backed with 22 mm OSB)
-10 cm thick diagonal traps on ceiling corners
-One 60x120 10 cm thick panel on the door.
-60x120 10 cm thick panels on the sidewall first reflection points (1 on each)

Still need to do:

-Get sorbothane pads
-Fix speaker stand height and make it more massive
-Fix grounding issue

Plan to do:

-Have two 60x60 panels for frontwall first reflection points (like right behind the speakers) Somehow mount them.
-Put some leftover rockwool behind diagonal ceiling traps.

Questions:

-How does the REW file look like?
-How can I improve it further?
-Any alternatives to sorbathane?
-Is 1.1m speaker height ok if my ears are at the same height?

[Soundman2020]

-Superchunks in 4 vertical corners (They made a HUGE difference)

They sure did! And so did the other things you did. That's a really nice improvement over what you had before.

-How does the REW file look like?

Very much better! The mode at 120 Hz is pretty much gone, and the one at 53 Hz is greatly reduced, in both intensity and decay. The decay on that one was about 1.4 seconds, and it is now down to less than half that, at about 700 ms. The intensity dropped by nearly 15 dB. The null at 90 Hz is also very much smoothed out now. I'm pretty sure that was fixed by the cloud, even though I still can't really pin down exactly what was causing it. But since it moved up in frequency, to about 105 Hz, I'm convinced that the cloud is doing that, and doing it well. You could try angling the cloud a bit more or a bit less, and see if that changes it. Also, you could put some absorption on top of the cloud, above the hard back: that might help a little more.

Do your speakers have bass roll-off controls on the back? They might be marked something like "room correction" or "bass tilt" or "low frequency shelving" or something like that. If so, make sure they are set to -6 dB. Also, of there are high frequency controls on there, then set that to +2.

Did you face the superchunks with plastic? If not, then it might be an idea to do that. Not for the first reflection points or the large absorber on the rear wall, but just for the superchunks. You are getting to the point where you are starting to suck out a bit of the highs due to the bass trapping, so it would be good to keep the highs in the room by putting 6mil plastic over the fronts of the superchunks (between the cloth and the insulation).

-How can I improve it further?

1) Add the horizontal superchunks across the top of the front and rear walls, and maybe also the side walls, with plastic on the faces. Or as you already planned, put more insulation in the existing diagonal traps.
2) Tilt the cloud at a few different angles, and see if that changes the null at 103 Hz.
3) Angle the panel on the door: Tilt it forwards, so that the top is further away from the door than the bottom. For example, keep the bottom right against the door, but the top 20 cm away.
4) Do the other things you said: mass on the speaker stands, decouple, fix height, and treat front wall.

-Any alternatives to sorbathane?

EPDM and Neoprene rubber are possible alternatives. Your speakers are not very heavy, so you'll have to experiment a bit with strips of that rubber, until you get the correct deflection ("compression").

The changes you make now will not be as spectacular as what you have already achieved with the initial treatment, but will still be noticeable and useful. You made a major improvement with that first round, so each additional round will just add some more to that.

-Is 1.1m speaker height ok if my ears are at the same height?

It would be better to raise both the speakers and your head, because of the reflections you are getting off the desk surface right now. As you raise the speakers, the angles improve and the reflections will then be below your ears. So if you can, raise your chair by about 15cm and raise the speakers by the same mount. 1.1m would be fine if you were not getting those reflections, but with the tight setup that you have there, there's just not enough space to be able to sit that low.

Congratulations! It's good to see such positive results, so well documented. You are getting much closer to have great acoustics, and I'm CERTAIN you must have noticed a big difference in the way your room sounds. The bass should be much tighter and cleaner now, you should have a greatly improved stereo image and soundstage, with better clarity and "air" in the highs, and an overall feel of more spaciousness in the music: less "mud".




- Stuart -

[rhuobhe]

Hello Stuart,

Glad to hear the results are as good as I hoped. Difference is very much noticeable indeed, it's exactly like you've described. I will work on your recommendations after taking a little acoustics break. Did enough construction work lately, will get back to music making for now

I certainly couldn't have done it without your help, thanks a lot

[Soundman2020]

Difference is very much noticeable indeed, it's exactly like you've described.

I certainly couldn't have done it without your help, thanks a lot

You're welcome! That's what we're here for.


- Stuart -