Stiffening a floor

[Marc Alan Goodman]

So I recently finished construction of a studio in Brooklyn. The control room sits on its own cement slab, but the live room and iso booths (about 1200 sq ft total including the lounge and bathroom) are sitting on top of an unfinished basement. The exterior walls of the space are 12" of brick from the early 20th century, and we left them as is. The floor itself is one layer of 3/4" plywood, two layers of 5/8" cement board, one layer of 9/8" plywood with a inset aluminum sheet (for radiant heat flooring), and 3/4" engineered hardwood on top, all glued together. We've noticed that the floor still has a small bit of resonance to it (if you hit a kick drum you can feel it on the other side of the 800 sq ft live room), and to top that off our immediate neighbor is having some noise issues. I believe the sound getting in to the neighbors house is passing through the floor and then through our shared basement walls. My combination solution to both of these problems is to stiffen the floor somehow and then add a few layers of drywall to the bottom, which will have the added effect of pushing our basement towards becoming a finished, usable space.

My question is: what is the best way to add mass to this floor from underneath? The joists are unevenly spaced due to early 20th century construction. It's been suggested that I should use 2x10s screwed and glued in place as stiffeners, and then just seal them off with two layers of 5/8" drywall. Will this significantly lower the resonant point of the floor? Is there a way to estimate how much mass we need in order to change the resonant frequency? To add to our difficulties, the ceiling has a number of pipes running below it (including some of the studios technical wiring) which will make it hard to get a good seal. Thanks so much for any help you can give.

[Soundman2020]

Stiffening might or might not work, depending on what part of the spectrum the problematic frequencies are in, with respect to the natural resonant frequency of the floor.

Let me explain: The amount of isolation you get in different parts of the spectrum, is governed by different physical characteristics of the floor. There are four regions, each controlled by a different aspect. From low to high, the four regions are :stiffness controlled, resonance controlled, mass controlled, and coincidence controlled. This might help get a better idea:

four-regions-of-isoaltion--transmission_loss.gif

As you can see, only the very lowest end is governed by stiffness. That region occurs only below the natural resonant frequency of the floor, and from the way you described the floor I suspect that its natural resonance is pretty low already, so it is unlikely that stiffness is the problem. Only very low, sub-bass frequencies live down there, way below the lowest end of the kick. So stiffening your floor would have no effect if the problem you are experiencing is not in the region controlled by stiffness.

It is more likely that the problem is in the actual resonance-controlled region: the floor itself is probably naturally resonant at the frequencies being produced by your drums. If that is the case, then the options open to you are simply to add a lot more mass to the floor, or to damp it somehow. Mass would drive the resonant frequency down lower, which is what you want, and damping would reduce the effect of the resonance. Doing both would be ideal. There are ways you could do that, but there's an even more important underlying issue here: How much more load can your structure take? You already have a lot of mass on that floor, so you are probably getting close to the structural limit of what your joists can handle. Only a structural engineer can tell you for sure, after inspecting your place carefully, but it is likely that you don't have enough structural integrity to add the amount of mass that you would need.

However, here's the good news: you might not have to! It is very probable that your problem is actually impact noise, not so much resonance, in which case simply decoupling the drum kit form the floor would be a major improvement. You could do that by building an ordinary low-cost drum-riser: Lay a couple of panels of 2" OC-703 on the floor side by side, put a couple of sheets of thick plywood on top (two layers, 3/8", screwed and glued), then set up the drums on top of that. I'm betting that will make a noticeable difference.

Worth a try!

- Stuart -

[Marc Alan Goodman]

Very, very interesting. The structure itself was actually built as a factory. The joists are nominal lumber around 10"x3", spaced an average of about 16" on center, so I think it can hold a bit more weight, and I have a structural engineer who can give me more exact numbers about what's safe.

I suppose my question now is: How much mass would I potentially have to add in order to lower the resonant point of the floor significantly? Is it in the realm of feasibility that I could push it below human hearing, or am I just wasting money and time trying to deal with the problem by adding mass?

[Soundman2020]

I suppose my question now is: How much mass would I potentially have to add in order to lower the resonant point of the floor significantly?

A lot! Actually a HUGE amount.

Since that is basically just a panel trap (membrane over an air cavity) the actual resonant frequency is given by:

f = 60 / SQRT (M*D)

Where M is the mass (surface density), and D is the depth of the cavity. (both in metric)

So basically, to halve the frequency you need four times the mass...

So if you can figure out what the resonant frequency is right now, making your floor four times more massive would force the frequency down by one octave. That might or might not be low enough, but it sure is a lot of mass!

This doesn't sound like a feasible solution.


- Stuart -