OOh! I know the answer to that! It's because you didn't sprinkle enough magical pixie dust on the plywood!(AKA why doesn't my floating plywood floor with 20 Hz springs and 2" air space not do anything for lows?}.
(I'll show myself out the door, now....)
OK, getting back into more serious things (because light-weight floated floors really are a joke!
) ...Steve, your comment was from the NRC article, where it says: "Adding resilient channels to one face of a single row of studs improves sound reduction considerably, allowing the sound absorbing material in the cavity to be effective." You added your own comment: "That seems to imply that the insulation will not be as effective without the RC."
It's a little cryptic, but when you take that in the context of the paragraph right above it, they are talking about a wall with NO decoupling first (just drywall on both sides of studs: "...gypsum wall board is solidly fastened to the wood studs both sides..."), as compared to a wall with RC on ONE side ("... adding resilient channel to one face of a single row of studs...."). The implication here seems to be that the insulation is not effective when the drywall is "firmly fastened", but is effective when it is decoupled. That ties back to the second paragraph in the article, which says: "... if the two faces are of the wall are not isolated from each other, sound absorbing material in the cavity is rendered ineffective".
Thus, they are not really saying that it is the RC that makes the insulation effective, but rather that it is the fact of decoupling (or not) that makes it effective, no matter how it is accomplished. RC is just one method.
However, there's still the issue that they didn't explain what they wanted to say very clearly.
The way I see it, is that they just did not word that very well: The insulation will still do what it always does (damping resonances in the cavity) regardless of whether or not the drywall is decoupled. However, the fact of having a solid mechanical connection between the faces has a negative effect that counteracts the positive effect of the insulation. Not decoupling doesn't somehow prevent the insulation from damping: it just means that the damping isn't very useful, because the coupling is really bad.
Simple analogy: you can pour water into a bucket with huge holes in it all day, but the bucket will never get full because of the holes. That doesn't mean that you are not pouring in water like crazy! It just means that the holes are thwarting your efforts. Plug the holes, and all of a sudden your efforts are far more "effective" at filling the bucket. You are still filling it the same way, just something else is negating that. And inside the wall, the insulation is still damping resonances, but the coupling through the studs negates most of the effort.
That said, I'm not sure I'd agree that there is no effect at all from having insulation in a coupled stud wall: there is an effect. But it would be much larger if one of the leaves was decoupled.
Take a look at page 30 of IR693, for example. The same full-coupled wall assembly (one layer of 16mm drywall on each side of 62mm metal studs, 400 mm OC) without insulation in the cavity shows Rw of 36, and with insulation that rises to between 40 and 43 (depending on type of insulation). That's a significant increase, so the insulation is having an effect.. just not a huge effect.
So there is an effect... the effect is just not as "effective" as the effect would be if the drywall was decoupled. (That's a confusing sentence, for sure! effectively...).
Not sure if I'm explaining that sensibly or not. To me, it seems that the article is just poorly worded. It's not that there is no effect at all: just that the usefulness of the effect is overpowered by the coupling.
(I'm hoping Andre can summarize the above into 10 concise words or less...)
Simple answer? No.is the optimal performance on the insulation in the cavity in any way dependent upon the presence of resilient channel?
(I bet even Andre can't be more concise than that!)- Stuart -