Floating floor questions! [Warning: long and incoherent]

[Dan_Fitzpatrick]

Greetings everyone, and thank you for an fascinating place to visit.

I've been studying this and other sites for a while now and have learned a lot. I'm a bit stuck on some of the theory though about floating floors and this forum seems to be the place to go for hardcore theory and math and stuff.   What I'm going to do here is describe MY understanding of things so far and ask some questions. Please correct me if my assumptions and understanding is wrong.

I'm contemplating building a room like in Scenario A in the attached drawing. I plan to use Kinetics KIP isolators type Q at 2" thickness. According to the  data, at 80 psi (720 lbs per KIP) the natural frequency would be 11hz (chart says 15hz, multiply by .71 for 2" thickness).

I estimate my room will end up around 8000 pounds (exact weight TBD) so i can use about a dozen of these KIPs (if i need to use more for structural purposes i can move to Type L and use less weight per KIP, but it will cost more.)

So 11hz is good. But i'm a bit suspicious of how easy this is because I know Paul had to do a bunch more math than this. But KIPs claim to have a simpler more straightforward behavior than Sylomer, apparently. For example, temperature isn't a factor apparently.

Now i think that the mass-air-mass resonance of the floor is a different number altogether that is not related to the resonance of the room on the KIPs. I may be wrong about this, but i don't think so. This is a totally different frequency and it is about 22hz according to my specs (Scenario A) and using Eric's MSMresonance03 excel file section "mass-spring resonance US"

OR -- gasp -- do the 2 frequencies somehow COMBINE?

the walls and ceiling also have their own mass-air-mass resonance which i don't know yet (as i'm on a Mac i can't use all the excel files), but it's probably in the 20's i imagine.

Now for a question: it has been said that floating a floor can make things worse or be a waste of time and money, so i want to know, is this true in my case. i suspect that one way to make things worse is if the resonance frequency of the room on it's spring was close to the Mass-air-mass frequency, then it could really be a problem (if it were close to audible range like if they were both 22hz; if it's 11hz for both then i suppose that's great). The two would reinforce each other and transmit a lot of sound into the slab.

the effect of the resonance of the room-spring-slab probably is a problem until 1.4 times that frequency, which is 15hz, still not near my mass-air-mass freq of 22hz. if it works the other way (22hz divided by 1.4) we are still ok, they just barely meet.

This whole theory of mine on how the floating floor can hurt things is just a guess, because i haven't seen any explanation but it's the best theory i've come up with.

Looking at the drawing again, i think that A must be better than B because in B both leaves are hard coupled at ALL frequencies while in A the leaves are only hard coupled around the mass-air-mass resonance frequency (the low bass unfortunately).

Scenario C seems to be the worst because we are trying to avoid getting sound into the slab, and here we have placed no obstacles in the way at all. Scenario B seems scarcely better. If you hard couple the inner leaf to the outer, does the airspace even have an effect at all?

But then i remember a bit of theory: with Scenario A or B i'll be transmitting a lot of 30hz or so sound into the slab (22hz * 1.4) ... does the theory say that from 30hz downward the airpspace will AMPLIFY these frequencies such that i would be better off with Scenario C?

if this is true, then it means that it isn't FLOATING the floor that is the problem, it's RAISING the floor at all (no dual-leaf floors? ever?), whether on springs or not. but then this would apply to ceilings as well ... it would be just as bad to have a dual leaf ceiling (assuming the ceiling outer leaf was attached to the rest of the building).

Apologies for this long and rambling post.  Can anyone tell me if i'm on the right track at all, or if i'm missing something major here.  Maybe someone can see what i'm missing by the idiotic questions i'm asking.

[In case anyone's wondering, i don't want to have a concrete floor, then the room will be just too hard to tear down later when i move in maybe 5-6 years. plus it just exceeds my threshold of budget and effort (yes paul i am in awe of you dude  ). ]

Thanks everyone for your help.

Dan  :)

[Bob]

I don't know what I'm talking about, so take this with a grain of salt.

Everything has a reasonance. Everything with an airspace has a resonance. TL is always worse at resonance.

But if you have a look at wall TL, you'll see that although it's worse at resonance (resonance dip near 50hz), that doesn't mean that it amplifies, actually it's the opposite it always lowers for any wall you're likely to build.

But the moment you actually put a spring in there, apparently it's possible to amplify to the point that it's actually louder in the next room.

the effect of the resonance of the room-spring-slab probably is a problem until 1.4 times that frequency, which is 15hz, still not near my mass-air-mass freq of 22hz. if it works the other way (22hz divided by 1.4) we are still ok, they just barely meet.

You had it the first time, by multiplying by 1.4.

Have a look at this (below). You'll see the resonance peak (B), and how it is still amplifying until it crosses the horizontal axis about 1.4 times that frequency (C) at which point the sound is just as loud in the next room as it is in the source room, and at higher frequencies the sound in the next room is quieter than it is in the source room.


from E-A-R's BASICS OF VIBRATION ISOLATION USING ELASTOMERIC MATERIALS


(I'm probably going to do Scenario C -- but because I have no room height. i.e. To build a room with a floating floor it would be less than 6'11", which is against building code where I live)

[Paul Woodlock]

Greetings Dan



Scenario A is the best with the following conditions I'll explain below

Scenario B is pretty pointless IMO

Scenario C is 2nd best


While you can probably reach the resonant frequencies you require with the current Scenario A design, I would advise making the floor MUCH HEAVIER.

Reason: Every load in the room will affect the resonant freqeuncy. Including people instruments, gear etc. Do you really want to weight all of that accurately? Nope! Furthermore live loads like people and instruments can change the load on the floor from day to day.

If you build a lightweight floating floor, the ratio of the dead load v. the live load will be LOW. Meaning the resonant frequency can easily change by a lot when more or less live load is placed upon it.  If the floor is designed near the limits of the elastomers properties the system could 'bottom out' with too much load ( i.e an extra person or two ) shorting out any decoupling.

"Sorry, you can't come in the studio, you're too FAT!!!!"  :twisted:

If you use a real heavy floor, the ratio of dead loads to live loads will be high, giving you much leeway with live loads and tedious to calculate loads like furniture, absorption frames, etc.

My floor design allowed 2 tons of leeway for live loads i.e adding 2 tons on the floor only changed the Resonant Freq by 1Hz or so. That was with a 5" thick concrete slab.

Now I appreciate why you don't want a concrete floor, but there are other ways to get the weight..

You could insert a steel layer ( A grid of steel plates )  in your floor. And as steel is 3 times as heavy as concrete you could, for example, replicate a 5" concrete slab with 1.6" thick steel plate. YOU may not need to go that heavy of course.


Dan, I wouldn't worry about making things worse in your case. You seem to have a good grasp of the situation, and my advice above pertains to other issues rather than having a too high resonant freqeuncy.

The folks who have to worry about making things worse are those who stick a couple of sheets of plywood over joists over 1/4" neoprene strips, and end up with not only a resonant freqeuncy in the audible range ( even up to the low mids ), they also create a huge drum skin on the floor which can have negative results on the studio acoustics. They would haev been much better off leaving the floor as your Scenario C.

The other thing not to forget:

Obviously the loads on the edges of the floor will be MUCH HIGHER due to walls and ceilings. As no floor will be infinitely stiff, you need to compensate by using either more closely spaced blocks or blocks with a bigger load area. Or a combination of both ( as I did ). I divided my floor into ZONES and performed seperate calcs for each zone. The zones included main floor area, walls, walls holding up the ceiling, Soffit wall, Heavy speaker stand area.

Hoep that helps


Paul

[Paul Woodlock]

....


from E-A-R's BASICS OF VIBRATION ISOLATION USING ELASTOMERIC MATERIALS

....

great idea bob!

I dunno why I haven't linked to that image before. It'll save some typing ache

Paul

[Bob]

Paul

Scenario B is pretty pointless IMO

Why? Isn't it like a wall or ceiling?

What's wrong with:
h) 3/4" CDX plywood
g) 30# roofing felt or Green Glue
f) 1/2" CDX plywood
e) 30# roofing felt
d) 3/4" CDX plywood
c) 2x6 at 12" OC, line the cavities with 6-mil plastic and dry playsand
b) 5/16" acoustic matting or Sill Plate Gasket
a) concrete slab over earth/dirt

[Paul Woodlock]

Paul

Scenario B is pretty pointless IMO

Why? Isn't it like a wall or ceiling?

What's wrong with:
a) concrete
b) 5/16" acoustic matting or Sill Plate Gasket
c) 2x6 at 12" OC, line the cavities with 6-mil plastic and dry playsand
d) 3/4" CDX plywood
e) 30# roofing felt
f) 1/2" CDX plywood
g) 30# roofing felt or Green Glue
h) 3/4" CDX plywood

What TL benefits are there with that? ( in addition to the non floating room within a room - Scenario C )

Paul

[Bob]

Paul

What TL benefits are there with that?

Wouldn't it keep some noise out of the concrete, and from there from flanking ?

( in addition to the non floating room within a room - Scenario C )

Please rephrase. I don't understand what you mean.

[Dan_Fitzpatrick]

Much thanks for the quick responses gentlemen!

Bob, thanks for your feedback, I always find your posts informative and interesting.

Thanks for including that link & graph. After reading the article on EAR I realized i need to go back and read all those papers again (i have read some of them, but it's been a while and they make more sense to me now). Maybe if I did that I wouldn't be so confused

Paul, I was so pleased (and a little surprised ) that you think i know what i am doing! Reading your thread has been a HUGE education for me. I don't think it's a joke at all when people suggest it should be made a book. It really should be (of course making all that paper DIY would be a chore ...  :lol: )

one thing i was considering to add mass to the floor is adding a layer of concrete stepstones that they sell in my part of the world ... sort of 1.5 x 12 x 12 inches (or so) with little stones embedded on top. that would probably help a bit with the dead-live ratio. don't know the exact weight or dimension off hand. or i could bite the bullet and pour perhaps 2 inches of concrete (mixing it myself), not in one solid layer but in a grid of separate wood-formed squares say 3x3, that way i could do it a little at a time and also i could get rid of it without needing a jackhammer. i have a rough idea that would be around 2 dozen bags but it would be possible. (my floor is about 10 x 12ft.)

one question i still have ... i know my resonance for the room-spring-slab system is good, and you suggest that the floor-air-slab system is adequate as well although it could be better. if i did add the 1.5 or 2 inches of concrete to my floor that should (as well as raising my dead-to-live weight ratio) also lower the floor-air-slab resonance ... can i just average out the density of the floor (5/8 MDF x 3 + 1.5 inch concrete) and plug those numbers into Eric's MSMresonance sheet (available on this site)?

there is a note in the spreadsheet that says do not use it for 2 massive layers. i don't think adding the stepstones would make it "massive" esp. because i think the stiffness is the key factor that changes the equation. but if i poured concrete in a grid of separate slabs would I still use the MSMresonance calculator ... or is there another excel sheet i need to use. I'd like to be able to figure out what HZ i'm getting for my effort ahead of time.

Thanks again,

Dan

[Howler]

I wonder what would result from pouring some sort of concrete/grout into 2'x2' forms,  maybe 6 inches deep. Place like big tiles on damping pads, and inject an elastic caulk between them. my room is 20x15, so it would take around 70 tiles for the room. Seems like they might resonate at higher ranges compared to a 20x15 membrane. They also might be more easily controlled. Seems like those are both good things.

Wonder what else you could use to make big tiles like that? How heavy each small membrane would need to be to dampen those ultra lows (set up in a matrix)?

hmmmmmm

Always talk of pieces on damping material. How nice it would be to roll out 1" or 2" pad under the top foor. Then put anyhing with the correct weight range on it.

At the new (old now) MSS Studios they suspended each section of the studio from giant shock absorbers. Two control rooms, dissimilar, and two tracking rooms (One larrrrge, one small), dissimilar. All the rooms possessed their own specific environments, connected by tiny halls. They built it inside the old Nat Guard armory where we used to practice basketball.

BTW, a fledgling but talented film company has bought the MSS facility and intends to use the studios as built for film and sound. Lots of room upstairs, and a huge full basement big enough for a tank. Most of the gear is gone, Neves sold, I think. But gear is secondary in a place that sounds like that. All the tracking snakes probably stayed, sooooo

[Bob]

Dan_Fitzpatrick:

This link is to a recent post by Eric about plywood over rigid rockwool over concrete.
SoundOnSound - Is a floating floor so important on ground floor

[Brian Dayton]

I don't know what I'm talking about, so take this with a grain of salt.

Everything has a reasonance. Everything with an airspace has a resonance. TL is always worse at resonance.

But if you have a look at wall TL, you'll see that although it's worse at resonance (resonance dip near 50hz), that doesn't mean that it amplifies, actually it's the opposite it always lowers for any wall you're likely to build.

But the moment you actually put a spring in there, apparently it's possible to amplify to the point that it's actually louder in the next room.

[

this assertion that there is a fundamental difference between a floor resonance and a wall resonance due to the presence of a spring is not correct.  

air is a spring, resilient channel is a spring, and furthermore (as i know i've seen at least Eric comment on), the air effect must also be taken into account for a floating floor for calculation of MSM

the core difference between what you see in those dramatic spring-mass isolator graphs showing boost at resonance and in a TL graph showing positive TL everywhere is this:

the TL graph compares airborne sound on one side to that on the other, hence the effect of mass (or mass/size, rather) IS TAKEN INTO ACCOUNT, and the "boost" due to resonance is LITERALLY AND IN FACT the dip below mass potential for that wall.

Eric, correct me if i'm wrong (but despite the cocktails that have lent ethanol to my bloodstream, i think my comments are in good form).


THE GRAPHS YOU SEE FOR SPRING MASS ISOLATORS ARE THE MAGNITUDE OF VIBRATION RELATIVE TO A RIGID CONNECTION.  

To make a reasonable comparison to a TL plot one would have to draw mass-law next to the actual TL curve and then you would see THIS EXACT SAME EFFECT.  enormous boost at resonance, decoupling above there, no effect below.


THERE IS NO DIFFERENCE, a spring resonance is a spring resonance.

for additional unique peek at this, take a look at the impact noise levels for these from IR-811

1.  raw 40mm concrete/15mm OSB
2.  as above with these
     A)  resilient channel and no insulation and single drywall
     B)  as above, add insulation
     C)  as A), add second drywall layer
     D)  double drywall + insulation

not the values at 50hz, that IS RESONANT BOOST

it has long, long, long fascnated me (and this came up in a thread at RO this week) that people panic about this resonance in a floating floor but simply pretend no problem exists in walls.  IT'S THE SAME PROBLEM, same scope, same magnitude

mass WILL BE A FACTOR on a floating floor as it is in a wall and there is exactly the same potential for a floating floor to actually amplify from one room to the next as there is for a wall.

finally, FWIW, the TL measuring process could not, i don't think, measure negative TL.  however, build a wildly undamped lightweight structure, put it in your test opening, measure noise levels in recieve room.  repeat immediately after removal with open hole.  you can demonstrate what is negative noise reduction.

but never could you measure negative TL (i don't think) with the existing standard, and indeed <10dB is hard to reach.

Brian

[Brian Dayton]

also, the wild boosts at resonance shown in those graphs are not likely to transpire (boost WILL transpire, that's not my point)

for example, steel as a material has damping of perhaps 0.001, essentially no damping, and predicted boost at resonance would be immense, enormous.

however, there is no way in bloody bejebus that .001 dampin gwill be seen in a realistic spring-mass system based on a steel spring.  actual constructions always have higher damping than the materials upon which it is based.

drywall, for example, has damping of ~.005 to .01 or so, but the damping factor of a wall might be anywhere from 0.1 to 0.4 (or whatever, wildly dependent upon variables)

a green glue sandwich has damping of maybe 0.6, a green glue wall may exhibit damping of >1

because damping is higher (due to a million reasons including friction, bending of fasteners, air leaks, blah blah blah) in real assemblies than in raw slabs of material, you have to elicit a MAGNANIMOUS change in damping to accomplish alot.  

Eric once commented way back in the archives on this, on the impact of damping, and observed that to accomplish alot you must make large changes in total system damping.  


i fear i'm getting away from my original point.  my point was this:

1)  your concrete or wood slab has mass, hence it will have some capacity to resist vibration from airborne sound
2)  the boost at resonance of your floor is RELATIVE TO THIS MASS RESISTANCE, and NOT RELATIVE TO ZERO RESISTANCE, and negative TL across a floating floor is ABSOLUTELY NO MORE LIKELY THAN ACROSS A WALL

however, worsening the performance RELATIVE TO NO FLOATING FLOOR is INNATE AND UNAVOIDABLE


and this is exactly, exactly, totally seen in all cavity wall assemblies not featuring wild levels of damping.  look at IR-761 and the dip below 2x4 wall performance for steel studs, resilient channel, etc.

extend that data to 30hz, and you'd see it for double studs too.  THIS IS THAT RESONANT BOOST DUE TO SPRING-MASS

just because TL is not negative doesn't mean there is no boost

B

[Bob]

Brian:

this assertion that there is a fundamental difference between a floor resonance and a wall resonance due to the presence of a spring is not correct.

As soon as I wrote "But the moment you actually put a spring in there, apparently it's possible to amplify to the point that it's actually louder in the next room. " I immediately popped back to the first line and typed "I  don't know what I'm talking about, so take this with a grain of salt. " since it looked odd, like physics changed somewhere between the two sentences.

I suspect that I'm still missing something though, because

just because TL is not negative doesn't mean there is no boost

doesn't make sense to me. I'm not sure how a boost/amplification (receiving room louder than source room) would appear on a TL graph, since I think negative dB which the TL graph shows would still not be a boost/amplification, just incredibly quiet.

the TL graph compares airborne sound on one side to that on the other, hence the effect of mass (or mass/size, rather) IS TAKEN INTO ACCOUNT, and the "boost" due to resonance is LITERALLY AND IN FACT the dip below mass potential for that wall.

Makes sense. Had to read it twice though. Helped to know what you meant by the "mass potential for that wall." (i.e. the mass law for walls: something like, doubling the mass raises the TL straight line by 6db, and that line is sloped so that it's less TL for lower frequencies).

THE GRAPHS YOU SEE FOR SPRING MASS ISOLATORS ARE THE MAGNITUDE OF VIBRATION RELATIVE TO A RIGID CONNECTION.  

That makes sense too. Had to think about it a bit.

also, the wild boosts at resonance shown in those graphs are not likely to transpire (boost WILL transpire, that's not my point)

The E-A-R graph (above) shows both a big peak (little damping), and a smaller peak (much more damping) under it.

2)  the boost at resonance of your floor is RELATIVE TO THIS MASS RESISTANCE, and NOT RELATIVE TO ZERO RESISTANCE, and negative TL across a floating floor is ABSOLUTELY NO MORE LIKELY THAN ACROSS A WALL

This means to me that there is a dip (less TL), but not an amplification in the next room. At least not nessessarily an amplification in the next room.

however, worsening the performance RELATIVE TO NO FLOATING FLOOR is INNATE AND UNAVOIDABLE

This also means there is no amplification in the next room at any frequency, but still the floating floor is not as good as not building the floating floor at the resonance frequency and +- a bit, so why bother building it.

Thus the floor resonance warnings that Paul and Eric keep posting.

[Dan_Fitzpatrick]

Brian, thanks for your comments, they have been pretty helpful although i haven't understood everything.

however, worsening the performance RELATIVE TO NO FLOATING FLOOR is INNATE AND UNAVOIDABLE

can you elaborate when you have time  :)  if floating the floor unavoidably worsens performance, why?

thanks!

Dan

[Bob]

Brian:

however, worsening the performance RELATIVE TO NO FLOATING FLOOR is INNATE AND UNAVOIDABLE

The additional floor layers represent additional mass too, relative to no floating floor. So the plus would be the +6db of TL if you managed to double the mass (i.e. same thickness of concrete would be required, or the same mass made out of steel plates would be thinner), and then subtract the resonance dip effects from the TL.
Since the resonance dips in IR761are often around 10dB to 12dB, I think it's plain that the resonance dips may be domanant of the two (mass vs resonance), hence your "INNATE AND UNAVOIDABLE", unless one were to add 3 times the thickness of the concrete slab as the floating floor (i.e. if a 6" slab, then an 18" thick slab floating above it) in which case the mass law's 12dB would be higher than the 10 to 12 dB resonance dip.

[Brian Dayton]

hey fellas

i should have added this:  however, worsening the performance relative to no floating floor is unavoidable IN THE RESONANCE REGION.  overall performance may well be better, or far, far better, depends on where that resonance is.

i'll follow up in the morrow, sleep beckons my battered brain just now

Brian

and bob:  you're posts are never bad, and if my "that is not correct" sounded at all foul, it wasn't meant to be in the slightest

[Dan_Fitzpatrick]

worsening the performance relative to no floating floor is unavoidable IN THE RESONANCE REGION.  overall performance may well be better, or far, far better, depends on where that resonance is.

Hi Brian,

Is that the resonance of the room-elastomer system or the mass-air-mass system ... i AM right to say these are two independent numbers right?

in other words, can my 11hz room-elastomer freq (Scenario A, top of thread) make the floor isolation at 22hz (the current planned floor's mass-air-mass frequency) worse. and, if i did add mass to the floor and lowered the MAM frequency to say 19 or 18, could that be a worse situation considering i have the 11hz room-elastomer frequency now closer to overlapping the MAM frequency.

how does the room-elastomer frequency affect the mass-air-mass frequency, or DOES it affect it. or do they affect each other?

any insight greatly appreciated, this has been a huge help already tho everyone.

thanks!

dan  :D

[Paul Woodlock]

Paul

What TL benefits are there with that?

Wouldn't it keep some noise out of the concrete, and from there from flanking ?

I may be wrong, but all I can see is it would increase very slightly the TL of the high freqeuncies which would already be completely inaudible outside  even without the room within a room.

( in addition to the non floating room within a room - Scenario C )

Please rephrase. I don't understand what you mean.[/quote]

Sort of explained above. I mean that I don't think it would give any practical improvent over Scenario C, and apart from losing room height is an awful lot of time and money.

Again, I may be wrong, but that's what my guts tell me.




Paul

[Howler]

I can't quite figure from your posts Mark. Is this to be built on a concrete pad? Is the pad on ground? I've seen many studios, killer studios, some studios that cater to LOUD bands, that were built on pads with no floating floors, and I cannot see for the life of me how they would have profited from building another floor atop the pad. If flanking sealing is taken care of correctly at the sill, there is NO leakage. Unless you plan to use the floating floor as a tuning device, in my mind it's pointless. And it seems that concrete on ground is more passive than any floating floor could ever be.

This ain't math talkin......

"If you ain't got mosquitos, don't stir the pond"

[Bob]

Howler:

If flanking sealing is taken care of correctly at the sill, there is NO leakage.

I agree. But how do you do that.

Cutting the slab isn't viable everywhere. Some places have insects. Some places have soft soil that lifts and sinks. Some places the slab is related to the foundation. Some slabs are built on top of stuff that assumes that the slab is supported by the foundation or else it'll tilt (tilt will shatter drains), or when more weight is added to it it will settle more.