Give me your wisdom: how is plasterboard used in your practice? 2

[Greenalleycat]

Hopefully I'm not pushing the boundaries here. I'm writing my Master's project which relates to plasterboard usage in construction (very simplified, it's a part of a much larger project ongoing for my client).

As part of this, I am comparing New Zealand practice to international practice, primarily in Europe & USA (lol we use it as our primary wind and seismic bracing system).

The main uses for plasterboard here are 'standard' (substrate for plaster and paint), wet area, bracing, fire, and noise. For us, bracing is all based on a standardised test, and fire and noise systems are all tested systems with details provided by manufacturers. Standard and wet area boards have a manufacturing standard.

I have dug around on Google to compare, but I know this only ever teaches you so much. So I'm hoping the smart minds on here wouldn't mind sharing their 20s summary of how plasterboard is used in your industry.

 

[Greenalleycat]

After reading this, I'm afraid I have no idea what ductility means.

Is the TLDR of this just that: the NZ testing suggests that plasterboard is simply more ductile (performs better in an EQ) than what the codes would lead us to believe?

Your tl;dr is correct - plasterboard (as used here at least) performs much much better in a large EQ than what your codes would lead you to believe

Trying to put everyone on the same page for what we are talking about with ductility... (apologies for any obvious explanations here)

Ductility as a material property is the ability to deform plastically prior to snapping e.g. pulling a piece of steel and watching it neck
The opposite of ductility is being brittle i.e. just snapping without prior warning e.g. jumping on a piece of timber and it goes boom

Ductility in the structural engineering sense is based on that principle and, in seismic deisgn, is used as a way of reducing earthquake design loads
Basically, we define an imaginary earthquake and say you can choose to resist this elastically (by strength alone) or with ductility (by allowing permanent damage to something sacrificial to dissipate the energy from the larger shaking)

In the context of a 'system' e.g. a portal frame or wall, obviously not everything goes ductile (permanently deforms) at the same time
So defining the yield point requires a push test and figuring out where there is a 'system' yield (i.e. you have developed a failure mechanism, not just when the first joit yields)
The ultimate point is normally where the strength drops to 80% of the peak
The ductility of the system is therefore ultimate displacement / yield displacement - this
I'm sure we've all seen variations of these diagrams - this is for a RC concrete frame


The implementation of this 'ductility' is very simple - code says the earthquake shaking is 0.90g (for typical buildings in my city)
By allowing 'nominal ductility' (ductility 1.25) this reduces to 0.71g, by allowing high ductility (say ductility 3.5) it drops to 0.26g
The USA R-value is basically the same thing I believe
Plasterboard can easily achieve ductility 10-13 by this definition - you'll struggle to get that out of many other systems

That isn't the whole story of course
The point of 'ductility' in your primary lateral system in the seismic sense is energy dissipation
Energy dissipation isn't just ductility - it's related to the whole hysteresis loop, and plasterboard has pretty ugly (pinched) hysteresis loops
So each cycle dissipates less energy than other typical systems...it's just that this thing can handle a lot of cycles without ever really failing catastrophically

 

[Greenalleycat]

Less astethic? You do mud your joints? The joints here are mudded in almost all homes except for modular homes. In modular homes they have plastic pieces that cover the joints. In schools they use vinyl covered drywall, but that is for durability and has nothing to do with shear loading.

The drywall trade does a board count and will hang sheets vertical or horizontal to keep the number of joints to a minimum.

I had to go searching. It seems you may have some products we do not have. I am not sure the difference.

View attachment 4886

Yea all joints are plastered. When I say less aesthetic I mean for critical light angles - industry seems to have decided that horizontal joints get less complaints
Perhaps you guys do stuff differently
Reducing fixings in the body of the sheet also mitigates risk of callbacks

And yes, we have a metric (or imperial) ass load of different plasterboard options
We have 'standard' board (your good ol bog standard drywall), wet area plasterboard, bracing board, fire-rated boards, and noise-rated boards + some random specialty stuff like X-ray rated board

How many of those do you have equivalents for?

 

[Brad805]

I am in Canada. We are a metric country, but most residential trades still work in imperial. It is very weird, but it has been that way for a long time. Our rebar is all metric. Go figure.

Nobody specs anything to do with the drywall here. The joints and layout is all up to the tradesman. In your common 8'-0" wall height most will sheet horizontally, but we have people who want 9', 10' and taller walls in their houses. In those cases they may orient them differently to save joints and tape seams. They are easier to install horizontally, but butt joints do not have factory edges, so that does not always result in great joints. The maximum sheet length here is 12'-0", so the room size can dictate joint layout as well.

One of the largest drywall suppliers in North America is CGC. They have fire rated, noise and all those typical ones you mention.

Our job when we have done residential jobs stops when the framing is done. A well done joint should not be visible in any light unless we are talking about very light colors. Even then you should not see it unless the painter is poor. If it is a butt joint you need to be good at your job or you will most certainly see those.

 

[Greenalleycat]

Interesting, thank you. We (as engineers) don't spec horizontal/vertical either - the manufacturers and trades have worked through those details and the bracing rating as the same either way.
We have much longer sheets here - GIB products up to 6m lengths of certain boards (~20') to reduce the butt joint risk.

We often review the fixing installations of bracing plasterboard as it is part of the engineered system
Considering there is an entire trade dedicated to installing plasterboard it is impressive how bad most of them are at doing it...

 

[JoshPlumSE]

Eng16080: After reading this, I'm afraid I have no idea what ductility means.

LOL. I know exactly what you're talking about.

I went to a really good presentation about ductility in column design for highway overpasses a long while ago. I think it may have been given by Nigel Priestley. He was a Kiwi, so my sense is that he was advocating for a change in thinking among US engineers to move us closer towards they do in NZ.

Anyway, he was talking about how the US detailing requirements were good at making sure various elements of a structure were "ductile", but that our Force Based design provisions were leading to less ductile STRUCTURES.

The example was a highway overpass where one column was like twice the height of the other column. And, the question is how do you design the columns for the earthquake. Let's say you split the force equally between the two columns, then this is the size and reinforcement you'd get for each column. But, when the earthquake comes it doesn't directly impart a FORCE on the structure, it imparts a DISPLACEMENT. Right? The force is merely a by product of the impedance of the structure to movement.

If you had a base isolated structure, for example, you'd get very little force in those columns. He then went through some quick calculations that demonstrated that those two columns could be easily designed to accommodate the displacement imparted from the of the earthquake without collapsing.

Finally, to let the lessen sink in, he pointed out that most of us would have based the lateral force distribution based on stiffness. So, the short column would have to be designed to resist something like something like 75% of the lateral force. But, that would require the column to be bigger, hence when you make another iteration of the force distribution, you have more than 90% of the force being resisted by the short column. And, he then calculated the displacement at which that short column would fail catastrophically and it was a very, very small displacement.

The point was that you shouldn't always try to take the brute force of the earthquake. Rather, you should consider allowing more flexible and ductile structure behavior at times. Pointing out that this did certainly mean that this short column will get damaged in an earthquake that isn't even the design level.... But, that allowing this would do a better job of ensuring that the wouldn't be total collapse in the "Big One".

I had to revisit my notes on this lecture a few times before I really understood what he was talking about. Not sure I could be trusted to implement it on a real project with the confidence and wisdom that he did. But, it helped me to understand how the NZ codes are a written around a different philosophy.

 

[human909]

Some great thought provoking concepts here. A great discussion triggered by Greenalleycat and a great contextual summary by JoshPlum.

I can't say I am experienced in high seismic zones or complex seismic detailing.

But the emphasis on ductility is important throughout structural engineering and it isn't recognised as mush as it should be. Even simple multi bolt shear connection rely on yield in order to spread the load. Plasticity and ductility are great for sharing loads and absorbing energy.

 

[canwesteng]

I didn't think we were designing for displacement with earthquakes, but for acceleration. You could design a weak column that can displace a certain amount, but if there is a whack of inertia at the top, it's going to want to keep going when the ground motion changes direction, and that can be a larger distance than the ground moves.

 

[Martin Gillie]

Hopefully I'm not pushing the boundaries here. I'm writing my Master's project which relates to plasterboard usage in construction (very simplified, it's a part of a much larger project ongoing for my client).

As part of this, I am comparing New Zealand practice to international practice, primarily in Europe & USA (lol we use it as our primary wind and seismic bracing system).

The main uses for plasterboard here are 'standard' (substrate for plaster and paint), wet area, bracing, fire, and noise. For us, bracing is all based on a standardised test, and fire and noise systems are all tested systems with details provided by manufacturers. Standard and wet area boards have a manufacturing standard.

I have dug around on Google to compare, but I know this only ever teaches you so much. So I'm hoping the smart minds on here wouldn't mind sharing their 20s summary of how plasterboard is used in your industry.

In the UK it can be used to assist racking capacity against wind on timber stud walls (no earthquakes). It can't be the only thing doing this - OSB or only ply is required on the other face but some contribution is allowed. Widely used for fire protection too.

 

[Eng16080]

The USA R-value is basically the same thing I believe
Plasterboard can easily achieve ductility 10-13 by this definition - you'll struggle to get that out of many other systems

Assuming this value is in fact accurate and comparable to the USA R-value, this would be a higher R-value than any other lateral resisting system. The next highest R-value (per ASCE7) for steel moment frames and a few other steel/concrete systems is only 8.

The results of the NZ testing are in stark contrast to how the (US) codes treat the material in terms of seismic resistance. Whether the testing is accurate or not, for me it would be too risky to go against the codes and use plasterboard (sheetrock) in this capacity where I would be entirely reliant on the testing to justify my design decisions.

If the codes were revised to account for this apparent discrepancy, then I would more strongly consider using the material in such a capacity. I do wonder, though, has there been no similar testing of plasterboard (sheetrock) shear walls here in the US which would presumably show similar results to the NZ testing?

 

[Greenalleycat]

Assuming this value is in fact accurate and comparable to the USA R-value, this would be a higher R-value than any other lateral resisting system. The next highest R-value (per ASCE7) for steel moment frames and a few other steel/concrete systems is only 8.

The results of the NZ testing are in stark contrast to how the (US) codes treat the material in terms of seismic resistance. Whether the testing is accurate or not, for me it would be too risky to go against the codes and use plasterboard (sheetrock) in this capacity where I would be entirely reliant on the testing to justify my design decisions.

If the codes were revised to account for this apparent discrepancy, then I would more strongly consider using the material in such a capacity. I do wonder, though, has there been no similar testing of plasterboard (sheetrock) shear walls here in the US which would presumably show similar results to the NZ testing?

Couple of important things here - we don't design to the full theoretical ductility from testing. We limit it substantially and only design to 3.5 (not 10-13). This testing is also calibrated to our residential housing standard - it has limited applicability outside of that standard e.g. for commercial. Finally, the bracing values are pretty low so you'd struggle to actually brace something big with this without a lot of walls, which is usually not favourable to architects and clients.

Also - there probably has been some testing in the US. USG Boral was here 2017-2021 and they had to do our bracing testing to be able to sell to the market. Their boards performed comparably to our primary supplier (GIB).

https://productspec.co.nz/media/2ggfllxv/usg-boral-plasterboard-bracing-manual.pdf

I have a vague recollection (can't find a source for it so I may be BSing) that the US moved away from plasterboard bracing following the Northridge earthquake

 

[Eng16080]

Thanks for the follow-up. This is more aligned to my understanding.

 

[XR250]

The normal drywall practice is to have joints at the centerline of a single stud. Our code at the time suggested a second stud in cases where the drywall was being used as as a shear element. Having hung drywall I know how easy it is to have your drywall only bearing on a stud 3/8" or so. They did not put in the second stud.
View attachment 4879

Another thing to consider is that the connection to the bottom plate is usually compromised. For an 8 ft. wall in the US, studs are 93" + 1 1/2" bottom plate + 3" top plate = 97 1/2" Let's say they put the 1/2" ceiling up first, so you have 97" to cover with 96" of drywall so only 1/2" of lap at the base best case. 0" lap worst case if the ceiling goes up last.

 

[Brad805]

Displacement-Based Seismic Design of Structures - 2nd Edition by Dr. Priestley is a very good reference for the topic of ductility.