Need some tips on cheap acoustic treatment

[abdo123]

I don't know, those just seem like random lines beaten into submission by a monkey that has no clue.

Target spectrogram value should be somewhere around 200-400ms.

All of your values are around 200ms except from 100Hz to 200Hz

 

[Flaesh]

Is this caused by +2 Lf trim?

No. You can measure without sub for check. Strongly delayed energy due to some kind of reflections, IME it's unusual for the near field.

 

[Tupisac]

No. You can measure without sub for check. Strongly delayed energy due to some kind of reflections, IME it's unusual for the near field.

I'm talking about +2 LF trim setting on my LSR 305 monitors. Those are pretty close to a wall and according to manual it should be set at -2 at such position.

Anyway, this wavelet spectrogram is a pretty nifty feature, here is help file I'm currently reading: https://www.roomeqwizard.com/help/help_en-GB/html/graph_spectrogram.html

Nicely shows all the echoes and reflections. Mids look problematic too:

 

[Tupisac]

Seems like it's a tiny bit better with monitors very close to a wall - like it shifted the problem area up up a bit. Before that they were as far from the wall as mounts allowed.

LF Trim does basically nothing. LF -2:

LF +2:

Also, I've did a bit of experiment and dug old mattress from the attic. Held it up on the slanted ceiling and then on the front wall. There is a bit of improvement around 200 Hz area, but barely visible (basically only with dotted line).

Mattress on front wall:

Mattress on the slanted ceiling:

 

[Tupisac]

I'm sorry, left the amplitude in log scale. The difference with the mattress is more visible in linear.

No mattress:

Mattress on front wall:

Mattress on the ceiling:

Looks like it fixes something around 200 Hz. I guess couple of those monoliths might do audible difference. It breaks something in the 10k area tho.

 

[kemmler3D]

Could be that the mattress is actually reflecting the 10khz range back sooner than the ceiling was and causing some bump in response there? Very odd, not expected, but if the fabric is dense enough covering the mattress, really HF might not make it through?

If you want to fix in the low 100s of Hz you really need a limp mass or membrane trap... if you were willing to build a QRD, such a trap is no worse for DIY IMO, just needs different materials.

 

[Inner Space]

Could be that the mattress is actually reflecting the 10khz range back sooner than the ceiling was and causing some bump in response there? Very odd, not expected, but if the fabric is dense enough covering the mattress, really HF might not make it through?

Yes, this is a reflection off the mattress, surely. Most use slick, mercerized cotton for the cover fabric, which is probably more reflective than drywall at HF.

 

[Tupisac]

I've tried drawing some monoliths into my room plan, but 10 cm thick blocks are just too much of an eyesore is such small space.
I think the only viable solution is to rip the ceiling down and stuff a bunch of rockwool under the dryboard - but it's a tad too much for now, I've just finished painting

But there is hope - I have 30 cm deep bookshelf planned just above the monitors. Position looks perfect, covering the connection point of ceiling and a wall, almost like a proper bass trap. Drivers will be almost flush with it's edge, maybe slightly recessed like 1 cm. At a first glance it should actually cover huge part of a mirror reflection from a ceiling at my listening position.

How good are books as absorbers in 100-200 Hz range?

 

[kemmler3D]

Books are not good absorbers, they are considered okay diffusers, but at a higher frequency.

Treatments don't necessarily need to be on the walls near your desk to work, especially low frequencies. You could in theory put bass traps under the couch or something.

Consider the fact that the wavelengths are as big as the room, or bigger, and you will realize why it doesn't matter as much exactly where you put the low frequency treatments.

Shelves or no, I would still try thin foam or similar on the walls near your desk, clearly from the mattress experiment there is potential for really unwanted reflections there.

 

[sam_adams]

I've tried drawing some monoliths into my room plan, but 10 cm thick blocks are just too much of an eyesore is such small space.
I think the only viable solution is to rip the ceiling down and stuff a bunch of rockwool under the dryboard - but it's a tad too much for now, I've just finished painting

But there is hope - I have 30 cm deep bookshelf planned just above the monitors. Position looks perfect, covering the connection point of ceiling and a wall, almost like a proper bass trap. Drivers will be almost flush with it's edge, maybe slightly recessed like 1 cm. At a first glance it should actually cover huge part of a mirror reflection from a ceiling at my listening position.

How good are books as absorbers in 100-200 Hz range?
View attachment 256358

Yeah . . . that's probably not a good idea.

 

[kemmler3D]

Yeah . . . that's probably not a good idea.

Agreed. Bookshelves on the back wall seem to be popular, but whatever is wrong at the MLP, the shelves there will only make worse...

 

[sam_adams]

Melamine acoustic foam would be an ideal product to use in this situation. It's light weight, easy to attach to walls or ceiling, and it's fire rated. Other plusses is that it is fiber, oder, and mold/fungus free. It is easy to cut if you want to move beyond simple rectilinear shapes. The big con is that it is relatively expensive compared to other solutions. As far as placement is concerned, take a look at our good friend Bob Golds' RT60 analysis that includes some very important information from the late Eric Desart.

 

[Tupisac]

Here is the whole picture:

My desk is it the north-east corner, placed along the northern wall. Sub is below the desk on the right side. On the eastern wall I have a fitness ladder installed and in SE corner there is a niche with a wardrobe. Then on the S wall there's a door leading to a corridor, tiny bookshelf and then big double Ikea Kallax shelves. On the W wall there's a balcony window with curtains and a couch.

I see this potential places for a bit of foam or boxes filled with rockwool:

- thin stripe (5-8 cm) behind books on the N wall
- over the wardrobe in the SE corner
- few boxes inserted into Kallax spaces and placed on the top. Also, maybe small monolith directly in the SW corner.
- small monolith (10 cm thick tops) in the NW corner, under the long bookshelf behind the couch

Also, maybe a bit of foam for hf treatment glued to the walls in the NE corner.

What you guys think?

 

[kemmler3D]

Behind the books won't do anything for HF since the sound will bounce off the books first.

On the wardrobe also won't be maximally effective, since it's not a first reflection point. It won't hurt but you won't hear an obvious difference just from that.

Absorption material in the Kallax might be pretty helpful.

If you can stomach it, I would look at putting some foam on the slanted ceiling above the desk, behind the monitors, and on the walls directly to the left and right of the desk. If your problem is high frequency comb filtering, this is where most of your problems are coming from.

 

[Tupisac]

If you can stomach it, I would look at putting some foam on the slanted ceiling above the desk, behind the monitors, and on the walls directly to the left and right of the desk. If your problem is high frequency comb filtering, this is where most of your problems are coming from.

I only care about 150-250 Hz range as shown on spectrograms in earlier posts. You guys suggested looking outside of basic mirror reflection points for that.

I wonder which of those locations I've specified in my last post are potentially most effective - I guess where I can stuff the biggest mass of absorbing stuff, but that's just a wild assumption.

Also, what about height? As seen in couple of videos, usual triangular bass trap for a corner spans from the floor to the ceiling. Would it work when placed only in the upper corner, like over the wardrobe?

 

[kemmler3D]

I only care about 150-250 Hz range as shown on spectrograms in earlier posts.

Ah, got it, sorry.

I guess where I can stuff the biggest mass of absorbing stuff, but that's just a wild assumption.

Basically, yes. The lower the frequency you want to treat, the thicker the material needs to be. Putting ~30cm inside some of the kallax cubes could actually be helpful.

Also, what about height? As seen in couple of videos, usual triangular bass trap for a corner spans from the floor to the ceiling.

This is mostly to add more material overall, the long dimension of the trap won't tend to change the performance that much.

Would it work when placed only in the upper corner, like over the wardrobe?

For the same length and quantity of material, it *should* work about the same. The real issue is that you need the material to be physically located at a place where the velocity of the air due to sound waves is relatively high. This tends NOT to be the walls, so the corner bass trap thing is more because it's not practical to put them in other places, not because they work best there. Also, depending on the room itself, the nodes might be closer to one corner than the other at a given frequency. You can find this out by moving your mic around the room while playing tones.

 

[Tupisac]

Also, depending on the room itself, the nodes might be closer to one corner than the other at a given frequency. You can find this out by moving your mic around the room while playing tones.

That's really interesting tip.

Like, would it work if I just played pink noise spanning 150-250 Hz range and check potential treatment places with a mic for the loudest signal?

 

[Tupisac]

Yeah . . . that's probably not a good idea.

I don't know if I understood it correctly, but I think it's a very different case with my shelf full of books situation than described in the article.

What I have now is a one boundary situation, marked as 1B in the article. Well, can't do much with that. Speakers are staying on the wall because of mediocre amount of available real estate.

The two boundary situation (2B) in the article involves a 45 degree tilt and basically infinite second boundary at 90 degrees to the first, like in the satellite mounted near the ceiling.
My case is different in two aspects - first, there is no 45 degree tilt and second, more important in my opinion - technically there is no second boundary. I can get the drivers to be basically flush with the bookshelf edge. Along with books it should work like the extension of the front baffle of the speakers themselves. Kind of like in the solution at the end of the article...

 

[kemmler3D]

That's really interesting tip.

Like, would it work if I just played pink noise spanning 150-250 Hz range and check potential treatment places with a mic for the loudest signal?

I *think* that's how it works but I am somewhat overstating my knowledge here as I haven't done it that way myself. In theory, all of the nodes need to be 1/4WL away from the wall, or more - (which is partly why the treatment needs to be so thick) but in practice walls are not perfect reflectors, and you have more imperfect reflectors beyond the wall, so it can vary.

 

[Thomas_A]

I'm sorry, left the amplitude in log scale. The difference with the mattress is more visible in linear.

No mattress:

View attachment 256267


Mattress on front wall:

View attachment 256269


Mattress on the ceiling:

View attachment 256270


Looks like it fixes something around 200 Hz. I guess couple of those monoliths might do audible difference. It breaks something in the 10k area tho.

Experiment with two mattresses, one each from behind the speaker to cover about 40-50 cm against center of the middle and a bit up on the wall. Then get or build "real" damping panels behind. 10 cm deep frame with rock-wool covered with perforated board, and covered with 6 cm well-formed acoustic foam.