With timber floors (or a wooden desk in the case of bookshelf speakers) what is the best "feet" to use. I have read quite conflicting recommendations in this regard (what? audiophiles not agreeing?).
On my Overnight Sensations I use those sticky felt discs that are deisgned to go on furniture to protect timber floors. This was what I had on hand and does a good job at protecting the speaker and the surface it is sitting on. I am about to build some floorstanders which will sit on my solid timber floor.
Is it a good or bad idea to decouple the speaker from a timber surface that it is sitting on from a sonic perspective? Other easy options I can think of would be a harder rubber pad which I imagine would have less damping effect on any resonance.
On my Overnight Sensations I use those sticky felt discs that are deisgned to go on furniture to protect timber floors. This was what I had on hand and does a good job at protecting the speaker and the surface it is sitting on. I am about to build some floorstanders which will sit on my solid timber floor.
Is it a good or bad idea to decouple the speaker from a timber surface that it is sitting on from a sonic perspective? Other easy options I can think of would be a harder rubber pad which I imagine would have less damping effect on any resonance.
How do you want to decouple from the floor? Hanging on ropes from the ceiling?
It's more a room acoustic question than a multi-way loudspeaker imo.
It's more a room acoustic question than a multi-way loudspeaker imo.
Transfer of vibrations through contact can be a problem. Even between driver and cabinet this can be more of an issue than standing waves within the box.
I'm not sure I see your concerns with damping to the same level. The cabinet itself should have the mass to support cone movement even if it were floating.
I'm not sure I see your concerns with damping to the same level. The cabinet itself should have the mass to support cone movement even if it were floating.
They are part of the speaker so I would imagine the issue is relevant to any speaker build.
Hi,
There is three ways to approach the situation: either coupling or decoupling or both.
_Coupling will give you a 'solid' reference for your box so your loudspeaker won't move when drivers do. But this won't prevent your room to contribute to message reproduced.
_Decoupling will make the box 'floating' upon your room so vibration should not propagate through material to the room and so limit the contribution of your wall, furniture,... to the music you listen to.
What is interesting is that in theory both requirements are needed for a theorical ideal but they are conflictual.
The 'best' approach to me is to do both at the same time: use of spikes on a rigid and heavy plate to couple the support and use of polymer materials ( sorbothane or other techniques) to decouple the loudspeaker box to the support ( plate).
It is not always practical however ( can be bulky, the nature of floor don't always allow for coupling through spike, shelf isn't rigid enough,...).
Imho you should try the three approach and find the one more practical to use for you.
Ysdr: how do you think decoupling is done in the industry? Hanging machine from ceiling or through the use of different technique?
In some factory you need to use microscope and other very sensible tools for quality check of items produced and this close to machine which produce heavy vibrations. Decoupling is known for a long time and efficient solution exist.
There is three ways to approach the situation: either coupling or decoupling or both.
_Coupling will give you a 'solid' reference for your box so your loudspeaker won't move when drivers do. But this won't prevent your room to contribute to message reproduced.
_Decoupling will make the box 'floating' upon your room so vibration should not propagate through material to the room and so limit the contribution of your wall, furniture,... to the music you listen to.
What is interesting is that in theory both requirements are needed for a theorical ideal but they are conflictual.
The 'best' approach to me is to do both at the same time: use of spikes on a rigid and heavy plate to couple the support and use of polymer materials ( sorbothane or other techniques) to decouple the loudspeaker box to the support ( plate).
It is not always practical however ( can be bulky, the nature of floor don't always allow for coupling through spike, shelf isn't rigid enough,...).
Imho you should try the three approach and find the one more practical to use for you.
Ysdr: how do you think decoupling is done in the industry? Hanging machine from ceiling or through the use of different technique?
In some factory you need to use microscope and other very sensible tools for quality check of items produced and this close to machine which produce heavy vibrations. Decoupling is known for a long time and efficient solution exist.
Sorry probably my poor use of terminology. It seems to me that speakers sitting on say carpet have a different connection to the room that those sitting on timber or concrete. My idea of felt pads I guess is not far different to carpet. I can imagine that timber could resonate a certain frequencies which may be a bad thing. Like krivium suggested, perhaps best to try different materials and see what sounds best in my particular environment.
Gilera,
For decoupling purpose you could try some commercialy available solutions:
Auralex have some isolation pads availlable:
Studio Monitor Isolation Pads | Auralex Acoustics
For polymer ( sorbothane) you have this kind of things:
Speaker Decoupling
For coupling, spikes...
My experience with decoupling is that if you have a solid foundation to boot on they are effective ( they change the rendering). If the surface you place them on is flexible it may behave strangely and can be impredictable ( sometimes renforcing coupling at some frequencies...).
In the second case coupling can be what is needed.
For decoupling purpose you could try some commercialy available solutions:
Auralex have some isolation pads availlable:
Studio Monitor Isolation Pads | Auralex Acoustics
For polymer ( sorbothane) you have this kind of things:
Speaker Decoupling
For coupling, spikes...
My experience with decoupling is that if you have a solid foundation to boot on they are effective ( they change the rendering). If the surface you place them on is flexible it may behave strangely and can be impredictable ( sometimes renforcing coupling at some frequencies...).
In the second case coupling can be what is needed.
Experimentation sounds good because it is not clear what vibrations will be transferred, until you measure..
1. Normally damping is said to be good, and it is but when used in this position it will have total control over only a limited scope of the overall movements. As with any damping it can only act upon where it is targeted.
2. Mass. Say a 12" woofer has a cone with MMS=100g. If the speaker weighs 1kg (10x) and we hang it up, it will stay still while the cone moves. Ok, there is a small vibration, lets make the cabinet 10kg, or 100kg. What difference is the floor going to make here?
Allen while we are talking about mass i wonder about mass loading a cabinet ( add mass over the loudspeaker cabinet):
I've heard change when doing this.
Would you think it is from the 'better ground reference' or from increasing the tensioning on cabinet structure ( changing the stiffness of cabinet build)?
I know only measure could tell but intuitively what would you say about it?
I've heard change when doing this.
Would you think it is from the 'better ground reference' or from increasing the tensioning on cabinet structure ( changing the stiffness of cabinet build)?
I know only measure could tell but intuitively what would you say about it?
On loading the cabinet, I estimate that a 100g cone acting evenly on a 30kg cabinet hanging in space, with a frontage averaging 4 times the cone area will produce pressure at -37dB to the main cone in a non-resonant manner. However, this piston will act in a dipole manner so there will be cancellation on top of that. Further, a cabinet on the floor may rock, rather than move freely back and forth... Doubling the cabinet mass to 60kg may reduce this by a further 6db.
Panels on the other hand... will be resonant, and potentially: for some resonances this includes the whole cabinet, so the front and back separately plus the sides, and with no dipole cancellation.
Panels on the other hand... will be resonant, and potentially: for some resonances this includes the whole cabinet, so the front and back separately plus the sides, and with no dipole cancellation.
I use ball in socket industrial feet with damping rubber that rests on my floor.
It worked better for me than 'poor man's spikes', which were ball socket threaded rods.
You can see them in this picture:
I used similar feet on my subwoofer.
It worked better for me than 'poor man's spikes', which were ball socket threaded rods.
You can see them in this picture:
I used similar feet on my subwoofer.
I have puzzled about this subject for almost 30 years, ever since the spikes and cones came into wide spread use by audiophiles in the late 1980s.
I can hear the effects of spikes on speakers. Most of the time it is subtle improvement, but sometimes it is a very noticeable improvement. Occasionally it makes things worse. And I am discounting those situations where the speaker actually causes an audible resonance in the floor structure, or vibrates itself across the floor. Of course something must be done in those situations. I am talking about an improved sound quality when there does not seem to be any problem that need fixing.
I am (was) a structural engineer, and I dealt with and worked through a lot of structural vibration problems over the years. Having said that, I do not understand why spikes and cones make a difference. I have never heard a theory that makes sense. There are some explanations floating around that seem to make sense conceptually or in an idealized sense. But these concepts don't stand up to scrutiny when you actually work through the math (Thank you Allen 🙂 you saved me the effort). Or even more discouraging, actually try to measure a difference in the sound.
If anyone knows of any research or white paper on this subject, I would love to review it...
So I believe they work, but I suspect that no one really knows why. It is one of several audio mysteries which our current understanding of the sound-to-ear-to-brain interface can not yet fully explain...
I can hear the effects of spikes on speakers. Most of the time it is subtle improvement, but sometimes it is a very noticeable improvement. Occasionally it makes things worse. And I am discounting those situations where the speaker actually causes an audible resonance in the floor structure, or vibrates itself across the floor. Of course something must be done in those situations. I am talking about an improved sound quality when there does not seem to be any problem that need fixing.
I am (was) a structural engineer, and I dealt with and worked through a lot of structural vibration problems over the years. Having said that, I do not understand why spikes and cones make a difference. I have never heard a theory that makes sense. There are some explanations floating around that seem to make sense conceptually or in an idealized sense. But these concepts don't stand up to scrutiny when you actually work through the math (Thank you Allen 🙂 you saved me the effort). Or even more discouraging, actually try to measure a difference in the sound.
If anyone knows of any research or white paper on this subject, I would love to review it...
So I believe they work, but I suspect that no one really knows why. It is one of several audio mysteries which our current understanding of the sound-to-ear-to-brain interface can not yet fully explain...
Last edited:
I used to run a mobile disco and the speakers were all on castors.
Never had any problems with that.
Never had any problems with that.
For bookshelf speakers on desks or cabinets I use 3M bumpers.
For speakers on polished floors I use plastic glides (see pic). Used to use spikes that went into centre of the X on PH2 screws in the floor but that was a pain.
The above two are arranged to give 3 point contact to the hard surface below them. 2 are located at the front outer edge and 2 are located in the centre of the back. I use 4 as I find it more stable than using just 3.
For carpet I use spikes so they contact the floor surface below.
For subwoofers I've tried SVS Soundpath or other similar methods.
Isn't the aim to stop the speaker from rocking? If you want to decouple from the floor then I imagine you have to go different routes. In the past I tried granite slabs under the speaker and these had rubber pads under them but was a waste of time and effort.
For speakers on polished floors I use plastic glides (see pic). Used to use spikes that went into centre of the X on PH2 screws in the floor but that was a pain.
The above two are arranged to give 3 point contact to the hard surface below them. 2 are located at the front outer edge and 2 are located in the centre of the back. I use 4 as I find it more stable than using just 3.
For carpet I use spikes so they contact the floor surface below.
For subwoofers I've tried SVS Soundpath or other similar methods.
Isn't the aim to stop the speaker from rocking? If you want to decouple from the floor then I imagine you have to go different routes. In the past I tried granite slabs under the speaker and these had rubber pads under them but was a waste of time and effort.
Attachments
Last edited:
Allen,
Thank you for your answer, it make sense to me. Always a pleasure as it give me source of inspiration!
May i ask how you get the -37db ? ( i didn't have this much of a scientific course during my school days as i was a literary).
Thank you for your answer, it make sense to me. Always a pleasure as it give me source of inspiration!
May i ask how you get the -37db ? ( i didn't have this much of a scientific course during my school days as i was a literary).
Yes, the 300:1 mass ratio goes to excursion, and the 4:1 area ratio gives back 12dB.
-(log(300/4))*20
-(log(300/4))*20
How about some small solid rubber balls, like those super balls. Tee nut with a bolt, drill ball to receive the bolt head and glue the ball on bolt head.
Thank you Allen! I expected something much more complicated, including speed of moving mass and other things at play... Anyway thank you it'll be quite usefull to me.
Puppet, it could work but it is a bit of a gamble: for decoupling usually you need to find a material you know it's shore and a given load under which it have an effect ( either damping or decoupling), otherwise this is a shot in the dark imho.
Wesayso's solution could be the cheapest indeed: you have access to spec of material and load under which they are effective but without the price targeted for 'audiophile' market. It has to be seen if it is ok with audio requirements though.
I've seen industrial solution which was effective at very low freq ( under 5hz iirc) which wasn't 'this' pricey given the freq and efficiency.
Puppet, it could work but it is a bit of a gamble: for decoupling usually you need to find a material you know it's shore and a given load under which it have an effect ( either damping or decoupling), otherwise this is a shot in the dark imho.
Wesayso's solution could be the cheapest indeed: you have access to spec of material and load under which they are effective but without the price targeted for 'audiophile' market. It has to be seen if it is ok with audio requirements though.
I've seen industrial solution which was effective at very low freq ( under 5hz iirc) which wasn't 'this' pricey given the freq and efficiency.
Last edited: