The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Sanity-check needed on idea for timber floor impact noise reduction Dynamics Processors (HW)
Old 4th August 2014
  #1
Gear Head
 

Sanity-check needed on idea for timber floor impact noise reduction

I'm doing a large-scale renovation on a house, which includes building a studio, so I want to stop impact noise from the first floor (suspended timber) from coming down into the ground floor.

First of all, this is what I've read about the sounds that I need to reduce...

Airborne - sounds that travel in between the joists, such as people souting, music.

Impact - e.g. sounds from someone walking on the floor, their steps vibrating the floorboard which in turn vibrate the joists.

There are three things that make this difficult in my case.

1. I want to reveal about 3 inches of the bottom side of the joists, to be visible from the ground floor. So, for this I will be boarding between each joist. This is no problem in itself, but it will prevent me from creating a false ceiling to aid soundproofing.

2. On the first floor, I don't want to raise the finished floor by that much because of headroom. I can't really go more than 40mm.

3. I will have underfloor heating installed on the first floor - a water pipe system. Because of the headroom issue of point 2., I will be placing the pipes between the joists, and using biscuit screed to absorb the heat, with the floorboard going staight on top. I've heard that biscuit screed is a much better option than having thermal metal plates just underneath the floor boards.

Now, the Airborne is no issue, because I'll just place some sort of insulation/acoustic boards between the joists. It's the Impact treatment that's more of an issue, because I don't want to use batens to raise the floor too much, and also I can't have a acoustic underlay (e.g. InstaLay 65 Under Screed Resilient Layer (E-FC-13)) because it would have to sit on top of the screed and therefore block the heat from leaving the UFH pipes.

What I was thinking was, to have the acoustic underlay (say 25mm or whatever) there to stop the joists from touching the floorboards, but not to stop the heat, so I was thinking I could cut it into strips just wide enough to cover the tops of the joists. I would then bring the insulation higher than I would otherwise (because in effect the joists are now slightly taller), and fit the UFH pipes as normal, and fill it will the biscuit screed as normal, and then put the flor board on top. Here is a picture of what I mean...

https://drive.google.com/file/d/0B10...it?usp=sharing

This is a cross-section - on the left and right are the joists, which everything else self-explanatory.

Could this work? Or is it a bad idea, and if so why? Do you know of any other method/system that could help me?

Thanks.

Last edited by meirionwyllt; 4th August 2014 at 09:49 PM.. Reason: link for image did not work - changed to regular hyperlink now
Old 4th August 2014
  #2
Gear Guru
Pic

The picture is not visible. Trying editing it in again. Impact Transmission loss will need a decent resilient layer with the mass on top floating on it.
Regupol perhaps.
Also the Mass on top can be usefully increased by cement boards, plasterboard etc.
DD
Old 4th August 2014
  #3
Gear Head
 

Hi DanDan, thanks for replying. I've now changed the image hyperlink to a regular hyperlink (from my google drive), so it should work now.

Would a floor construction like that work d'you think? Using Regupol or similar in the position shown in black in the picture.
Old 5th August 2014
  #4
Gear Guru
No Contact

You need to fully float. No contact.
DD
Old 5th August 2014
  #5
Gear Head
 

OK, how about modifying it slightly, cutting more strips of the membrane, and gluing them to the sides of the joists too...

https://drive.google.com/file/d/0B10...it?usp=sharing

A lot of work, but would it do the job?
Old 6th August 2014
  #6
Gear Guru
Height

I don't know what biscuit screed is and what restriction you are working under. I would be inclined to install these on the joists http://www.auralex.com/sound_isolati...tion_uboat.asp and lay a layer of heavy board. Regupol on top of this, then your screed, floorboards, carpet.
Insulation and sealing underneath as you have it.
DD
Old 6th August 2014
  #7
Gear Head
 

If the extra headheight upstairs was something that I could negotiate on then yeah I would certainly do it the way you say, but I cannot, unfortunately.

Biscuit screed is just a dry screed mixture of sand and cement. It's dry so it doesn't set into a hard block and remains grainy.

Would it work as shown in the picture, d'you think?

I will check out the U-Boat.

Thanks.
Old 6th August 2014
  #8
Gear Guru
Builders

It should work. It would be better of experienced builders could pop in here. I am kinda theoretical.
DD
Old 6th August 2014
  #9
Gear Head
 

Hi, I've had a look at the U-Boats, but aren't they more for stopping sound from going up into a floor above? It's the other way that I want - to stop impact noise from upstrairs from reaching down into the studio. Could these U-Boats be placed on the top of the joists instead? I'm guessing that I need to stop the joists from vibrating in the first place.

Thanks.
Old 6th August 2014
  #10
Gear Guru
Symmetry

Sound Transmission is completely symmetrical.
What doesn't go up doesn't come down!
DD
Old 6th August 2014
  #11
Gear Head
 

OK great.

So you don't think it matters that the joists would get a chance to vibrate from impact (i.e. people jumping upstairs)?
Old 6th August 2014
  #12
Lives for gear
 
norton's Avatar
If you leave joist material exposed you will hear impact noise transmission. Plain and simple. There's no point in going through the effort to creates mini ceilings between joist cavities if you are concerned with sound transmission from above.
Old 6th August 2014
  #13
Gear Guru
Que

I am afraid i don't understand that question.
Let's work from the top down.
People jumping and walking about, footfall. The first improvement we might add would be mass. Then a resilient layer, then another layer of mass. The heavier and thick each layer of mass the lower the resonant frequency and the better the decoupling between above and below. Such systems are symmetrical if both sides are built equally. This doesn't apply in your case unless https://www.youtube.com/watch?v=OdQDXs75Ulo

DD
Old 6th August 2014
  #14
Lives for gear
 
Rod Gervais's Avatar
 

OK:

With people walking around the concern is impact noise - so we need to decouple the finished floor from the structure below......

With music from below we need to both decouple and add mass........

Norton is right when it comes to impact noise - and I would point out that air borne noise faces the same challenge..... the transmission of air-borne noise is not limited to the space between structural elements - the structure itself also acts as a transmitter......

In order for you to use the biscuit screed you refer to you are going to have to provide a structural element to your design that I do not see now........ I would not hold out a lot of hope of rigid insulation below the screed holding up over a 20 or 30 year period.

Building up above the floor becomes problematic due to the added insulation value each subsequent layer adds - which then decreases the amount of heat passing through to the space above.......

You don't mention what level of isolation you are really looking to achieve........ given your constraints I would not expect much when all is said and done.......

Rod
Old 6th August 2014
  #15
Gear Head
 

norton - No, I agree entirely, I have now given up on the idea of exposing the bottom of the joists, it would be counter-productive.

DanDan - Ah OK, sorry, I forgot that you said that I'd need Regupol as well. And when you say Regupol, do you mean...

Regupol 4515 Acoustic Underlay from CMS Danskin Acoustics
or
Regupol 6010BA Acoustic Screed Underlay from CMS Danskin Acoustics

I've looked into U-Boat and unfortunately my joists are too wide (2.25") to fit inside the U-Boat which is 1.5". Do you know of any similar product?

Lastly, why I would need decoupling above AND below the joists? I understand the Regupol side, but could you please explain how decoupling the bottom side would benefit me? It's just that I'm short of head height as it is.

Thanks!
Old 6th August 2014
  #16
Lives for gear
 
jhbrandt's Avatar
Quote:
Originally Posted by meirionwyllt View Post
I'm doing a large-scale renovation on a house, which includes building a studio, so I want to stop impact noise from the first floor (suspended timber) from coming down into the ground floor.

First of all, this is what I've read about the sounds that I need to reduce...

Airborne - sounds that travel in between the joists, such as people souting, music.

Impact - e.g. sounds from someone walking on the floor, their steps vibrating the floorboard which in turn vibrate the joists.

There are three things that make this difficult in my case.

1. I want to reveal about 3 inches of the bottom side of the joists, to be visible from the ground floor. So, for this I will be boarding between each joist. This is no problem in itself, but it will prevent me from creating a false ceiling to aid soundproofing.

2. On the first floor, I don't want to raise the finished floor by that much because of headroom. I can't really go more than 40mm.

3. I will have underfloor heating installed on the first floor - a water pipe system. Because of the headroom issue of point 2., I will be placing the pipes between the joists, and using biscuit screed to absorb the heat, with the floorboard going staight on top. I've heard that biscuit screed is a much better option than having thermal metal plates just underneath the floor boards.

Now, the Airborne is no issue, because I'll just place some sort of insulation/acoustic boards between the joists. It's the Impact treatment that's more of an issue, because I don't want to use batens to raise the floor too much, and also I can't have a acoustic underlay (e.g. InstaLay 65 Under Screed Resilient Layer (E-FC-13)) because it would have to sit on top of the screed and therefore block the heat from leaving the UFH pipes.

What I was thinking was, to have the acoustic underlay (say 25mm or whatever) there to stop the joists from touching the floorboards, but not to stop the heat, so I was thinking I could cut it into strips just wide enough to cover the tops of the joists. I would then bring the insulation higher than I would otherwise (because in effect the joists are now slightly taller), and fit the UFH pipes as normal, and fill it will the biscuit screed as normal, and then put the flor board on top.

This is a cross-section - on the left and right are the joists, which everything else self-explanatory.

Could this work? Or is it a bad idea, and if so why? Do you know of any other method/system that could help me?

Thanks.
meirionwyllt,

Rod basically summed it up when he said, "given your constraints I would not expect much when all is said and done." That's the honest answer.

What you mentioned above for AirBorne sound isolation will NOT work.

Probably the BEST thing you can do at this point, if you are stuck with this 'idea', is just lay some 3/8" or 8 mm Regupol 6010 and put a laminated floor on it. You will ONLY get about IIC36 from that and about STC-0 for airborne sound.

I do not recommend spending money or time without really knowing what you will get when all is said and done. There are plenty of designs that actually work and have testing data for them. I would recommend that you have a look at (ir811) NRC-CNRC Fire Resistance and Sound Insulation of Floors: Sound Transmission and Impact Insulation Data - (link is on my publications page)

There are no magic products, no physics defying voodoo.. well, not yet anyway. When it comes out, we'll have a look at the testing data and see.

Why not build your studio area on the ground floor? It's much easier, cheaper, and it just works.

Cheers,
John
Old 6th August 2014
  #17
Gear Guru
Phew

There you go. Two guys who actually know from extensive build experience.
DD
Old 6th August 2014
  #18
Lives for gear
 
Rod Gervais's Avatar
 

Quote:
Lastly, why I would need decoupling above AND below the joists? I understand the Regupol side, but could you please explain how decoupling the bottom side would benefit me? It's just that I'm short of head height as it is.
Decoupling the floor (in the manner described) deals (pretty much) with just impact noise from above......

In order for you to add enough mass above that decoupled floor to really mean anything (from the perspective of air-borne transmissions) the mass will circumvent your heating system when all is said and done....... besides which - you've already said that you can't add a whole lot over the existing floor height due to a lack of ceiling height..

Now - we do (when we have no choice) deal with isolation on top of floor assemblies rather than below....... but these assemblies then become very thick........

I did this in an apartment building over the top of a commercial space below - and it worked quite well for their purpose - which is normal living arrangements. (It was not possible to do any work in the space below)

So the folks upstairs could listen to music - watch TV - and not bother the folks in the business below........ the assembly in that case was about 2 1/2" thick over the top of the existing sub-floor.... and it would have been thicker than that if the owner was using a heating system in the floor......

Now, those systems can be designed to achieve the levels of isolation needed for (relatively) loud music - however it would take (much) thicker assemblies than described above.

The problem you face here is one of not only decoupling - but also mass........

When the floor joists pass direct through to the space above - the only mass you have (from the perspective of isolation) is the mass above the isolating member you're talking about here. (For the purpose of this discussion I am deliberately ignoring the air spaces between the joist here to focus on the joist themselves)

Now - when you add mass directly to the underside of those joists (or even in between) you WILL increase the level of isolation...... however the amount you increase it is limited by the transmission of sound to the space above directly through the floor joist.......

Once you reach the point where the joist itself is the weak point in the equation - you could add mass up to the point where the joist were ready to collapse and you would not increase the level of isolation between the spaces to any degree that would be meaningful.......

It's really that simple........

Rod
Old 6th August 2014
  #19
Gear Head
 

OK thanks for your advice, I was merely trying to exhaust all possibility of not adjusting the height - so, possibility exhausted.

What I need to know now are two things...

Do you know of any similar U-Boat products that could help me on the underside of the joists? U-Boat is too narrow for my joists.

And also, which of the two Regupol items above where you referring to?

Thanks.
Old 6th August 2014
  #20
Gear Head
 

Hey Rod, thanks for the detailed explanation - much appreciated.

One thing I'm not quite sure about throughout this thread is - what is meant by "mass" in this sense? I'm sure it would all make sense to me once I know this.

Thanks.
Old 6th August 2014
  #21
Lives for gear
 
Rod Gervais's Avatar
 

What would you be using U-boats for in this case (Not that U-boats have any real value anyway - just curious how you are picturing this)

U-boats are of little value period........ I remember speaking directly an engineer who was involved with the original design - and his thoughts were that it was not worth the effort/cost to have them tested because they would not fare all that well - and they were pictured as simply a "poor man's" way of getting a little sumthin sumpthin - better than nothing - but not a whole lot........


However the entire concept of the U-boat is to be able to rest a structure on a product that is sitting on top of another product......

You don't have enough height upstairs to build a new structural deck resting on the existing - and downstairs is a ceiling........ so even if you could invert the boats and use them - you would still have to screw through them to attach the ceiling - at which point they just because more useless than they would be if building a floor on them........

I find myself confused here........

To work below the ceiling you need to begin looking at things like resilient channel (RC) which will give you about a 6dB boost in isolation over what you would get if attaching drywall directly to the bottom of the floor joist - resilient clips with Hat channel (about a 6dB boost over the RC) or fully decoupled framing, which will get you about another 6db increase in isolation over what you'll achieve with the clips.

In all cases (of course) the increases related directly to the same amount of mass being installed below the assembly.......

Now - you are very limited to the amount of mass you can install when using RC (2 sheets of 5/8") ......... you get an increase in mass (up to 3 sheets of 5/8" if memory serves me) if working with clips systems (however need to be very careful with clip spacing otherwise things do not work well) - and are limited in the amount of mass only by structural capacity if working with an independent frame......... so it could be designed for 4 or 5 sheets if you wanted..

Rod
Old 6th August 2014
  #22
Lives for gear
 
Rod Gervais's Avatar
 

Mass is the "panel" side of things - so plywood is mass - drywall is mass - a concrete slab is mass....... etc.......
Old 6th August 2014
  #23
Gear Guru
Sinking Boats

Interesting thoughts Rod. I suggested U Boats on top of the joists supporting an added Mass layer of MDF or such. That is simply because I sorta trusted Auralex.
As we don't have data on them working or not (do we?) why do you suggest that they wouldn't do much or anything?

DD
Old 6th August 2014
  #24
Lives for gear
 
Rod Gervais's Avatar
 

Quote:
Originally Posted by DanDan View Post
Interesting thoughts Rod. I suggested U Boats on top of the joists supporting an added Mass layer of MDF or such. That is simply because I sorta trusted Auralex.

As we don't have data on them working or not (do we?) why do you suggest that they wouldn't do much or anything?
DD,

As I said above - I was told (by someone who worked in their engineering department at the time the product was developed - who was very involved in the testing of their products) that these products have never been tested......... this is first hand - not a case of "he said she said".

Having said that - even assuming a product was made that would fit over the joist directly (which these won't)- I have a huge issue (from the perspective of engineering) with simply sitting a bunch of plywood loosely on top of floor joist and expecting things to work over a long period of time..........

In fact - it would not be code compliant here in the states - the deck needs to be sheathed if the space is occupied - and there are specific fastening schedules that need to be adhered to, all of which require a direct connection of the sheathing to the joist........... and the minute you begin fastening through the boats to the joist below you screw up what little isolation the boats might have offered..........

As far as what I suggest.......... it's pretty simple - what they offer in the way of isolation is little at best......... if they were "the real deal" from that perspective they would have warranted testing. It isn't as if the cost of a single set of tests would have broken the bank when it comes to this company - they provide product testing on a regular basis....... reality is that any test results would have been poor (at best) - they knew this - and that is why they never had them tested........

I am not trying to bad mouth them for this - in fact it's a smart business move - but by the same token when the properties of a thing are unknown - when there is no test data to examine - it is impossible to predict (to any degree even remotely close to reality) if that product will help anyone succeed in their quest for isolation.

Rod
Old 6th August 2014
  #25
Lives for gear
 
avare's Avatar
 

Quote:
Originally Posted by DanDan View Post
Interesting thoughts Rod. I suggested U Boats on top of the joists supporting an added Mass layer of MDF or such. That is simply because I sorta trusted Auralex.
As we don't have data on them working or not (do we?) why do you suggest that they wouldn't do much or anything?
Consider for starters what is the effective load/deflection and bottoming out load. We have no information.

Well informed,
Andre
Old 7th August 2014
  #26
Lives for gear
 
jhbrandt's Avatar
Big +1 to Rod and Andre

One has to remember that companies sell things because they can.

When building an isolated structure - especially one that is to isolate both air borne and impact sound down to at LEAST 20 Hz - there are several very important things to consider if it is to be a successful isolation build;
(Rod, Andre... this is off the top of my head. So if I missed anything, please add to the list)
1. Frequency range of desired isolation.
2. Degree of isolation in Decibels per frequency band for air borne AND impact.
3. Use. i.e.; floor, wall, ceiling, etc.
4. Live and static load of #3.
5. The above 4 points will determine WHAT isolator to use.
6. Once the isolation product is chosen, the size and spacing of the isolator is determined by the load so that optimal isolation is obtained. i.e.; Both impact and air borne sound transmission is attenuated to the degree required by #1 and #2. This is done by engineering calculations for the material used. If the isolator is a spring, then 3.13 * SQRT(spring deflection in inches) = system resonance. Similar calculations are done with a rubber, neoprene, Sylomer, Sorbothane, etc.
-- AND you can NOT do any calculations without the DATA for the isolator.

So whatever you do, ONLY use products with DATA.
I don't think the engineers 'guessed' how they would shoot a rocket to the moon.. "Hey, Joe said that he heard down at the rocket store that they have a new module you can put in the boosters to keep them on course.." "Ok! great, let's get 10 of those and put Roy in the cockpit and shoot him to the moon."

In all honesty, we live and we learn. Thank goodness that the things that I have 'learned' haven't killed me. At my age, I really appreciate the value of planning and testing.. I'll say this, that my opinions CHANGE with new information and I stand by the scientific process. I've probably DONE what many who will read this have done. A lot of things. Not just acoustics and a lot of things didn't work. -- The more you get into this area of acoustics, the more you know - you don't know. LOL!

I can not emphasize enough that there are PLENTY of good products available that HAVE the data...
Use those.

Cheers,
John
Old 7th August 2014
  #27
Gear Guru
Das Boot

A trio of cruisers gunning for those U Boats.....Dive! Dive! Dive!

Thanks for the contributions guys, if I may address Rod first.

Good calls, I have never seen these things, perplexed to hear that they don't fit over joists. Your point re plywood floating around up there is well taken.
As I consider it a system of small areas or 'points' of resilient support would also allow the floor/ply to bend under foot. A large Kick drum.
Presumably when these devices are actually used one uses a rigid floor system of two or more layers bonded together both vertically and horizontally.
Overall, I hear you, good luck with that and so on.

Now to the more general issue brought by the three of you. How do you calculate or predict or have any idea of how this will perform without Data?
The text books and our Lab work when training deal with isolating motors and such which are idealised scenarios. The supporting boundary is massive and immovable.
In this case it is neither massive nor immovable, and I strongly suspect that we also have no Data on it's behaviour. I would expect the 'bottoming out' Data from the various favoured products to be somewhat applicable, but resonance?
Standards here are very different. Fire and thermal specs are high, but one can visibly see the deflection of a floor/ceiling. I don't see a way of getting resonance calculations into these real life scenarios.

So how to proceed? Well to be fair or pedantic perhaps, I would suggest looking again at Auralex. Some claim success with U Boats. https://www.gearslutz.com/board/photo...ally-work.html Let's see if Auralex have Data in other forms. e.g. Installation Guides. Ditto Regupol. I expect Auralex will at least have a suggestion as to how close to place the submarines to each other to address varied load predictions above. With the other materials it probably boils down to simple choice of thickness to achieve the optimum deflection under a range of (largely guestimated) loads.

In term of the OP's issue, if one wanted to address the footfall at all seriously I think the underfloor heating needs to go. Also the requirement to expose joists below. Then it becomes a regular soundproofing exercise. http://www.customaudiodesigns.co.uk/...tic-floors.htm

DD
Old 7th August 2014
  #28
Lives for gear
 
jhbrandt's Avatar
Quote:
Originally Posted by DanDan View Post
A trio of cruisers gunning for those U Boats.....Dive! Dive! Dive!
LOL!

Quote:
Originally Posted by DanDan View Post
Now to the more general issue brought by the three of you. How do you calculate or predict or have any idea of how this will perform without Data?
You do it one of two ways:
1. You measure the product yourself and test.
2. You don't use the product at all.

Quote:
Originally Posted by DanDan View Post
The text books and our Lab work when training deal with isolating motors and such which are idealised scenarios. The supporting boundary is massive and immovable.
In this case it is neither massive nor immovable, and I strongly suspect that we also have no Data on it's behaviour. I would expect the 'bottoming out' Data from the various favoured products to be somewhat applicable, but resonance?
Yes we do. Structural engineers work with this sort of data all the time. It can be determined by the span. But an engineer is required to go in and measure the sag and other factors to determine the strength, resonance, and load ability of a floor or ceiling section.

Quote:
Originally Posted by DanDan View Post
Standards here are very different. Fire and thermal specs are high, but one can visibly see the deflection of a floor/ceiling. I don't see a way of getting resonance calculations into these real life scenarios.
A great reference would the the one written by Norm Mason in 1966. ASHRAE Lecture - Controlling Vibration Problems in Sensitive Structures. - http://www.jhbrandt.net/ashlec.pdf
Page 4 explains this problem.

Cheers,
John
Old 7th August 2014
  #29
Gear Guru
Engineer

Thanks John, really great looking link. I am well aware of the higher engineering possibilities. Note however my comments begin with 'in this case'. I really cannot imagine an engineer at the OP site measuring deflections and so on. Thus as I said we have little or no Data on the supporting joists in terms of deflection, stiffness, resonance. So as parts of your link suggest, very wide goalposts, or as I said guesstimates. Generous over engineering by guesswork.
This is why I am suggesting that the manufacturers installation guides may be a realistic guide here. But apart from our banter which is very interesting technically for us, what about the OP's case? Any suggestion how to proceed?
DD
Old 7th August 2014
  #30
Lives for gear
 
avare's Avatar
 

Quote:
Originally Posted by DanDan View Post
Now to the more general issue brought by the three of you. How do you calculate or predict or have any idea of how this will perform without Data?
The text books and our Lab work when training deal with isolating motors and such which are idealised scenarios. The supporting boundary is massive and immovable.
In this case it is neither massive nor immovable, and I strongly suspect that we also have no Data on it's behaviour. I would expect the 'bottoming out' Data from the various favoured products to be somewhat applicable, but resonance?
The floor/ceiling partition has 2 parallel spring systems. The mass-isolator-mass system and the mass-air-mass system. The resonance of both systems can be calculated with the appropriate data.

To others reading this thread: The MAM resonance is ignored too often by amateurs trying to improve isolation. Below the MAM resonance, the isoaltion is same as a single homogenous structure of the same mass as the sum of the leaves.

Appropriately calculated,
Andre
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump