Where to place duct between leaves; inevitable third-leaf

[Note: I had meant to post this under my studio build thread, 'basement studio-room project in Boston area'; perhaps it might be better to respond there (?)] I'm getting close to the time where the ducts above the studio will be replaced with the Johns Manville Superducting. **** IMPORTANT NOTE: These ducts will supply the above room only and NOT the studio; so I would like to isolate the inside of the ducts from the studio as much as possible. **** Placing the ducts between the leaves means that some cross-sections through the ceiling WILL involve other than 2-leaves -- this is unavoidable (please don't simply comment on this fact... UNLESS you can design a way to have the duct between the leaves and not have anything else but a 2-leaf cross-section at any point). SO, in other words, some cross-sections connecting 2 points will be either across 2-leaves or 3-leaves, or possibly only 1-leaf, or as great as 4 leaves). I have created 4 different diagrams of possibilities, which will help visualize the compromise that HAS to be made... I obviously want to make the best compromise. SO, is it better that: 1.) The inside of the duct be 2-leaf with respect to the studio, and single-leaf with respect to the above room? 2.) The inside of the duct be single-leaf with respect to the studio, and single-leaf with respect to the above room? 3.) The inside of the duct be single-leaf with respect to the studio, and double-leaf with respect to the above room? Remember (although I don't know how much it would apply), I'm trying to block sound coming OUT from the room and going into the above room(s) and not really the reverse. It seems to me that sound that gets into the duct is NOT equivalent to sound that gets through the above floor because the sound that gets into the duct has to travel to the end of the duct and come out the register... The superduct will likely attenuate some of that sound it would appear... Alternatively, sound that does gets into the duct may not pass that well straight up through the duct and then through the floor, but may better pass through the length of it and out the register. Diagrams: 1.) Simply 'ignoring' the duct's presence and placing the sheet-rock between the joists in the same position across the board (i.e., whether there will be a duct there or not). 2.) This is the same as the first option, but adds an 'extra' layer(s) of sheet-rock below the duct, creating a 2-leaf system between the duct and the studio. 3.) Here there is no sheet-rock along the subfloor, and the duct is moved up against the subfloor; sheet-rock is added below the duct. 4.) Same as the previous option except the layer(s) of sheet-rock below the duct are thicker. [Note that in the diagrams any cleating, between-joist- and fire-blocking and insulation and caulking is omitted for the sake of clarity -- these of course would be used in the actual construction; also, the steel joists of the inner ceiling leaf are 'I' shaped (back-to-back 'C' or 'S' sections)... I just used rectangles for simplicity] thanx for any help, as usual, K

I'd go with #4 with maybe an extra layer of beef in that cavity. The extra mass will partially compensate for the reduced "air" in the "mass-air-mass" equation. I could be full of s--- though, so we'll see what others have to say. :roll: :P --Keith :mrgreen:

Hey Kenni I was curious about the Johns Manville Superduct. When I was preparing to put my duct in I inquired about this duct. No one here had heard of it. If you get a chance, could you post a pic of the duct and let me know what the inside and outside are made of? I'd like to check into how the superduct stacks up against flexduct (which I had to end up using). Thanx!

camistan wrote:

Hey Kenni I was curious about the Johns Manville Superduct. When I was preparing to put my duct in I inquired about this duct. No one here had heard of it. If you get a chance, could you post a pic of the duct and let me know what the inside and outside are made of? I'd like to check into how the superduct stacks up against flexduct (which I had to end up using). Thanx!

Even better, you can go to the Johns Manville website and request a small sample piece. I did that... although I was able to find the material in the meantime, and so the sample came AFTER I bought the large pieces.... It was like 1.75 for a square foot, and came in 4 feet by 10 feet sheets, NOT as finished duct -- you need to be able to cut the pieces and build the ducts (which can be done with staples and the appropriate pressure (or heat) tape)... The material comes in 1", 1.5", and 2" thicknesses (maybe other thicknesses too?); I got the 1". There are also 2 different 'stiffnesses' of the board; I got the less stiff stuff. They have several pdf files of the specifications and construction techniques. Anyway, there is a pretty good looking foil layer on the outside, and center portion of dense fiberglass, and then a black 'polymer' liner on the inside. Like you, I had a lot of trouble locating a source at first. But on the JM website, there is a utility to find a supplier based on zip code. I found a place right here in Reading, MA.; an HVAC industrial supplier. When I called though, THEY didn't even hear of it! But after a little while on the phone, they managed to find info about it on their database -- and in fact had some in stock, so I went up and bought it. http://www.jm.com/locator/mqlocator.exe?link=find If I get a chance I'll try to take some close up pics of the sample... thanx, K

sharward wrote:

I'd go with #4 with maybe an extra layer of beef in that cavity. The extra mass will partially compensate for the reduced "air" in the "mass-air-mass" equation. I could be full of s--- though, so we'll see what others have to say. :roll: :P --Keith :mrgreen:

Keith, Actually, this may be a good application for using the big roll of lead sheeting I have... It could supply the mass in a much thinner form, therefore preserving a somewhat larger air gap.... (It was kind of like my thought of using cementboard in similar areas... and I just MAY do that in the areas around the ducts(?)) Plus, lead is heavy AND limp... It seems that placing limp (non-resilient) materials between stiffer materials is the idea behind the 'constrained layer' concept utililized by stuff like GreenGlue and QuietRock (is this not right?).. I was also wondering? I know that having a larger air gap is always better, but are the good effects only the result of the distance across the gap, at any given point, OR are they also due to the total volume of the dead air space? I was thinking that maybe the 'air spring' in a mass-spring-mass arrangement is not like that of a solid (or semi-solid) spring, in that the air can compress in ALL directions -- not just along the distance spanning the 2 mass leaves. In other words, when a shorter distance between mass leaves arises in only a few isolated spots, the air separating them can 'squeeze' laterally as well as in the directions towards the mass leaves. If this is true, then shorter gapped areas can benefit (some?) from adjacent larger gapped areas... that is, due to the overall volume between the mass leaves and not just the linear distance between them..... thanx, K

Luftweg wrote:

If this is true, then shorter gapped areas can benefit (some?) from adjacent larger gapped areas... that is, due to the overall volume between the mass leaves and not just the linear distance between them

Within reason, this is certainly true. If you had a double stud wall with only 1 inch between the studs, but 8 inches between drywall surfaces, the 1 inch part would only be 1 1/2 inches every 16 or 24 inches ... this would be fine. but the larger the narrow space, the more potential for problems, since i think air can't move too far too fast. but anyway, i would also do number 4 as Keith suggested.

I'd go with #2, with some changes - put most of your sheet rock "beef" up against the floor, then make your superduct ducts as tall as you possibly can, so they come within about 1/2" of your upper layer of the lower ceiling leaf (the one on the steel joists) - separate the bottom of the superduct from the top of your lower layers with 1" of fiberglass insulation, compressed to about 3/4", and you're there. The extra beef on top will help that leaf; the deeper duct will slow your air velocity down under the same CFM delivery, so will quiet the air noise; the deeper duct will also increase the air gap of the MAIN space; and by keeping the air gap at the bottom down to about 3/4", this will RAISE the m-s-m resonance of that third leaf to around 160 hZ, which is easier to stop than an octave lower (which would be caused by a 3" gap at that point - all these #'s assume 2 layers of 5/8 wallboard on the steel, and 1 layer below the superduct. I'm off to bed after first night shift (get to do four this time around :? ) if it's not clear what I mean, say so and I'll try to redo your sketch to show it... Steve

knightfly wrote:

I'd go with #2, with some changes - put most of your sheet rock "beef" up against the floor, then make your superduct ducts as tall as you possibly can, so they come within about 1/2" of your upper layer of the lower ceiling leaf (the one on the steel joists) - separate the bottom of the superduct from the top of your lower layers with 1" of fiberglass insulation, compressed to about 3/4", and you're there.

Thanks Steve. Although I'm a bit confused with 2 things: If I put the bottom of the duct within 1/2 inch of the inner-leaf sheetrock, how can I use the fiberglass at a 3/4 inch compressed thickness (perhaps you meant either 1/2 or 3/4 inch separation between the duct and inner-leaf?) Also, my diagram #2 shows additional sheetrock directly BELOW the duct, but above the inner-leaf layers (leaving a space between them). So, I'm thinking you are saying to omit that extra sheetrock below the duct and move it to above the duct.

knightfly wrote:

The extra beef on top will help that leaf; the deeper duct will slow your air velocity down under the same CFM delivery, so will quiet the air noise; the deeper duct will also increase the air gap of the MAIN space; and by keeping the air gap at the bottom down to about 3/4", this will RAISE the m-s-m resonance of that third leaf to around 160 hZ, which is easier to stop than an octave lower (which would be caused by a 3" gap at that point - all these #'s assume 2 layers of 5/8 wallboard on the steel, and 1 layer below the superduct.

[Important:] **** I just want to ensure that you know that these ducts will not be supplying the studio ****, but rather will be supplying the room above (and as such I would like to isolate the inside of the duct from the studio as much as possible). I am using the same internal cross-sectional area for the SuperDuct as the original ducts had, so there would be no increase in velocity (or decrease in volume). I don't really care about the any air noise coming from the ducts that goes into the above room (I'm sure these new ducts will be quieter anyway than the old galv steel ones).

knightfly wrote:

I'm off to bed after first night shift (get to do four this time around :? ) if it's not clear what I mean, say so and I'll try to redo your sketch to show it... Steve

YES! Please Steve... a representative diagram would be very cool, as I am not entirely understanding of all this..... Maybe I can also try to do a couple diagrams up with what you have written here in mind... thanx, K

Steve... I actually re-read your reply a couple times and it's starting to make sense to me (this is a reflection on my thickness, and not your post).... So, what you are saying is that keeping the bottom air space small will help with both the sound going straight up, completely through both the bottom and top of the duct, AND will help with any sound that makes it from the middle of the duct and out the above floor registers? If that's true, then that IS what I'm looking for... Not so sure yet of the exact positions of the sheetrock... AND do you think that I should use the 1/32 lead sheeting that I have wrapped around the ducts, just above it, just below it, or not even using it anywhere here?... Or, should it replace some of the sheetrock layering? The one last thing is that I want to mention that the main duct (which will also be SuperDuct) is going to run perpendicular to the joists, next to the center beam, and these between-joist ducts will need to connect to them... I will post a pic of the structural setup of the soffit/lintel/header around the centerbeams and supporting the steel joist ceiling that I had engineered for me -- but I still don't know about the exact positioning of the main ducting yet (but have a couple possibilities in mind)... thanx, K

Didn't realize those ducts were NOT to supply the studio - in that case, I'd move the duct upward and more of the mass BELOW the duct (more balanced), but do NOT let the extra air gap be wider than 3/4", and 1/2" would be even better. For the underside of the drywall BELOW the duct, the gap between that and your 2 layer ceiling needs to be small in order to keep that resonance higher frequency - this makes it easier to stop with the same amount of mass. I'd stay under 3/4" between the bottom of the "duct beef" and the top of your two ceiling layers, and damp between the two with slightly compressed fiberglass insulation (this is where a 1/2" gap gets 3/4" of insulation) Moving more of your mass down below the duct should help isolate the inner air path of the duct a bit more from studio noise... Steve

I have a theoretical question that is bugging me.... If having a smaller distance air gap raises resonance frequency, and if that frequency is 'easier' to control, why is it recommended to have as large an air gap distance as possible in the overall construction? My first guess here would be that with a larger volume air gap, the 'spring' would be larger and thus more capable to handle a higher spl (although it would create a lower resonance frequency?). Now, if only a small area will have a short gap -- which is also contiguous or directly connected to the main air space -- would that affect the resonance of the entire air gap system, or the resonance of that small area? Then again, the third leaf air space, where the middle of duct would be, would be its own isolated air space, so maybe that gives it its own resonance too (although, that third leaf would be directly connected to the outer leaf of the overall room). Somehow I was thinking this duct-between-the-leaves thing would be a tricky thing.... thanx, K

Here is an option reflecting what Steve recommended: There is the 'standard' amount of sheetrock above the duct (3 layers, just like elsewhere); below the duct is alot of sheetrock too and this keeps the air gap down to 1/2 to 3/4 inch... As I said before, I may likely also wrap the duct in some 1/32 inch lead sheeting, OR will put some lead between some of the layers of sheetrock... Now, the inside dimensions of the old duct were 5x5 inches = 25 square inches. If I follow the between-joist dimension, which is 13 inches (due to the sistering of the joists, which, btw, I had not pictured before in my diagrams), then allow for 1 inch each side (for the thickness of the Superduct), I would have an inside width of 11 inches. 25/11 = 2.27 inches, for the inside height of the duct. Adding an inch on top and bottom (for the duct thickness) gives 4.27 inches... So, the Superduct outside dimensions would be 13 by 4.27 inches... Then, the depth of the joists are 7.25 inches. Subtracting the 1.875 for the 3 layers of sheet rock above the duct gives 5.375 inches. Subtracting the duct height of 4.27 inches gives 1.105 inch. Finally, the planned distance between the bottom of the ceiling joists to the bottom of the floor joists is 1 inch... That leaves 2.105 inches for for the added sheetrock below the duct... This is not really enough for 3 layers of 5/8 sheetrock, AND to leave a 1/2 to 3/4 gap between them and the inner leaf sheetrock (it would be more like 1/4 inch... [note: the diagrams have not been entirely proportional; they are just rough sketches, so some things might look 'off' from my measurements above] Perhaps this is a job for cementboard? .... or maybe just some more lead? Certainly this needs more thought.... and a full blown drawup... K

Theory question musings - If building a "good" barrier, you only have mass-air-mass and don't concern yourself with 3-leaf effects - under these conditions, you want a wide air gap and two heavy (preferably slightly dissimilar) masses. Or at least different coincidence frequencies per layer in one of the masses - although this has, to my knowledge, not been quantified and documented. What HAS been documented, is that the ideal situation is for your mass-air-mass resonance to be at least an octave or MORE below the lowest frequency you want to stop, because at resonance the wall is weakest and it only starts to get good at more than an octave above resonance (msm resonance, that is) When you must by necessity add a third leaf, you enter the land of compromise - we know that the third leaf will weaken the barrier - we know that the msm resonance between the third leaf and one of the other two leaves will be a weak spot; and we can calculate the msm resonance resulting from the third, lighter leaf and the leaf closest to it - now, if we consider the two heavier leaves as being the main barrier to sound, and we make them as heavy and as wide apart as possible, then the weak point caused by the third leaf (with as narrow a gap as possible to raise the weak point's frequency) will have the hardest time penetrating the OTHER two leaves, which have been properly chosen for most mass and widest air gap (giving us the lowest msm resonance, as well as the best isolation at higher frequencies) - ONE of those higher frequencies is the one our third leaf is allowing through, so the OTHER two leaves will have an easier time stopping that frequency. Man, what a rambling statement; if I ever figure out what it means, I'll let you know :roll: Meantime, gotta sack out; been fighting strep throat the last week and have to cover two extra 12-hour night shifts to boot. Lemme know if any of this made sense... Steve

Here is a slightly more enhanced version showing a possible design cross-section: K

I think that looks really good -- and nice drawing, by the way! 8) I'm a little concerned about the amount of weight you're adding to that upper floor... Should I be? --Keith :mrgreen:

sharward wrote:

I think that looks really good -- and nice drawing, by the way! 8) I'm a little concerned about the amount of weight you're adding to that upper floor... Should I be? --Keith :mrgreen:

Thanx! The weight issue has been addressed, hence the sistering of joists, which were 'laminated' together with deck screws, and importantly, Titebond II glue, while tightly clamped together. The above floors are bedrooms, and dead load is still under 20 lbs (with 3 layers 5/8" sheetrock) and even figuring for 40 lbs live load, the doubled joists were determined to be good for over 13 feet span (more likely 13.5 or 14) at L/360.... Even at 50 lbs live load, the max span is 12 feet 4 inches (L/360).... And the longest actual span is only 11 feet 9 inches. I will ask the PE again to make sure there are no support issues just to make sure.... Remember, only the outer-leaf sheetrock is being added to the above floor; the inner-leaf sheetrock will be supported by independent (steel) ceiling joists that will in turn be supported predominantly by the end walls and lintel/headers (alongside the centerbeam, supported by sidewall hybrid wood/steel members), and then to the floor. The added weight that IS attached in the coffer-opening between the joists where the duct is, is only an isolated exception (and should not be much of an issue since it is only one space out of 11 such between-joist spaces).... I suppose I could do something for that particular opening to beef it up structurally, but it hasn't been worried about. Right now though, my main structural concern is the centerbeam --not because of the studio room construction, but rather a provisional thing in case I ever want to add an extra floor to the house; I would be boxing in this beam with the room and won't be able to beef it up easily afterward -- so I'm looking into the possibility of bolstering the beam NOW somehow (possibilities include: adding lallys :evil: ; or FRP reinforcement; box lamination with aircraft birch ply or LVL (?)). This was an afterthought, and the PE is busy right now, but will go over it with me sometime in the next few weeks... And of course also I am trying to ensure that the concrete walls and floor are treated properly... so those 2 things are on my mind.... thanx again, K

Cool -- the earlier drawings didn't show the sistered upper joists, but I see now that the newer drawings do show that. :) Wow, adding a floor to the house, eh? :shock: That sure would be a massive undertaking! --Keith :mrgreen:

sharward wrote:

Wow, adding a floor to the house, eh? :shock: That sure would be a massive undertaking! --Keith :mrgreen:

Ha. It's not even 'in the works' or remote planning stages; I just wanna make sure that I don't block-in something structural that I may later wish I hadn't .... Interestingly many of the same houses in the neighborhood have had 2nd floors added.... so it seems that it CAN be done -- and probably fairly practically too... K