# Best (affordable) way to decouple my speaker stands?

OP

#### JaccoW

##### Active Member
I'm loving all the suggestions out here. Even if I only end up using one of them it's cool to see what others have come up with.

[edit: sorry I replied to the wrong post. This was meant to be directed @JaccoW ]
The sorbothathe hemispheres should work for you. You just need to make sure they are loaded properly to get the best results. You want them to be loaded close to the max rating. Then the natural frequency will be less than 10Hz and you will have good isolation down to 20-30Hz.
View attachment 171778
You can easily calculate the natural frequency if you know the deflection. Attached are some hemisphere deflection curves from Sorbothane (small are the 0.75" dia, medium are the 1.25" diameter, large are the 2" diameter).

The natural frequency is just fn=(1/2*pi)*(g/D)^0.5 [see here: https://ocw.mit.edu/courses/mechani...ti-dof-vibration/MIT2_003SCF11_rec10note1.pdf]

g is acceleration from gravity; D is deflection. If you use the Sorbothane deflection curves convert inches to feet and then you can use g=32.2 ft/s^2 in the calculation.
I'll see if I can find the metric equivalent of that formula, but for now I have the 30mm (1.18") sorbothane hemisperes incoming which are rated at 8-16kg for 4 spheres. With the current load of the speakers + stands + marble plate sitting at around 13kg that should be well within its useful range.

EDIT: You might enjoy the generic formula from the Product guide of Sorbothane itself.

Hi

I have had good results with these.

The price seems to have increased a bit since I last bought any but you may be able to find similar cheaper. No measurements though just my own feelings.
Yeah @spiral scratch and @muslhead mentioned them earlier in the thread as well. A quick search for Wagner MP-4E seems to turn up 10x10x2cm (4"x4"x0.8") for around €5 (\$5.60) a piece. Just ordered some to experiment with... though I only realized after ordering that I might have ordered a lifetime supply if you are supposed to cut them up into smaller pieces.

Last edited:

#### GalZohar

##### Senior Member
do you know the word SBIR ?
A speaker must be against the front wall, always.
Supposedly SBIR should be reduced if the ratio between your distance to the wall and the distance to your speakers is large enough (so the 1st reflection is significantly lower energy than the direct sound). In any case, you'll have SBIR at any distance except in-wall, and if you can't listen at all because of neighbor issues then none of it matters anyway.

OP

#### JaccoW

##### Active Member
I think I might have a relative/semi-scientific way to measure how well decoupling works. It's a bit rough right now but once I get some more options in I might be able to clean up the data a little bit. Right now there are some differences in the length of the run and some accidental hotspots that show up.

The idea is this;
I use a accelerometer app on my phone and measure at different points on and around the speakers. I then play the same (bass-heavy) song for 20 seconds and record the results. In this case Rebirth of Cool by DJ Cam Quartet but I think The making of a Cyborg from the Ghost in the Shell movie might be a good candidate as well.

Right now I took some measurements at 4 places:
• On top of the speaker
• On top of the foot of the speaker
• On the marble plate
• On the ground right in front of the speaker.
Which brings us to the following rough results:

Top of speaker:
As expected, lots of vibration. Some of the peaks are going as high as 0.4276 m/s^2.

Foot of speaker stand:
Much more controlled but still a lot of vibration. Maximum value at 0.2510 m/s^2.
Noticable vibration when I put my hand on it.

Marble plate:
Separated by some silicon bumbers between the stand and the marble plate this level drops the vibrations in half to peaks around 0.1135 m/s^2.
By hand I can hardly feel any vibration anymore.

Ground:
Funnily enough we see a rise here again. Maybe it's the effect of the sound coming out of the speakers?
Peaks rise to 0.1400-0.1600 m/s^2. This can actually be felt by hand again as this feels like it vibrates more than the marble plate.

Connected to the ground with hard rubber hemisperes (which were supposed to be dampening) so let's see what the foam blocks do here once they come in next week.

Conclusion:
Like I said, this is a first attempt to measure how succesfull a certain type of decoupling is. The peaks you see near the end are mostly me bumping into the phone when pressing for the recording to stop. I probably need a couple more measurements;
• Noise level baseline silent (no music playing)
• In front of listening position
And maybe I need to couple the phone using spikes.

I'm open to any comments and improvements regarding this method but at least I do measure what I feel here.

More to come!

D

#### Deleted member 23424

##### Guest
I think I might have a relative/semi-scientific way to measure how well decoupling works. It's a bit rough right now but once I get some more options in I might be able to clean up the data a little bit. Right now there are some differences in the length of the run and some accidental hotspots that show up.

The idea is this;
I use a accelerometer app on my phone and measure at different points on and around the speakers. I then play the same (bass-heavy) song for 20 seconds and record the results. In this case Rebirth of Cool by DJ Cam Quartet but I think The making of a Cyborg from the Ghost in the Shell movie might be a good candidate as well.

Right now I took some measurements at 4 places:
• On top of the speaker
• On top of the foot of the speaker
• On the marble plate
• On the ground right in front of the speaker.
Which brings us to the following rough results:

Top of speaker:
As expected, lots of vibration. Some of the peaks are going as high as 0.4276 m/s^2.

Foot of speaker stand:
Much more controlled but still a lot of vibration. Maximum value at 0.2510 m/s^2.
Noticable vibration when I put my hand on it.

Marble plate:
Separated by some silicon bumbers between the stand and the marble plate this level drops the vibrations in half to peaks around 0.1135 m/s^2.
By hand I can hardly feel any vibration anymore.

Ground:
Funnily enough we see a rise here again. Maybe it's the effect of the sound coming out of the speakers?
Peaks rise to 0.1400-0.1600 m/s^2. This can actually be felt by hand again as this feels like it vibrates more than the marble plate.

Connected to the ground with hard rubber hemisperes (which were supposed to be dampening) so let's see what the foam blocks do here once they come in next week.

Conclusion:
Like I said, this is a first attempt to measure how succesfull a certain type of decoupling is. The peaks you see near the end are mostly me bumping into the phone when pressing for the recording to stop. I probably need a couple more measurements;
• Noise level baseline silent (no music playing)
• In front of listening position
And maybe I need to couple the phone using spikes.

I'm open to any comments and improvements regarding this method but at least I do measure what I feel here.

More to come!
Nice. Watching this with interest. My M16 are on spiked stands which sit on a suspended wooden floor, as does my sub. Had been planning on buying some sorbothene blobs but am intrigued by the MP4E suggestions. I look forward to your results.

OP

#### JaccoW

##### Active Member
Nice. Watching this with interest. My M16 are on spiked stands which sit on a suspended wooden floor, as does my sub. Had been planning on buying some sorbothene blobs but am intrigued by the MP4E suggestions. I look forward to your results.
The Sorbothane hemispheres won't arrive until after christmas so I'm curious as well!

#### storing

##### Active Member
Forum Donor
If you want to optimize you need more data and also you need to now what your optimizing for.
Best absorption per wait. or per volume. or cost?
Do you care abut damping? or only abut decoupling...

I guess my question was: what works best for speakers (let's say like the ones the OP has), irregardless of cost, where 'best' means lowest sound level (preferreably shown as spectrum because it's likely different applications work on different frequencies) for the neighbors below. I realize there are many variables here (floor type for one) but still would be interesting to see real-life case studies. I searched around a bit but didn't find anything. (At the same time I wonder how it affects sound in the room itself but that's something else warranting its own case studies).

#### spiral scratch

##### Member
You don't need to cut up those AV pads, but the pads are rated for 50psi so one shouldn't need to use so much pad to decouple lighter objects. They are a modular design so If you want to stack the pads they will be more stable if they are wider than 1 inch.

#### mcdn

##### Active Member
Forum Donor
I think I might have a relative/semi-scientific way to measure how well decoupling works. It's a bit rough right now but once I get some more options in I might be able to clean up the data a little bit. Right now there are some differences in the length of the run and some accidental hotspots that show up.
Brilliant! One suggestion is to measure in the middle of the floor as well as close to the speaker. I had the same problem with suspended wooden floors in typical 19th century houses in the UK and they flex a lot in the centre of the room.

OP

#### JaccoW

##### Active Member
You don't need to cut up those AV pads, but the pads are rated for 50psi so one shouldn't need to use so much pad to decouple lighter objects. They are a modular design so If you want to stack the pads they will be more stable if they are wider than 1 inch.
True. These are designed to withstand 50psi, which translates to 3.51535kg/cm². Meaning these 4"x4" / 10cm x 10cm pads can carry about 800lbs / 364kg before they collapse I guess?

Most websites I come across say they work best between 80-100kg (175-220lbs) though and 1/16th of these pads will offer that 50psi / 3.51kg/cm² load capacity. I have 8 of these on the way so I can test the 4 x 1/16th and 4 x 1/4th and 4 x 1 options.

The location is a question though. Do I put these between the speaker stand and the marble plate or underneath the marble plate?

#### cicastol

##### Member
The best way to reduce vibrations induced by loudspeakers is to coupling them to a mass with spikes ( granite slab or equivalent at least the same mass of the loudspeaker ,more is better) doing "mass loading" and then decouple the whole with viscoelastic feet .
Photos of my DBR62 coupled with spikes to a granite slab and then decoupled with Polysound Sylo (silicone foam isolation pads),stainless steel feet are glued with cyano to the granite slab for maximum rigidity.

#### Attachments

• IMG_20211105_153940 (1).jpg
286.1 KB · Views: 177
• IMG_20211105_160907.jpg
334.6 KB · Views: 175

#### Lambda

##### Major Contributor
The location is a question though. Do I put these between the speaker stand and the marble plate or underneath the marble plate?
I would say isolate the marble plate from the ground.
But if you can used some sort of soft absorbent material between the speaker stand and the plate as well.

I think I might have a relative/semi-scientific way to measure how well decoupling works. It's a bit rough right now but once I get some more options in I might be able to clean up the data a little bit. Right now there are some differences in the length of the run and some accidental hotspots that show up.
This is a nice test!
I wound suggest replacing Music by a Sine sweep from lets say 20-200Hz over 30seconds or so.
Put some strong impulse and a second of silence in front of the sweep so synchronize this to the start of your accelerometer recording.

So later you can correlate the aptitude in the accelerometer recording to an frequency by looking at the time since the start pulse.
if this makes sense.

Last edited:
OP

#### JaccoW

##### Active Member
I would say isolate the marble plate from the ground.
But if you can used some sort of soft absorbent material between the speaker stand and the plate as well.

This is a nice test!
I wound suggest replacing Music by a Sine sweep from lets say 20-200Hz over 30seconds or so.
Put some strong impulse and a second of silence in front of the sweep so synchronize this to the start of your accelerometer recording.

So later you can correlate the aptitude in the accelerometer recording to an frequency by looking at the time since the start pulse.
if this makes sense.
Maybe combined with the Genelec Boink that does a 16, 18, 20, 22, 26, 30, 35, 40, 45, 50, 60, 70, 80, 90, 100, 120 and 150Hz test tone?

I also came across this one that has individual constant frequencies for 30 seconds each. Add them to a playlist and test away.

20Hz-20kHz sweep

16-150Hz Boink

#### Lambda

##### Major Contributor
maybe this helps:

20-200Hz in 20s

#### Attachments

• sweep.zip
96 KB · Views: 26
OP

#### JaccoW

##### Active Member
Just got some reminders from this post. Maybe I should continue where I left off last time and start measuring again.

#### Farenheit

##### Member
Attention!:

As you can see in the attached photo, tomorrow I am going to place a camera wheel under my 57 Kg (125 lb) JBL 3100S to isolate it from the ground.

The floor is driftwood.

If the camera wheel doesn't blow up and give me a heart attack, I'll tell you the results...

OP

#### JaccoW

##### Active Member
Attention!:

As you can see in the attached photo, tomorrow I am going to place a camera wheel under my 57 Kg (125 lb) JBL 3100S to isolate it from the ground.

The floor is driftwood.

If the camera wheel doesn't blow up and give me a heart attack, I'll tell you the results...View attachment 223955
Are you still alive?

#### Farenheit

##### Member
Are you still alive?
Total disaster, but we are in a scientific forum and I had to prove it scientifically.

The speakers danced to their own beat and he wasn't about to lose them around the house.

Note: My health is fine.

Replies
45
Views
2K
Replies
45
Views
2K
D
Replies
39
Views
1K
Deleted member 46664
D
Replies
4
Views
654
Replies
57
Views
5K