acoustic treatment for a small live room
 

Hi, i know there are thousands of posts talking about acoustic treatments, and i've read a lot. I think i understood some points but i need to improve myself on some other points.

My first acoustic treatment was a set of two 2 inches rockwool pannels on each walls of my room, the sound was dead but there was still a weird noise, like vibration in the sound.

I did not do any measurements of this first set-up.

Then i decided to use some good monitors, a nice desk, and i thought it would be nice to listen and play in that room : a part of the room for mixing and a part for playing acoustic guitar and singing.

I think this must be hard to achieve as the room is small.

2,90 X 3,10 meters (9,51 X 10,17 feet), ceiling at 1,95 m (6,39 feet)


Any advice would be warmly apreciated : What kind of treatment should i focus on ?

Should i add some wood planks on my super chunks in order not to dead the sound room ?

I did several new tests.

Room empty,

then 2 5CM fiberglass pannels in the rear corners.

Then 2 10CM fiberglass pannels in the rear corners

Then 2 15CM fiberglass pannels in the rear corners

Then 2 25CM made of 15Cm fiberglass pannels + 10 CM Metisse pannels (recycled garments tranformed in a kind of wool pannel)

Then i added to this last set up, 2 superchunks in the front corners.

Any idea how to reduce this 50hz and how would you call that kind of set up ? dead ? almost dead ? definitely what to not to do ? Good ?

Anyway, before and after, i can say the room sounds mat, maybe too much. But it's OK for listening music and recording voices.

I hope i am understandable.

Bye

G21

Well, there's certainly a good improvement in the response with each additional step you are taking, and the last one is looking quote good, actually... except for that 50 Hz resonance, which I'm assuming is modal? If so, your best bet for dealing with that is with a large, tuned membrane trap (panel trap), placed at the location of highest pressure for that problem. However, it's not so easy to tune such a trap, but it can be quite effective.

Yes, the room is heading for being a bit too dead in the high end: you might want to consider adding some reflective surfaces over your bass traps, to get some of the high mids and the high end back into the room again.

Ahhh! That's probably a big part of your problem! The back wall is always the one that gives you the most trouble, and needs the most treatment.... That 50 Hz issue is almost certainly your 1.0.0 mode, which is predicted to be at 55 Hz, based on the dimensions you gave.... If you don't treat the back wall, you wont be able to deal with that very well...

Thank you for taking time to reply.

In fact, the first post can be considered as a try. The second post is more a step by step with measurements in order to see the improvements or not.

Ok for the back wall. I'm going to build an helmholtz bass trap. I suppose i can hang it horizonally ?

And then i will try to add some relective surfaces on the bass traps of the front corners.

i'm also worried about the hollow point at 230. I don't know what to do about it. Time will tell...

Why Helmholtz? Are you aware of how hard they are to tune? A room mode is a very tight, high-Q resonance, just a few Hz wide. A Helmholtz resonator is a very tightly tuned, high-Q device, with a peak just a few Hz wide. Getting those two peaks to align is a lot harder in reality than the text boos make it sound: slight imperfections in your building materials, or workmanship, even changes in temperature and humidity (that make the materials expand or contract fractionally), can throw your tuning off. If you have built Helmholtz resonators before, and understand the process of tuning them, and are prepared to go through that, and have the equipment to do it, then by all means, that's a good approach. But you will need a very thick front panel (probably 40 mm at least), a very deep cavity behind it (probably 30cm or more), very tiny slots or holes, and very wide slats or hole spacing. Such a device takes up a lot of space in your room, and isn't very efficient (unless it is very large).

On the other hand, membrane traps are easier to build and to tune, much thinner, and don't need high precision to make, nor do you need an extremely thick front panel. You could build a limp membrane in less than one third of the thickness you'd need for a Helmholtz device tuned to the same frequency, and you could even build several of them, tuned to different frequencies, in the same total space that one single Helmholtz device would take up.

Your images only show up to 200 Hz, so I'm not seeing what you are referring to there. But a dip at 230 Hz is probably either SBIR or modal. My money would be on SBIR.

Great advices ! i found out a file showing how to calculate a membrane trap. I'm going to give it a try.

I have to disagreee with Soundman2020 here.

About helmholtz resonator: Q can be very high, or very low. Depends on the design.

More about it here:
Velocity based vs. pressure based absorbers

and more specifically here:
helmholz not very popular?

In my experience (My iso-room is almost those exact dimensions and had a wicked 50hz mode issue) the membrane absorber(s) are the way to go. Easy to build and tune. When I get back to my own computer I can post some plans.

Alright, if I recall correctly, the traps were based on this calculation: Multi-layer Absorber Calculator

Unfortunately, my sketchup files are apparently on my studio computer, not my laptop, but a picture or two can give you the idea.

They were placed so (on this wall... In the end I used 7, roughly the same on each wall, except for one where the door is.):
http://oi63.tinypic.com/qxo20m.jpg

The walls were then filled so:
http://oi64.tinypic.com/18hvu0.jpg

The result of 2 walls, so 4 traps plus fiber (Again, the mdats are on my studio computer, when I'm next there I'll post them if I remember).
Before:
http://oi65.tinypic.com/de728k.jpg
After:
http://oi68.tinypic.com/31623d1.jpg

All 4 walls and corners done, as I said, I need to get when I'm next in the studio.

OK, apparently you can't use dropbox links anymore on Gearslutz... I'll try and find another solution.

And apparently ImageBB doesn't work either... Anyone have a solution to posting pictures here now?

Found a solution via tinypic. Don't recommend it for anyone somewhat security conscious.

I found some building instructions for membrane traps working with a membrane made of 10 mm Medium wood. 15CM deep. 1M X 1M. It works for 55/57 peaks.

I though it would be better to fix 2 of them on the wall behind me. At middle height of the wall.

I'll try with one, then i'll see if it's worth going that way.


Thank you for sharing your experience.

G21

The opposite is true:

A HH array is easily "tuned" (predictable) and does often not need to be very precisely tuned since unless very shallow; they have relatively low Q compared to membrane absorbers.

Membrane absorbers are very hard to "tune" (predict in terms of fc in design stage) and have high Q compared to Helmholtz arrays. The only time I use membranes is for very low frequencies (< 30 Hz) and/or if imited available build depth.

I'd recommend hardboard over wood for panel traps for a few reasons. First, the density is much more consistent, which makes calculations more reliable. Second, although being denser, hardboard is much more flexible than wood (less depth and stiffness needed to meet mass requirements), allowing it to vibrate more freely, which is of course the goal. Third, cost per kg/m2 for hardboard is cheaper than wood. Finally, you save those millimetres of depth in your trap build, which in a room 3m wide, every millimetre counts.

I read A LOT. And very often, membrane bass traps are easy to calculate, hard to build.

I can't comment on my personal experience. I never had to used it. But can comment on helmholtz; very effective, and with a good design, Q can be wide enough to cover a big band and still doing its job!

My excel sheet works only with MDF or multiply. MDF is like hardboard but a bit thiker.

Thank you in advance.

G21

Use the calculator I linked above. It works pretty spot on. I calculate lower than my intended frequency for 2 reasons: Variable density of the material I'm working with (I weight the plate and calculate its average density for the final calculation, but variances in the material density can effect the frequency centre, usually moving it higher), and to account for the tensioning of the plate on four sides, which will slightly increase the resonant frequency with my construction method. In the last few years, all of my membrane traps have been successfully tuned within 2Hz of the target frequency. With an absorption coefficient (theoretical, hard to measure since we're tackling very specific modes) of 0.8 over 2 octaves, I can say that my target frequency has been tackled.

MDF is like hardboard, but lighter, and stiffer. At lower thicknesses you can use mdf to good effect (I have used it in the past), but the trade-off is that you have to make the traps deeper.

How are membranes hard to build? Drilling precise holes is a lot harder.

Ok... I found a hardboard (isorel) dealer.

Are we talking of the 3mm one ?

I will post the building instructions i found.

Who said holes was the only way to do it?

I found a lot of informations on this link. (In french)

Les Bass traps a membrane souple et rigide

I think i will go for one of these two shape :

The first one, i am used to perform that kind of work.

The second one is considered best.

Holes are the only way to do it. They can be round holes, square holes, or long rectangular holes (slots), but without holes you just have a sealed box. If you wanted to pick on something to argue about, you should have focused on me saying drilling. ;)

Nonetheless, whether you drill or use slats, I stand by my statement. For the general DIYer achieving the precision required for Helmholtz is more difficult than it is for membranes. It's not impossible, just more difficult.

The nice thing about hardboard is that the thickness nicely corresponds to the mass (kg/m2). 3mm is usually 3kg/m2, 5mm about 5kg/m2. In order to be accurate, you have to weigh the material before finalising your design, as its mass will determine the depth you require in the trap to reach your desired frequency.

Multi-layer Absorber Calculator

https://www.gearslutz.com/board/atta...1&d=1552046623

300 mm deep HH array incident angle from Soundflow.

One panel has 6 mm gaps and the other one 7 mm gaps between the slats. So even if you mess up by this much; the performance is almost identical. There is no precision required to build a HH array.

A membrane on the other hand due to the high Q requires quite a lot of attention. Membranes are a last resort for me if I need to combat very low frequencies and/or the available build depth is very limited. For all other low frequency absorption needs I rely solely on HH arrays since it´s she safest bet (easy to predict and build plus offers a wide bandwidth so you don´t need to worry about missing your target frequency range).

Thanks for the links.

The author of this web page points to issue

As J Eklund, the unpredictability function of the type of membrane.
The low durability in the time leading the detuning.

+ 1 to Jens comments.

Jens is a serious gearslatz!

Okay !!!!

As my room is quite dead, even if i have only treated the corners and as i wish to improve highs, can i build this HH in the corner. At this very moment, corner traps are 40cm deep, with a face of 80cm. that is to say an average 25 cm depth.
i choose the incorrect wool density. It's around 1500 Pa.s/m2. So they works like this :

Multi-layer Absorber Calculator

But if i transform this setup in an HH one (i will have to seal the boundaries), i can get this :

Multi-layer Absorber Calculator

Do you think i can follow this idea ? before keeping working on the walls ?

Things seemed so easy at the beginning....

(What is the posts limits beyond wich i will be able to link to some images ?)

Did you mean 400mm thickness?

1500 Pa.s/m2 ...which is 1.5 kPa.s/m2 doesn't exist. Did you mean 15kPa.s/m2?

Yes, 40cm

and

Well. There are pannels made of recycled garments. It is mainly cotton and a bit of polyester. Impossible to find some accurate informations about it except its density that is 20kg/m3.

In France it is named "metisse".

http://assets.locomotivehosting.com/...pdf?1548346249

I have read that it should be compared to hemp pannels.

Would you mind posting all the relevant data that you inputted to achieve this spectacular result finally?

2 parameters are clear, slot width (6mm), and trap depth (300mm). I'm assuming the 90mm is the slat width, or slot spacing, but correct me if I'm wrong. What's the depth of the slats? What about air gap, if any? GFR of the absorption?

I'd love to get to this result, but no other models come even close.

Wool (300 mm) = Rockwool Flexibatts (8,8 kPa*s/m², 30 kg/m3)

Slats 16 or 19 mm thick

Hole spacing = 90 mm

Model = Miki