avatar

thomas northward

Silverado

Posts: 139 Member Since: 10/02/2011

Lead

Nov 5 11 10:21 AM

Tags : :

This is an area where a lot of mistakes are made, and it's never good news when a designer / acoustician / structural engineer comes into a studio because there is a noise transmission issue and the conclusion is... "We have to tear it down".

This is something we see every year. Often in DIY spaces - but sometimes in professional spaces that were somehow modified along the years and where the conditions for the floating floor to function have been altered.

You cannot float properly on:

- Rockwool
- Fiberglass
- Any type of foam - with 2 exceptions
- Tires
- Any sort of rubber
- Hockey pucks
- Tennis balls
- Matresses
- Your grand ma.

Why: because none of these behave as a proper spring system, with a known behavior that can be calculated. Some like Rockwool or foam once loaded will stiffen and the tiny bit of elasticity you had the first days will disappear really fast and transmission of the LF and LMF will be back with a vengeance. All these are good for is attenuating footstep noises, if you're lucky.

What you need to achieve proper floating is:

1 - A Mass-Spring system that can be properly calculated - so a system you have a clear control over.
2 - Products that are designed for this exact task (I will name a few) and come with a clear guide  to their behavior under load - which is often the result of intense lab testing and quite a few real-life tests.
3- Seek the lowest possible natural frequency for your system. Never exceed 10Hz within a studio design.

What's a mass-spring system.

What's the natural frequency of a spring system:

It can be seen as a single degree of freedom mechanical behavior. It's basically the spring's natural oscillating frequency, and is dependent on mass and stiffness. Having this information is absolutely essential, and should always be communicated by the designer every time there is a floating floor calculation.

f(0)= (1/(2*Pi))*sqrt(K/M)

with f(0)= Natural Frequency in Hz
K= N/m
M= Mass in KG or daN (about equivalent)

The rest of the calculation for a floor is a bit more complex, but if some want then I will go through it at a later stage.

This natural frequency information is very important because it will determine both the areas where:

- transmissibility is >1, so when the floating floor system will actually resonate and transmit more to the structure.

- transmissibility is <1, so the area where the floating floor system is efficiently decoupling.

The natural frequency being the resonant frequency of the system, this is the area where transmissibility is >1. Therefore this must be as low as possible. The transition to the efficient decoupling area is gradual and varies with the type of spring system used.

As a rule of thumb, a floating floor should have at least a 90% efficiency (decoupling value) at 20Hz to make sure it is efficient over the usual musical bandwidth of 20Hz-20kHz.

Which means you should always keep your f(0) under 10Hz.

Quote    Reply   
avatar

thomas northward

Silverado

Posts: 139 Member Since:10/02/2011

#2 [url]

Nov 14 11 7:09 AM


Ooops, my bad.

In Europe:

- Sylomer    
- BSW Regufoam
- Products from AMC Mecanocaucho (note: we work with this company)    

         - Akustik + Sylomer
         - Vibrabsorber (/+Sylomer)

In North America, if I can't work with any of these 3 above that I know well, I will work with

- Mason Industries which have very good products,

and sometimes with

- Kinetics Noise, but to be honest I am not a fan of their KIP / RIM system and I avoid these if possible.


Thomas Jouanjean Northward Acoustics

Quote    Reply   
avatar

strangeandbouncy

Platinum Blonde

Posts: 2,150 Member Since:02/02/2011

#3 [url]

Nov 14 11 7:45 AM

Hi Thomas,


   that is most useful indeed! I am not sure I would be so forthcoming with such useful information!


   bless you, kind Sir!


     kIndest regards,


      ANdyP

Ruh Roh . . . . .

Quote    Reply   
avatar

burp182

Gold Finger

Posts: 316 Member Since:03/03/2011

#5 [url]

Nov 14 11 6:11 PM

Thomas,
What about the Auralex U-Boat isolators? I know of two rooms that used them as the foundation for their floors and they seemed to get good results. There has only been about a 4 year time span but no one has indicated diminishing results yet. They are certainly inexpensive and easy to use.
Thoughts?

Carl Dito

Quote    Reply   
avatar

thomas northward

Silverado

Posts: 139 Member Since:10/02/2011

#7 [url]

Nov 15 11 7:44 AM





Thomas,
What about the Auralex U-Boat isolators? I know of two rooms that used them as the foundation for their floors and they seemed to get good results. There has only been about a 4 year time span but no one has indicated diminishing results yet. They are certainly inexpensive and easy to use.
Thoughts?





-burp182






I would not recommend these. They don't function down in the LF where all the big issues are  - but surely do have a positive effect on impact noise transmission, to a certain extent. So there is a plus, although not significant for us as the most important part is left out.To put things in perspective, the lightest REGUFOAM density is the 150 and it takes 10 to 15kN per m² (1 to 1.5 tons per m²) for it to reach it's ideal load. We can reach a proper natural frequency with a lighter floor if we reduce the amount of Regufoam and place it a certain way (it's seldom that there is more than 20% of the total floor surface covered by this type of decoupling interface) so we can augment the load per mm² on the Regufoam and reach it's optimum frequency. 90kg/m² of floor weight is the minimum usable if we push the load concentration on the Regufoam and yet keep a safe structure. These 90kg/daN will rest on a small and carefully spread area of Regufoam and load it to the equivalent of 10 to 15kN/m² - or since these systems work in N/mm², 0.010 to 0.015N/mm².

The issue with these 'light' floors is that they remain a bit bouncy. So for a Mastering studio in a home it's fine, but not for a drum booth for example, where the movement will be perceived.

Another problem is the fact that if you want to properly float, then your walls (and ceiling) have to rest on the perimeter of that floor too - so they are decoupled too. Which creates very heavy local loads, called section loads because they usually apply high pressure near a wall/floor connection in the foundations / load bearing structure, and products like U-boats can't handle these high very local loads properly. With the lightest Regufoam floated floor we have done, we reach up to 4kN (400kg) of pressure per mc' (on a strip of 0.3 m², so 300/1000mm) which is already the equivalent of 13kN / 1.3 tons of pressure per m².

A normal European house is at the limit of overload if the studio is not on ground floor (they support on average 3.5 to 4kN/m² of usable load per level, a bit more on ground floor) and a North-American wood-based house has likely already collapsed, unless on the ground floor with a proper concrete foundation and there is no basement.

So U-Boats are designed for very light floors in comparison (maybe 45kg/m², with no real box-in-box possible). These light floors have a high natural frequency, over 10Hz, and because of their structure usually also act as some kind of resonator - which can be a positive thing, but most of the time isn't. And there is no control over the behaviour under load, it's clearly a shot in the dark, which a design company like ours simply won't allow. We want control and work with what gives us control.

We have to be control freaks if we want things to click from day 1.

Not to say that U-Boats don't have a use where budget is limited, and don't help at all. But it's clearly not a professional system.

I'll post a PDF example of a current floating floor design on big springs (fun, we don't get to do a lot of these!) to show the kind of loads we are dealing with in pro environments where very strict criteria / performance levels have to be met.

Hope this helped.




Thomas Jouanjean Northward Acoustics

Quote    Reply   
avatar

tom eaton

Platinum Blonde

Posts: 1,335 Member Since:25/01/2011

#9 [url]

Nov 15 11 10:56 PM

Thomas... great info.  How do you account for changing uses on floating floors such as a grand piano showing up on a spot that usually only holds a 120lb lead singer?  With such fine calculations I assume there is a load range for best floor performance, but in a large room where you could have anything from a single guitar amp to a large choir how do you maintain the desired float performance?

Thanks!

tom

Quote    Reply   
avatar

LeforaGuest

New Forum Friend

Posts: 0 Member Since:25/05/2017

#10 [url]

Nov 16 11 4:59 AM

In reply to Tom question, and considering that I'm anything but an expert, if you consider the weight of the entire structure to be floated (in the order of tons), a grand piano or a baby singer would make the same difference to the load seen by the floating material, that is zero.

As soon as you start talking about lightweight floated (or better decoupled) floors, the load you impose (let's call it dynamic 'cause it keeps on changin') will probably make a difference in the resonance frequency of the floor. More in the neighborhood of various Hertz.

Thomas am I right or completely off the mark?

If you consider that a 10cm concrete slab is 2500Kg/sqm, and you probably need more than that to support a completely floated room (including walls sitting on upmentioned floor and ceiling supported by walls), you see that putting a 700Kg piano on a floor that is supposed to carry 20tons would make little difference to its perfomances.

At least that is my understading.

PS: Thomas, if you could take a look at my control room rebuild thread down in the Home Studio section, I'd love to have your input on what we've done. If you remember the conversation we had many moons ago via e-mail you'll probably remember the issues we had atthe studio too. Thanks.

Quote    Reply   
avatar

johnr

Platinum Blonde

Posts: 1,561 Member Since:25/01/2011

#11 [url]

Nov 16 11 4:25 PM


If you consider that a 10cm concrete slab is 2500Kg/sqm

-judah

Ronnie, I think you might have a stray zero there. 2500kg/cubic metre is a closer figure (and that's on the high side), which would make a 10cm slab 250kg/sqm. I suspect your general point is still valid in many situations.

Quote    Reply   
avatar

thomas northward

Silverado

Posts: 139 Member Since:10/02/2011

#12 [url]

Nov 16 11 5:28 PM

What Ronnie said is about right.

An average professional control room - with a fully floated bunker shell usually weights anywhere between 50 to 90... Tons.

A live room being usually bigger, weights very often way over 120 tons. A grand piano or choir is peanuts in terms of overall load modification.

But it's indeed in the case of a heavy bunker. So the static load is high and the dynamic load is pretty much negligible. And extra load actually is beneficial as it would further lower the (already low) natural frequency.

What happens with lighter floors (although we always check that for heavier rooms no matter what) is that we look carefully at the optimum static load and the range of dynamic load that can be used. A normal room should be able to withstand 350daN/m² of extra dynamic load (so anything that isn't a permanent load like people, pianos, equipment etc) without the "spring" bottoming-out before that extra load capacity is reached.

Like with big rooms, adding dynamic load will increase the performance for a while as the natural frequency gets lower, until the system stiffens where you can see the natural frequency go up again.

See example graph attached.

With the lighter floor like the "lightest floors" we've done with around 90kg/m² of weight where there are no walls resting on it, this calculation becomes critical and we have a very narrow range to play with. Therefore these floors can only be used in mastering rooms or some mix rooms and provisions are made where equipment is to be installed. There is no way you can use this to float a live room properly. For that we need to move into a different load range to attain a safe & usable dynamic load range.

We do these calculations in-house during the design phase but they always end up on a structural engineer's desk. A good structural engineer (that understands how all this works) to check calculations is essential. And also to check if the building is 100% capable of supporting such loads.

Carefully planned ventilation is always essential to all floors too.

As a side not, these may seem like expensive features, but they seldom represent more than 10% of the total cost of the studio build, mostly depending on how easy it is to work in the space wrt load repartition. It isn't expensive to float properly. But it takes detailed calculations and... A building that will allow this.

I've also attached an example of load repartition on a spring decoupled floor sub-structure to give an idea of how it looks like. Each number is for a separate spring system. Not as linear as most would think it is... This is for a Control Room and can allow an extra load of 350daN/m² (so a total extra of 14.7 tons in this particular case!). Which is standard and... more than enough. Natural frequency of that floor is ~3.5Hz. Some of the springs take over 2 tons of pressure on their own, some close to 2.5 tons. This CR weights over 66 tons empty of equipment.

The more you dig in this small area of studio design, the more complex it gets as there is a multitude of little things to think about.



Click here to view the attachment
Click here to view the attachment

Thomas Jouanjean Northward Acoustics

Quote    Reply   
avatar

LeforaGuest

New Forum Friend

Posts: 0 Member Since:25/05/2017

#15 [url]

Nov 17 11 4:31 AM


Ronnie, I think you might have a stray zero there. 2500kg/cubic metre is a closer figure (and that's on the high side), which would make a 10cm slab 250kg/sqm. I suspect your general point is still valid in many situations.

-johnr

They call me fast fingers crappy brain for a reason. Anyway, yeah, my bad, I was thinking about cubic meter and put it all wrong.

Quote    Reply   
avatar

thomas northward

Silverado

Posts: 139 Member Since:10/02/2011

#17 [url]

Apr 14 12 8:52 AM

Fascinating.
What, Thomas, would your opinion be of a large Whisper room (tm?) floated upon six large 5/8 filled truck innertubes?

-bockaudio

I'd go with pieces of Sylomer ;)

It's not that expensive.

Thomas Jouanjean Northward Acoustics

Quote    Reply   
avatar

thomas northward

Silverado

Posts: 139 Member Since:10/02/2011

#18 [url]

May 11 12 4:30 AM

We recently had the opportunity to work with springs, which doesn't happen often. It requires a combinations of load capacity, ceiling height and... Budget for it to click together.

This is a fun project. clients are great to work with too!

So, here's an example of a floor with estimated natural frequency of 3.5HZ to 4.5Hz.



Click here to view the attachment
Click here to view the attachment

Thomas Jouanjean Northward Acoustics

Quote    Reply   
avatar

stitch

New Forum Friend

Posts: 2 Member Since:12/07/2012

#19 [url]

Jul 12 12 6:19 AM

Wow Thomas! Thanks so much for this informative thread! Unfortunately they don't teach this stuff at school! 

Would it be correct to assume that you can properly float a lighter construction (wooden joist/gypsumboard construction of 65 kg per square meter) if you use thinner stripes of sylomer/regufoam? 

Quote    Reply   
avatar

sebastian

New Forum Friend

Posts: 3 Member Since:13/07/2012

#20 [url]

Jul 14 12 1:58 AM

Stitch, it can be "properly" floated, but only for a narrow range of load. The problem is that the dynamic load (that is, the variation in load as people and equipment come and go) varies so much that it is not possible to design the spring elements to work effectively over such a large range of load. With a heavy concrete bunker, the total load on the springs (weight of the structure plus the dynamic load) does not vary much - say the bunker weighs 20 tonnes, then even if 2 tonnes of people come and go, that is a load change on the springs of only 10% and the springs will still be working in the desired load range. But if a lightweight floor weighs 1,000kg and 1,000kg of equipment and people come and go then the load has changed by 100% - it is not possible for springs to accommodate that range of load without being bouncy and unworkable.

Quote    Reply   
Add Reply

Quick Reply

bbcode help