Advice with acoustic panels

[Emiyanez]

Hi,

I recently bought a pair of ELAC UFR52 on sale after a demo (which I really liked), but I am having a hard time properly placing them in my room. Unfortunately, I am quite limited in terms of positioning and can only separate them a maximum of 15cm from the front wall. This is causing a severe cancellation (SBIR) right at 200Hz, making the speakers sound a bit "thin".

A fairly wide PEQ boost (Q=1, +4dB) partially solves the issue both in terms of measurements and subjective perception. However, since the crossover point between the woofers and the concentric midrange driver is at 220Hz, I am worried this boost might push the 4" midrange driver past its linear excursion limits at the very bottom of its range, potentially increasing intermodulation distortion.

By measuring the speakers at different distances, I confirmed the dip is pure SBIR from the wall behind them; pulling them out past 0.6 meters completely solves the issue, but unfortunately, that is not an option in my room.

I came across the GIK Acoustics 244 bass traps with FlexRange technology, and I am wondering if installing them on the wall right behind the speakers would successfully mitigate this reflection. Looking at their charts, the 90Hz option seems to be a reasonable candidate since it targets the problematic frequency area while leaving the lowest sub-bass untouched.

Would this actually work in practice given the tight clearance? Also, would a standard 120x60cm panel behind each speaker be enough to catch the reflection, or would I need to cover a much larger surface area to see a real improvement at 200Hz?

Thanks a lot in advance for your help!

 

[amirm]

Those panels, while not a proper "bass trap," do have coefficient of absorption reaching around 0.8 from 200 to 250 Hz. So should work well for your application.

 

[Keith_W]

Let's do some maths.

Depth of your speakers: 275mm
Distance of speakers to front wall: 150mm
Distance of baffle to front wall: 150+275mm = 425mm
SBIR wavelength = 4x distance of baffle to front wall = 1700mm
SBIR frequency = c/lambda = 343/1.7 = 201Hz

So yes, it is plausible that having your speakers 15cm from the wall is causing that cancellation. Now let's redo the maths when you pull the speakers 0.6m into the room:

Distance of baffle to front wall: 600+275mm = 875mm
SBIR wavelength = 4x distance of baffle to front wall = 3500mm
SBIR frequency = c/lambda = 343/3.5 = 98Hz

As you can see, pulling your speakers out lowers the SBIR frequency. What happens if you push it flush against the wall?

Distance of baffle to front wall: 275mm
SBIR wavelength = 4x distance of baffle to front wall = 1100mm
SBIR frequency = c/lambda = 343/3.5 = 312Hz

Shorter wavelengths are easier to absorb. So yes, you could consider those acoustic panels, but I would push the speakers to be as flush with them as I can.

 

[Paul7777x]

Let's do some maths.

Depth of your speakers: 275mm
Distance of speakers to front wall: 150mm
Distance of baffle to front wall: 150+275mm = 425mm
SBIR wavelength = 4x distance of baffle to front wall = 1700mm
SBIR frequency = c/lambda = 343/1.7 = 201Hz

So yes, it is plausible that having your speakers 15cm from the wall is causing that cancellation. Now let's redo the maths when you pull the speakers 0.6m into the room:

Distance of baffle to front wall: 600+275mm = 875mm
SBIR wavelength = 4x distance of baffle to front wall = 3500mm
SBIR frequency = c/lambda = 343/3.5 = 98Hz

As you can see, pulling your speakers out lowers the SBIR frequency. What happens if you push it flush against the wall?

Distance of baffle to front wall: 275mm
SBIR wavelength = 4x distance of baffle to front wall = 1100mm
SBIR frequency = c/lambda = 343/3.5 = 312Hz

Shorter wavelengths are easier to absorb. So yes, you could consider those acoustic panels, but I would push the speakers to be as flush with them as I can.

I have to learn this wizardry.

Could you kindly point me to a good source for beginners please?

 

[JPA]

I have to learn this wizardry.

Could you kindly point me to a good source for beginners please?

You could probably get all the information you need by searching the web, but that would be tedious and haphazard. Instead of doing that I recommend the authoritative and very readable
"Master Handbook of Acoustics" by Everest
https://www.amazon.com/s?k=f.+alton...3&ref=nb_sb_ss_p13n-expert-pd-ops-ranker_1_13

 

[NTK]

I have to learn this wizardry.

Could you kindly point me to a good source for beginners please?

Here is a quick explanation of this particular problem, sometimes also called adjacent boundary effects (e.g. in Dr Toole's book). For future questions on basic acoustics, may be we should have a dedicated thread for it.

When the sound wave reflected from the wall behind the speaker is of opposite polarity to the direct sound, we have cancellations (destructive interference). These cancellations will be most severe when the difference in the total distance of the reflected wave and the direct wave is 0.5 λ, 1.5 λ, 2.5 λ, ... etc. (where λ is the wavelength). The wavelength of a sound wave is given by the speed of sound in air divided by the frequency, λ = c/f *(where c is the speed of sound in air = 343 m/s at 20 °C, and f is frequency).

The distance to the wall behind the speaker is measured from the front of the speaker. The radiation point of the sound (acoustic center), is actually slightly in front of the speaker front baffle, but we can approximate it to be flush with the front baffle. Therefore, the the first (lowest) cancellation frequency, the wavelength equals 4 × the speaker to the wall distance. With the speaker to wall distance being 425 mm = 0.425 m, λ = 1.7 m, and f = c/λ = 343/1.7 = 202 Hz.

Note that the 2nd dip frequency f2 will be 3 × f1, and f3 will be 5 × f1, etc.

Half way between f1 & f2, f2 & f3 etc., we will have constructive interference, and therefore bumps in the frequency response.

Picture source: https://www.genelec.com/monitor-placement

Note:
* Explanation of the formula λ = c/f : The amount of time of one oscillation is 1/f. In the amount of time for 1 oscillation, the sound travels 1 wavelength λ. Therefore, the wavelength is equaled to the speed of sound c, multiplied by the oscillation period, 1/f, and λ = c × 1/f = c/f.

 

[Paul7777x]

Here is a quick explanation of this particular problem, sometimes also called adjacent boundary effects (e.g. in Dr Toole's book). For future questions on basic acoustics, may be we should have a dedicated thread for it.

When the sound wave reflected from the wall behind the speaker is of opposite polarity to the direct sound, we have cancellations (destructive interference). These cancellations will be most severe when the difference in the total distance of the reflected wave and the direct wave is 0.5 λ, 1.5 λ, 2.5 λ, ... etc. (where λ is the wavelength). The wavelength of a sound wave is given by the speed of sound in air divided by the frequency, λ = c/f *(where c is the speed of sound in air = 343 m/s at 20 °C, and f is frequency).

The distance to the wall behind the speaker is measured from the front of the speaker. The radiation point of the sound (acoustic center), is actually slightly in front of the speaker front baffle, but we can approximate it to be flush with the front baffle. Therefore, the the first (lowest) cancellation frequency, the wavelength equals 4 × the speaker to the wall distance. With the speaker to wall distance being 425 mm = 0.425 m, λ = 1.7 m, and f = c/λ = 343/1.7 = 202 Hz.

Note that the 2nd dip frequency f2 will be 3 × f1, and f3 will be 5 × f1, etc.

Half way between f1 & f2, f2 & f3 etc., we will have constructive interference, and therefore bumps in the frequency response.

Picture source: https://www.genelec.com/monitor-placement
View attachment 538670

Note:
* Explanation of the formula λ = c/f : The amount of time of one oscillation is 1/f. In the amount of time for 1 oscillation, the sound travels 1 wavelength λ. Therefore, the wavelength is equaled to the speed of sound c, multiplied by the oscillation period, 1/f, and λ = c × 1/f = c/f.

Excellent. Thank you for the lucid explanation. I’m hooked now.

 

[Emiyanez]

Wow! Thanks a lot for the explanations and the advice. I really appreciate it!

 

[Sigberg Audio]

In my experience, ~10cm panels placed like you suggest won't do a great deal with SBIR dips, but your mileage may vary.

I would also try to move them even closer to the wall as suggested by others. Finally, these types of dips aren't necessarily as audible as they look to be from the measurements. The improvement you hear (if any) by moving the speakers out in the room might just as well be from the fact that the speakers are closer to you, not necessarily due to changes in the SBIR-related dips.

 

[Emiyanez]

Finally, these types of dips aren't necessarily as audible as they look to be from the measurements.

Thanks! These are the measurement of the left and right speakers from the listening position (quite far from them). Closer to the speakers, the drop narrows down quite a bit.

I can live with peaks and drops but a drop off this magnitude in this region is unfortunately really evident when listening to music. So much that, at the beginning, I thought that the speakers had a problem or that the wiring was inverted internally and there was cancellation around there crossover frequency. I had never experienced something that bad in my previous room (speaker placement was much better though).

My hope is that the panels may help a bit and EQ does the rest. I am fully aware it may not end up being perfect.

Thanks again for the input.

 

[Sigberg Audio]

@Emiyanez Thank you for sharing the measurements. I would start by doing myself a favor and using 45-105dB (so 60dB span) on the Y-axis of the graph. This is what is typically shared, and will make it look a bit less severe.

Also set the X-axis to 20-20,000hz.

Could you perhaps share a similar measurement with the speakers 60cm out and also as close to the wall as you can get it?

 

[Keith_W]

Also, don't forget that SBIR occurs with side walls as well! The worst thing you can do is space your speakers approximately equidistant to front wall and side wall, or approximately the same distance. For e.g. if it's spaced 0.6m from the front wall and 0.8m from the side wall, you would get two cancellations. Sum them, and you get an extra wide and deep cancellation.

 

[Emiyanez]

@Emiyanez Thank you for sharing the measurements. I would start by doing myself a favor and using 45-105dB (so 60dB span) on the Y-axis of the graph. This is what is typically shared, and will make it look a bit less severe.

Also set the X-axis to 20-20,000hz.

Could you perhaps share a similar measurement with the speakers 60cm out and also as close to the wall as you can get it?

Will do!

 

[YouDontHaveEnoughBass]

As others have said if you read the leading advice from the likes of Genelec you either want the speakers really far or really close (a bit out from the wall being the nasty SBIR zone); I'm in the same situation and opted for the latter where my speakers (ascilab f6bs) are wall mounted with maybe a 1cm gap between back of speaker and wall

I'd suggest the same here, estimate the circumference of the ports at the back (im guessing theyll be around 5cm) and aim for that to be as close as possible without blocking the ports

I'm always beating this drum (see username) but any perceived lack of fullness after doing this is probably not from an SBIR dip at that point and more likely from not enough bass shelf in the 100hz and below region; the Harman curve allows a bass shelf all the way up to 12+ db for a reason!

 

[YouDontHaveEnoughBass]

Question for everyone while we're on the topic - has anyone tried mounting sealed/front ported speakers directly onto an acoustic panel on the wall? This seems like the ultimate solution for typical living room hifi type setups yet I've never seen it tried

 

[Sigberg Audio]

I'm always beating this drum (see username) but any perceived lack of fullness after doing this is probably not from an SBIR dip at that point and more likely from not enough bass shelf in the 100hz and below region; the Harman curve allows a bass shelf all the way up to 12+ db for a reason!

I'll contradict you slightly here. I'm sure your setup sounds great, but as a more general comment; Many who lack punch and impact in the bass keep increasing the bass shelf until they completely mask the midbass. I usually run ~6-7dB shelf ideally together with plenty of energy in the 100-300hz range.

 

[rajpacksound]

Let's do some maths.

Depth of your speakers: 275mm
Distance of speakers to front wall: 150mm
Distance of baffle to front wall: 150+275mm = 425mm
SBIR wavelength = 4x distance of baffle to front wall = 1700mm
SBIR frequency = c/lambda = 343/1.7 = 201Hz

So yes, it is plausible that having your speakers 15cm from the wall is causing that cancellation. Now let's redo the maths when you pull the speakers 0.6m into the room:

Distance of baffle to front wall: 600+275mm = 875mm
SBIR wavelength = 4x distance of baffle to front wall = 3500mm
SBIR frequency = c/lambda = 343/3.5 = 98Hz

As you can see, pulling your speakers out lowers the SBIR frequency. What happens if you push it flush against the wall?

Distance of baffle to front wall: 275mm
SBIR wavelength = 4x distance of baffle to front wall = 1100mm
SBIR frequency = c/lambda = 343/3.5 = 312Hz

Shorter wavelengths are easier to absorb. So yes, you could consider those acoustic panels, but I would push the speakers to be as flush with them as I can.

Moving speakers farther from the wall lowers the cancellation frequency, while placing them closer raises it. Since higher frequencies are easier to absorb, using acoustic panels works best when speakers are positioned as close to the wall as possible.

 

[Soniclife]

Thanks! These are the measurement of the left and right speakers from the listening position (quite far from them). Closer to the speakers, the drop narrows down quite a bit.

I can live with peaks and drops but a drop off this magnitude in this region is unfortunately really evident when listening to music. So much that, at the beginning, I thought that the speakers had a problem or that the wiring was inverted internally and there was cancellation around there crossover frequency. I had never experienced something that bad in my previous room (speaker placement was much better though).

My hope is that the panels may help a bit and EQ does the rest. I am fully aware it may not end up being perfect.

Thanks again for the input.

Before you try treatment I would try rebalancing your overall sound. You have to much energy above 600Hz, and to little from 200 to 600. I'd try bringing the top end down with a shelf filter to better match the lower range, try 1 or 2 PEQ below 600 to balance the sound, and the same below for between 40 and 150Hz, don't aim for perfectly flat just broad adjustments, especially above a few hundred Hz. I don't think you are hearing the dip, but the general lack of bass.

 

[Emiyanez]

Quick update on the topic... I ended up purchasing the acoustic panels and they did not help at all. Luckily, I could send them back without major issues.

Before you try treatment I would try rebalancing your overall sound. You have to much energy above 600Hz, and to little from 200 to 600. I'd try bringing the top end down with a shelf filter to better match the lower range, try 1 or 2 PEQ below 600 to balance the sound, and the same below for between 40 and 150Hz, don't aim for perfectly flat just broad adjustments, especially above a few hundred Hz. I don't think you are hearing the dip, but the general lack of bass.

@Soniclife Thanks! I followed your advice and created a shelf filter of -3db above 500Hz and combined it with a pair of PEQ filters to fill the dip at 200Hz and get rid of a resonance at the really low end. I kind of like what I hear but, as I am aging, I have the feeling that I started having problems with high frequencies, so the extra energy above 500Hz makes the music more detailed and enjoyable.

 

[Soniclife]

Quick update on the topic... I ended up purchasing the acoustic panels and they did not help at all. Luckily, I could send them back without major issues.

Did you take any measurements, would be interested in seeing them if you did.

I kind of like what I hear but, as I am aging, I have the feeling that I started having problems with high frequencies, so the extra energy above 500Hz makes the music more detailed and enjoyable.

It might be that no attenuation at the top is correct, or that a smaller reduction works best. Looking at your response again I'd suggest changing from a shelf to a PEQ at 1khz to bring that area down and leave the top alone. Maybe also try GSonic, there is a thread on this forum about it, it's free software.