advertisement


Vibration Isolation Experiments

Mr Pig said:
Have you looked at Stack Audio's new isolation feet? They're not cheap but the few reviews they've had are outstanding. They use a novel method of dissipating energy and apparently they're quite hard to make.
Click to expand...
Yes, I have, also boxes of sand, a pendulum approach, SDS constrained layer damping steel sheets, all manner of footers with various approaches to dissipating energy and or coupling (Gaia, Stacore, Wilson Audio, Magico, Symposium, Wellfloat, etc). In the end I chose roller bearings as it was cheap and easy to experiment with and then spend more to get a high quality solution. I also chose springs (metal and air bladder) as I have a lot of experience using them in suspensions, understand the physics of using them to isolate and they are heavily used in industry to isolate noisy equipment or isolate equipment from the ground. And the Townshend approach was already well regarded and reviewed, which fitted with my knowledge and experience, with the Sonority approach an interesting variation, with a lot of listening and testing behind it.

A possibly interesting and related audible and measurable road traffic noise effect:

I used to work at Mercedes F1 in Brackley where there is an auditorium that looks out onto the A43. The Auditorium is 50m from the A43 and had always struggled with poor acoustics. An acoustician was called in and used tuned Helmholtz resonators to reduce the room modes (I then got him to treat my lounge after hearing how effective it was). But there remained some mid range noise coming from the main support columns and the windows and frames. Then a year or so later we had the fuel strikes and traffic on the A43 dropped to the odd car every few minutes. With no cars or lorries going past the noise in the Auditorium disappeared and then returned as soon as road traffic went past. This showed me that ground borne noise is significant. So we filled the support columns with acoustic foam and that problem disappeared, just leaving the window frames, which were more difficult to treat and then we ran out of money to treat the building further having got to an 80% solution.

At McLaren F1 we tested an active ride suspension corner actuator for the 1991 car on a corner actuator test rig. This test rig was sitting on footers on the floor. We ran the test from 1 Hz up to around 20 Hz and very quickly set various parts of the building resonating (complaints from many people a long way away in the building). So we purchased air springs to put under the feet and never had a problem again.
 
Curious what is between the bottom of the loudspeaker and the board it is sitting on? There’s a section on the Symposium site that discusses using a Svelte Shelf under loudspeakers. They recommend removing any spikes from the loudspeaker and placing it in direct contact with the platform. The platform is a big part of the implementation of Symposium products under loudspeakers whose purpose is energy dissipation.
 
John said:
Curious what is between the bottom of the loudspeaker and the board it is sitting on? There’s a section on the Symposium site that discusses using a Svelte Shelf under loudspeakers. They recommend removing any spikes from the loudspeaker and placing it in direct contact with the platform. The platform is a big part of the implementation of Symposium products under loudspeakers whose purpose is energy dissipation.
Click to expand...
Spikes were removed to give the speaker bottom direct contact with the board and enable better energy transfer than can be achieved with 4 spikes.

I am aware of the Symposium approach and am thinking about how that really works and how it could be achieved.
 
IanW said:
Spikes were removed to give the speaker bottom direct contact with the board and enable better energy transfer than can be achieved with 4 spikes.

I am aware of the Symposium approach and am thinking about how that really works and how it could be achieved.
Click to expand...
Do you know what the composition of the board the loudspeaker is placed on or it’s purpose?
 
John said:
Do you know what the composition of the board the loudspeaker is placed on or it’s purpose?
Click to expand...
A layered board to absorb energy that feels pretty acoustically dead.

The best person to ask is @Sonority as it is his design after a lot of research and testing.
 
IanW said:
Some photos taken during and after the installation process of the prototype speaker stands that isolate in all 6 degrees of freedom:

53E010A3-7509-4F65-99E1-55D8A929C85F by Ian Wright, on Flickr

509AF88A-2599-482D-809F-AD24DE57B6E8 by Ian Wright, on Flickr

DSC00001 by Ian Wright, on Flickr

Steve's roller shelves etc can be seen on his website:

https://www.sonoritydesign.co.uk/platforms/
Click to expand...

Nice. Looks a little like Nobsound springs but a heavy duty version with platform.
 
ryder said:
Nice. Looks a little like Nobsound springs but a heavy duty version with platform.
Click to expand...
I started out by purchasing some Nobsound springs. They provide some isolation which is good, but then I realised that the springs were too soft for my requirements. I purchased a set of stiffer springs but could not get a set with the right outer diameter to fit that well. The springs are too short and I calculated were typically isolating from 25 Hz upwards and I wanted much lower. The springs would rattle in use and so I coated them with heat shrink to get rid of the rattling and the transfer of higher frequencies. But I was never happy with the heatshrink installation (subsequently bought a heat gun to do the job properly on the bigger springs) and the heatshrink increased the stiffness massively (and the damping) as the spring was so soft in comparison. As I looked into the physics it became clear that the springs were too short and could never provide lower frequency isolation.

As a rule of thumb you need to the unit to be isolated to compress the springs by around 25mm and still have free travel to have a resonant frequency of just over 3 Hz and hence increasing isolation from around 5 Hz. Interestingly this is why for industrial machine isolation you would most likely specify the static loaded deflection as that defines the resonant frequency.

It was a useful part of my learning through experimentation and then looking further into the physics. And yes conceptually the Nobsound approach is an early version of where I ended up.
 
It
IanW said:
I started out by purchasing some Nobsound springs. They provide some isolation which is good, but then I realised that the springs were too soft for my requirements. I purchased a set of stiffer springs but could not get a set with the right outer diameter to fit that well. The springs are too short and I calculated were typically isolating from 25 Hz upwards and I wanted much lower. The springs would rattle in use and so I coated them with heat shrink to get rid of the rattling and the transfer of higher frequencies. But I was never happy with the heatshrink installation (subsequently bought a heat gun to do the job properly on the bigger springs) and the heatshrink increased the stiffness massively (and the damping) as the spring was so soft in comparison. As I looked into the physics it became clear that the springs were too short and could never provide lower frequency isolation.

As a rule of thumb you need to the unit to be isolated to compress the springs by around 25mm and still have free travel to have a resonant frequency of just over 3 Hz and hence increasing isolation from around 5 Hz. Interestingly this is why for industrial machine isolation you would most likely specify the static loaded deflection as that defines the resonant frequency.

It was a useful part of my learning through experimentation and then looking further into the physics. And yes conceptually the Nobsound approach is an early version of where I ended up.
Click to expand...

It appears that you have done your homework and have some technical knowledge on how isolation works. I have some Nobsound springs below my amps and they do work. However, some effort is required in getting to the optimal compression / spring configuration.

Your labour of love looks like a combination of Nobsound and Townshend. Impressive.
 
ryder said:
It


It appears that you have done your homework and have some technical knowledge on how isolation works. I have some Nobsound springs below my amps and they do work. However, some effort is required in getting to the optimal compression / spring configuration.

Your labour of love looks like a combination of Nobsound and Townshend. Impressive.
Click to expand...
Thanks.

Yesterday I got the Nobsound spring assembly out to experiment further with some isolation in my second system. I agree that they do provide some isolation (measured with the iPad accelerometer and subjective sound) and as you have stated, some effort is required to get to the optimal solution for what is available. Based on what I have learnt and having got some more heatshrink and the heat shrink gun, I will do some calculations and try again with some isolation under my Audio Synthesis DAX Decade.
 
Hi Ian, interesting thread on a much misunderstood couple/decouple quandary.

The popularity of spikes is unfounded as they certainly do not decouple and also do not act as one-way mechanical diodes. Those sharp points couple well to whatever they rest upon and IMO their only saving grace is that the points form a plane and provide stability.

Your positive results with springs is understandable. I'm considering, when I get the time, to DIY something along the lines of Townshends units. I have found a foamed silicone tube that claims to have a fast recovery rate and will try this as damping for the springs.

Something else that may interest you is to, instead of feeling for vibration by hand, use an old unwanted and unloved tonearm with cheapie cartridge. I used an old Magnaplanar uni-pivot arm with Nagaoka cartridge and hooked up to amplifier. Much more sensitive than a hand job :)
If you run a signal genny sweep it will clearly show up any resonances on your multimeter.
 
sailor said:
Hi Ian, interesting thread on a much misunderstood couple/decouple quandary.

The popularity of spikes is unfounded as they certainly do not decouple and also do not act as one-way mechanical diodes. Those sharp points couple well to whatever they rest upon and IMO their only saving grace is that the points form a plane and provide stability.

Your positive results with springs is understandable. I'm considering, when I get the time, to DIY something along the lines of Townshends units. I have found a foamed silicone tube that claims to have a fast recovery rate and will try this as damping for the springs.
Click to expand...
I have been looking for an alternative to heatshrink on the springs. A visco elastic tube that touches the spring on the inside is a possibility but you do need to be careful as increasing the damping too far will reduce the isolation that you see at higher frequencies. Or as you have suggested, just use the silicone tube as a spring replacement. I have some experience of doing this in F1 race car suspensions but found that the non linearity, hysteresis and need to get the correct load range will be more of a challenge for this requirement. Another alternative is to dip the springs in visco elastic fluid that solidifies its surface in the air but maintains the visco elastic nature beneath the surface.
sailor said:
Something else that may interest you is to, instead of feeling for vibration by hand, use an old unwanted and unloved tonearm with cheapie cartridge. I used an old Magnaplanar uni-pivot arm with Nagaoka cartridge and hooked up to amplifier. Much more sensitive than a hand job :)
If you run a signal genny sweep it will clearly show up any resonances on your multimeter.
Click to expand...
I have tended to use the accelerometers in the iPad for measurements (use of the palm of my hand or finger pad is just to confirm that something has changed). I have now purchased a standalone measurement USB accelerometer that has a wider measurement range and some processing software. I can then do overlays etc when I have exported the data to Python etc.

The use of a tonearm and a cartridge and an amplifier remains a possibility but then I would need an A to D to get the data into a usable form and enable some frequency based analysis. All doable of course.

Your method would work with a frequency sweep as an input and look for the magnitude at a series of frequencies.

But with the calibrated USB accelerometer a lot of the hard work has been done for me.
 
IanW said:
Another alternative is to dip the springs in visco elastic fluid that solidifies its surface in the air but maintains the visco elastic nature beneath the surface.
Click to expand...
Interesting idea. Have you tried this?

I use the tonearm mostly for speaker box vibrational info. Was surprised to find how noisy some speaker terminal mounting plates are
 
I have a suspended wooden floor over a void, I chose to damp the floor by fitting acrow props under each speaker.
Coupled with a reasonably basic isolation of the speakers from the from resulted in a large reduction in floor activity.
 
sailor said:
Interesting idea. Have you tried this?

I use the tonearm mostly for speaker box vibrational info. Was surprised to find how noisy some speaker terminal mounting plates are
Click to expand...
Experiments being conducted in this area over the next few weeks.
 
I tested adding more isolation
Currently my TAD E1 speakers sit on 4 x Black Ravioli on Townshend stands
My TAD D-1000 sits on Black Ravioli on top of an Electric Beach board on top of Black Ravioli sitting on a Creaktiv rack
Plus
Granite placemats on top of all kit except the speakers as they have a curved top.
A wrist weight and cloth/(sand ?) doorstops on them

Tested a few times and proven to sound better

More to come maybe so it will look even crazier
 
Is there such a thing as too much isolation? I suppose if we want as close to total isolation as we can get, someone needs to invent superstrong audiophile hooks and superstrong audiophile fishing line so we can suspend our loudspeakers from the ceiling, floating one inch above the floor... I wouldn't be surprised though if they then sounded completely different from what the speaker designer intended as my guess is that the designer iterated the design while it was firmly planted on a floor of some sort.

Thoughts on enoughness and toomuchness?
 
You must log in or register to reply here.


advertisement


Back
Top