AVS Forum banner
Status
Not open for further replies.

Using Liquid Nails for drywall?

32K views 46 replies 11 participants last post by  Brian Ravnaas 
#1 ·
I recently saw somebody else post a suggestion to use Liquid Nails to attach drywall to studs (assuming they are true), instead of screws. Is that really a possibility? I can see how it could potentially cut down on sound transmission through the screws, but can that really be done as a proper construction method?
 
#3 ·
I wouldn't suggest using liquid nail only. Proper constuction would be to use liquid nail to hold the drywall in place for a few seconds on the ceiling so that you can screw it down. As for the walls, since you are going to have to mud the seems anyway, why not do it right and put a few screws in the drywall to fasten it to the studs. It takes a little more time, but the corect way to do it.
 
#4 ·
Are you overlapping and stapling the paper edges on kraft-faced insulation to the face of the studs? If so, which I thought was the right way to do it, then the glue would only be adhering to the paper and not the wood. I doubt that would do much good unless you put liquid nails under and over the facing edge.
 
#5 ·
Assuming one layer of Gypsum/drywall is already good and well screwed to the wall studs, then lets talk about laminating a second layer.


Certainly gluing on the ceiling (no drywall screws, no laminating screws) is a dangerous thing, either glued to joists or glued to gypsum/drywall/laminated.


One fellow once wrote that "Liquid Nails For Subfloors" to laminate wall layers acted as Constrained Layer Damping to increase TL -- but I don't quite believe that. I'm also concerned that structurally it might be a little weak.


Regular "Liquid Nails" is more like Contact Cement, and I've read that Contact Cement can be used structurally on walls, but from a soundproofing point of view it's been said all over the place that the coincidence dip (a region that most drywall wall systems have less sound transmission loss at) around 2.5khz ends up dropping down to around 1.5khz, and since there's more voice/sound energy in a DVD at 1.5khz than at 2.5khz this means that the wall isn't as soundproof after glueing than without glueing.


Here's a couple of charts from a theoretical point of view, just me experimenting. I was surprised at the change in STC (i.e. that may not be valid, I don't know), although the coincidence dip change was right on the money. These are just two theoretical comparisons between two different wall systems, they are not meant to be an answer to the screwed vs glued question.

http://www.bobgolds.com/GlueOrScrew/home.htm
 
#6 ·
Code in my area allows the use of construction adhesive in effect to LESSEN the number of screws or nails needed, but not as a REPLACEMENT for screws or nails. For example, you may be required to space screws 12" on a vertical wall without the use of adhesive, but can space them 24" WITH adhesive. IN NO WAY can adhesive be used alone.


I suggest you read your code. It gives a substantial amount of information on the requirements for applying drywall.
 
#7 ·
Quote:
I recently saw somebody else post a suggestion to use Liquid Nails to attach drywall to studs (assuming they are true), instead of screws.
Oops. I just re-read that. You're not talking about laminating layers of gypsum together, you're talking about gluing gypsum to studs without screws -- and that doesn't work. Not for ceilings, and not for walls.


Gluing gypsum to studs in addition to screws is done to reduce pop-outs.


I remember reading somewhere that they used to glue gypsum to ceiling joists somewhere, until it started collapsing.
 
#8 ·
bob is dead-right, liquid nails won't damp anything. clay/chalk-filled rubber in toluene does not a CLD material make.


i think it was DOW that had ceiling-crashing problems a whiel ago, yikes!


code here concurs with what mikeWH posted, but a layer of 5/8" drywall is only about .015psi so there is no reason to believe that laminated sheets of the stuff wouldn't be an easy task for any adhesive. but code is code.


for the coincidence point to drop from 2500 to 1250 as your graphs indicate would require that the laminating adhesive yield properties equivalent to a solid slab of drywall 1.25" thick. the laminate would have to be 8 times stiffer than the single sheet. liquid nails is far too soft to drop the critical frequency that far, i suspect.
 
#9 ·
brianr820
Quote:
a layer of 5/8" drywall is only about .015psi so there is no reason to believe that laminated sheets of the stuff wouldn't be an easy task for any adhesive. but code is code
http://www.liquidnails.com/datasheets/dw24ds.pdf says it has 250psi, so that's about 10000 times more than it needs from a psi point of view.


But screws hold gypsum because screws are steel and are good under tension, and gypsum is rock and is good under compression, so they hold by resisting each other. They do this in earthquakes, and in humidity.


Adhesive sticks to the paper on either side of the drywall. When paper gets wet it falls apart (that's how they recycle paper). When a drywall that's screwed to the wall gets very humid and later drys it's not a problem, but if that wet paper were the only thing holding the drywall up, it'd be my guess that the paper would split down the middle of the paper (half staying in contact with the gypsum, and half staying in contact with the liquid nails), or variations depending on where the weak spots and tear spots and drywall break spots ended up being. With a subwoofer vibrating the walls, and if it's humid (several people breathing, moist basement), it seems to me the problem would occur.


Even if the problem isn't the paper, the drywall itself isn't good under tension. If the wall has been flexed enough, which happens over the years, some of the interior is going to be dusty. Again dust is not a problem with a compressible material as long as it doesn't escape, but dust (or 1" wide chunks) under tension is a problem.


Of course I'm just making all this up, but it sounds good to me.


Code is our friend.
 
#10 ·
tension -vs- compression is an interesting thought that i'll have to chew on. that is extremely little tension - on a wall. the dow ceiling failures... was that a paper-to-gypsum failure or an adhesive oxidized to zero-strength failure i wonder?


i'll conject that humidity wouldn't cause drywall paper failure, only bulk water might. is there data/info/write-ups on this?
 
#11 ·
brianr820:

http://www.svca.org/articles/Ceilingarticle.htm says
Quote:
I just got home and the living room ceiling is on the floor and it broke all my furniture ... In late 1960 homebuilders began to attach drywall ceilings with the use of glue and the practice continues today. What happened is that the glue failed. When I looked at the damage I noted that the paper face of the fallen drywall was not torn.
 
#12 ·
Quote:
tension -vs- compression is an interesting thought that i'll have to chew on
The easiest way to think about tension vs compression is to think of a garbage bag twist tie wire. Tie one end to the top of a steel fence, and now pull upwards on it. It's hard to pull. Without the paper wrapping, there's a good chance it'll cut through the flesh on fingers before it breaks. That's strength in tension. Now with it sticking straight up, stand on it. It bends and falls. That's weakness in compression.

Put a brick on the ground and stand on it, that's strength in compression.

Put an engine block on top of that brick, that's strength in compression.

Now take 1 pound of chain with quarter inch links, and use it to lift an engine. That's strength in tension. Now take a one pound brick and put the chain through the holes on each end so the brick becomes a link. Lift that same engine, and the brick snaps. That's weakness in tension.


There's lots of places where tension and compression are combined. In overpasses in earthquake zones, they have concrete columns that have been wrapped in steel tubes. The steel is tension around the concrete to keep the concrete dust/rocks together when they break (circumference tension), but it's the concrete that holds the span up (compression). Basement Lally Columns are the same sort of thing.

Quote:
the dow ceiling failures... was that a paper-to-gypsum failure or an adhesive oxidized to zero-strength failure i wonder?
I did a google search on that, and ended up with thousands of stock market quotes.
 
#13 ·
I always use adhesive when doing ceilings, because otherwise the only thing holding up the drywall is the resistance to pull-through of the screw heads (tensile strength). On the other hand, walls hang sideways to the screws (shear strength), and I consider the adhesive optional. It's still better with.
 
#14 ·
Quote:
Originally posted by BasementBob
Now with it [garbage bag twist tie] sticking straight up, stand on it. It bends and falls. That's weakness in compression.
Actually no. It's actually a phenomenon called buckling. Buckling is a failure mode that occurs at much lower compressive loads than the material capability. Aside from buckling, a twist tie can carry the same compressive load as it can tensile.


From a theoretical standpoint, if you could get the twist tie straight up and down, and stand on it such that your center of gravity was exactly centered on the twist tie, and not move at all, then it could support you. In practice of course, you can't do this (the margin of error would be extrememly small) with a twist tie.


But if you're careful, you can demonstrate this principal with an empty aluminum can (pop/beer). You can stand on it without it failing, but the slightest perturbation (touch the side with your foot) will cause it to buckle. Strictly speaking, this isn't failure in compression, it's failure due to buckling.


Of course the drywall paper-to-gypsum interface is anisotropic (i.e. it has different strengths in different directions). It doesn't fail in tension because of buckling; it's just weak in tension because the adhesion between the paper and gypsum is failing. A stack of books is similarly strong in compression but weak in tension.
 
#16 ·
Larry Fine

Quote:
I always use adhesive when doing ceilings, because otherwise the only thing holding up the drywall is the resistance to pull-through of the screw heads (tensile strength).
Do you mean glue to the joists, or laminate layers of drywall together, or both?
 
#17 ·
Both.
 
#18 ·
On walls screw the joints/edges glue everywhere else. Ceilings, not much choice but to screw. IMO
 
#19 ·
bob said: I just got home and the living room ceiling is on the floor and it broke all my furniture ... In late 1960 homebuilders began to attach drywall ceilings with the use of glue and the practice continues today. What happened is that the glue failed. When I looked at the damage I noted that the paper face of the fallen drywall was not torn.


quoting the article that he read.


if the paper face was not torn that implies that the adhesive had oxidized to basically zero strength at some point. think an old rubber band that's brittle and crumbles in your hand - similar polymers used in most construction adhesives.


later in the article it states that near the fireplace it was worse in a nearly-falling ceiling. heat would accelerate the aging process. so that's adhesive failure, not failure of drywall in tension/compression.


it's sensible, of course, that the tensile performance of drywall would be less than the compressive performance, but remember again that we're only talking (ceiling, straight down) .015psi of tensile stress, which peanut butter could probably handle.


i strongly doubt that you would have any problems with gluing drywall to itself with decent % of coverage and letting it sit (until the adhesive hit zero tensile strength, which is FAR and i mean FAR less of a problem with more advanced polymers than the bunk-grade hydrocarbon resins used in the 60's).
 
#20 ·
brianr820:


re the article:


It's also glued to the joists, rather than laminated layers of drywall -- joists only is MUCH less surface area for the glue, and an order of magnitude increase in pcf (1.5" wide joists on 16" centers, is approximately 10% of the area of the drywall available for glue at the bottom of the joists).


Anyway, it certainly shows that my 10000 times pcf estimate above is way out to lunch, otherwise it should have held.
 
#21 ·
i don't think your estimate is out to lunch, the material had just degraded to zilch in the way of properties. drywall and wood are porous enough to allow oxygen attack over time. same adhesive between sheets of metal of the same weight probably would have held...
 
#22 ·
Bob, don't forget that laminating a ceiling doubles the load on the same 1.5" of adhesive area and fasteners, except for the additional screws you install through the second layer. All the adhesive in the world between the layers won't do anything to help the bond between the first layer and the joists.
 
#23 ·
Larry:


My opinion is, at least for my own construction in my own home:

a) glue is optional/recommended on the joists/studs to reduce nail popping and rattles.

b) no glue between laminated layers. I note your statement about assisting installation with glue, but I'm fairly sure it'll reduce STC.

c) with two layers of drywall, a minimum number of screws on the first layer, and a full course of screws on the last layer. (as recommended in the USG manual)


I'm certainly not recommending gluing a ceiling to the joists as it's only means of support. IMO without screws it'll fall. With less evidence I feel the same way about laminating two layers on the ceiling - IMO without screws sooner or later, it'll fall.


I concur with you that the weakest point in a support structure is likely the first point it will fail, as you say "the bond between the first layer and the joists."
 
#24 ·
a glue without damping properties may reduce STC via impact on the high frequency (coincidence) region, but a glue with good damping properties will greatly improve performance in that area, EVEN IF it lowers the critical frequency a tad, but damping glues will not reduce critical frequency considerably or at all. and re-state that i suspect that liquid nails is nowhere near rigid enough to drive the critical frequency down by more than a single 1/3 octave band, if that.




witness: laminated glass offers improved sound performance due to the damping of the polymer central layer. the big coincidence dip becomes a flat region rather than a trough... and the polymer used to laminate glass is far from an efficient damping material. far. and it's stiff as far as laminating glues go and does cause the critical frequency to fall...


the impact in the coincidence region is totally dependent on the adhesive - an adhesive with properteis exactly matching the drywall will cause the dip to 1250hz, an adhesive that is soft will not cause anywhere near that dramatic of a dip, nowhere near. the Insul software doesn't take the damping or modulus of the laminating adhesive into question.


from http://irc.nrc-cnrc.gc.ca/cbd/cbd240e.html "Above the coincidence frequency, laminated glass (consisting of two or more layers of glass bonded together by thin plastic interlayers) can provide much higher TL than solid glass. Laminated glass may closely approach mass law performance above the coincidence frequency. This improvement is apparently due to damping (dissipation of vibrational energy) by the plastic interlayers. "


see http://www.domesticsoundproofing.co.uk/windows.htm for a nice picture
 
#25 ·
one more comment and then i'll shut the ol' pie hole.


if you leave the two layers of drywall un-bonded, you will avoid lowering of the coincidence resonance, but you may get another one for your trouble.


typically two layers of drywall with nothing filling the tiny air space between them exhibit only a couple of dB performance gain in the 500-1600hz region. Canada's NRC has hypothesized that this is due to the tiny air cavity yielding a cavity resonance.


you can observe this phenomenon in any of the NRC's published work, and also at www.pac-intl.com (seems to be a popular site on these boards).


at pac-intl compare their published data for 1+2 drywall to 2+2. their 3rd party data reports zero gain for over an octave in this region.


found here: http://www.pac-intl.com/tests_wl_wood.html#st


2db gain not 6 is reported for going from 1+1 to 2+2, etc. Those are the only two i compared just now, results may vary. Many competent labs have reproduced the phenomenon to varying extents, but it's not reflected by the Insul software.
 
#26 ·
brianr820

Quote:
the Insul software doesn't take the damping or modulus of the laminating adhesive into question.
True.

Quote:
2db gain not 6 is reported for going from 1+1 to 2+2
It depends on the wall system. For example, using NRC data, I can certainly find examples to show what you're saying with a staggard stud wall. But for a double stud wall it's well over 6db. Please see http://www.bobgolds.com/DoubleRSIC/WallComparison3.JPG



Certainly the coincidence dip reported by PAC for their RSIC is much smaller in their data, but you can see for NRC data of similar mass systems that the 500hz to 4khz region is well above RSIC's
http://www.bobgolds.com/DoubleRSIC/home.htm

Some of that can be attibuted to the extra wall thickness (air gap), and perhaps a bit more mass due to the 2nd stud wall vs the RSIC/HatChannel, but even still. Of course it's almost dishonest to compare lab results from one country to that of another country more than a decade later as I've done in this graph.
 
Status
Not open for further replies.
You have insufficient privileges to reply here.
Top