[GSBN] Lime plaster and expansion joints

Tue Feb 10 10:02:34 UTC 2009

I've been following these mails about cracks with great attention. 
Appart from respecting the suggestions made earlier by John and others I 
would like to suggest the use of gypsum (as was done on the façades of 
most buildings in Paris) because it is the only binding agent that does 
not shrink while setting. In order to make it water resistant (gypsum is 
somewhat solluble) it was often mixed with lime (and sand I believe). I 
have not yet found any good documentation that states mixing proportions.
It is also possible to protect a gypsum plaster with lime plaster. In 
that case it needs a mecanical bond that is improved by adding sand to 
the gypsum + scratching. When I say lime, I mean what we call "Calcic 
Lime" in Europe (lime that does not set under water). Gypsum is generaly 
not compatible with 'Hydraulic lime' (lime that sets under water and is 
comparable with a Calcic Lime-Cement mix).
Mind you, one should not use just any gypsum but I have no idea how the 
type gypsum for outside use is called in English. In French it is called 
'plâtre gros' (gypsum thick). A good website with english textes is 
http://www.platre.com/

An other solution (that I  adhere to ;-) for exposed façades is cladding.

I too worry about expansion joints letting in water. But then again, a 
nice veranda arround the building can take care of that.

I also heared a couple of days ago (at the St Astier hydraulic lime 
plant) that scratched lime plaster is much less likely to crack than 
smooth (troweled) plaster but that the latter is more water repelling. 
They also claimed that in the old days plaster was almost always allways 
troweled/smooth.

Bye,

André - cracking up - de Bouter
France

John Swearingen a écrit :
>
>
> On Mon, Feb 9, 2009 at 12:54 PM, RT <ArchiLogic at yahoo.ca 
> <mailto:ArchiLogic at yahoo.ca>> wrote:
>
>
>
>     We know that the steel reinforcing in reinforced concrete works
>     because the coefficients of thermal expansion are similar for the
>     two materials so I don't think that the cracking that Jeff is
>     talking about is due entirely to juxtapostion of the two different
>     materials, plaster and steel.
>
>
> Well.....I dunno about coefficients, but I have seen cracks along 
> steel members.  I think what happens is that the steel superheats 
> while the skin doesn't.  My theory goes something like thisL  The sun 
> heats the plaster and the plaster heats the steel and the steel gets 
> REALLY hot because it's backed by all this great straw insulation and 
> because it's really hot it gets really big, but the skin isn't hot 
> (except maybe right around the steel) and so the steel swells and the 
> skin doesn't, ergo crack.
>
>
>
>     Here in Canada where exterior temperatures can be at either
>     extreme of the thermometer, it's common practise to keep the
>     structural frame completely inside of the thermal envelope as much
>     as is possible in order to minimise the amount of
>     thermally-induced movement of the frame.
>
>     And brittle materials like glass and masonry (including plaster)
>     should be mechanically isolated so as to avoid the destructive
>     effects of movement of the frame, whether those stresses are
>     thermally induced or as a result of structural loads, even if the
>     frame is entirely inside of the thermal envelope. ie One would
>     never use fixed connections with masonry (and plaster) at both
>     ends of a structural steel member. One end can be a fixed
>     connection but the other should be a "roller" type connection ie
>     free to move.
>
>     In an infill bale wall, this might mean pre-stressing the mesh by
>     pulling down to the foundation and/or sill plate only and
>     providing a connection at the top of the panel that only restricts
>     lateral movement of the panel.
>
>
> Now who's gonna do that?  I would mention that often we might have a 
> steel brace frame inserted into a wood wood frame structure, with the 
> mesh attached to the wood.  So then you have to think of the steel as 
> kind of an alien material behaving differently than the rest.
>
>
>
>
>     But back to Laura's question:
>
>         ...lime plaster ... begs the question of whether we need to be
>         using
>         expansion joints on large expanses... lime having different
>         characteristics
>         than a cement stucco
>
>
>     Control joists are intended to deal with cracking due to stresses
>     from structural loads as well as those induced by temperature and
>     shrinkage (T&S)  so I think that it would be imprudent to omit
>     them or significantly change the spacing when used with lime plasters.
>
>     And I think that minimising shrinkage cracks in lime plasters is
>     as much or more about process as it is about specs .
>
>     For instance, lime mortars should be well-aged before application
>     in order to work properly, the mix for the scratch & brown coats
>     cannot be too rich (as is the case with PC mortar mixes), the
>     brown coat should be floated the day following application to
>     further compact it (unlike with PC mortar mixes) and the plaster
>     needs to be protected from dryouts due to sun and wind exposure
>     during the lengthier (than Portland cement plasters) setting
>     period -- all to minimise the amount of shrinkage that *will*
>     occur during the curing period.
>
>     I don't know what the "official" recommendations for control joint
>     spacing with lime plasters might be but given the incremental
>     degree of attention and amount of time that is needed for lime
>     plasters to be done properly and given that that incremental
>     attention and time is likely to be circumvented (probably the
>     biggest reason why the use of Portland cement plasters displaced
>     lime plasters) I would venture that a panel size should probably
>     not exceed 144 sq ft., with control joint spacing in any direction
>     not exceeding 12 ft whereas with Portland cement mixes, I wouldn't
>     want to exceed a panel size of 100 sf with max spacing between
>     control joints at 10 ft although the "official" specs say that
>     larger panel sizes and greater spacings are acceptable.
>
>
> All very well, but what about the fact that you're making a crack in a 
> wall that doesn't have a paper backing to deal with the water than 
> comes rushing in?
>
> John
>
>
>     -- 
>     === * ===
>     Rob Tom
>     Kanata, Ontario, Canada
>     < A r c h i L o g i c  at  ChaffY a h o o  dot  C a >
>     (manually winnow the chaff from my edress in your reply)
>     __________________________________________________
>     Do You Yahoo!?
>     Tired of spam?  Yahoo! Mail has the best spam protection
>     aroundhttp://mail.yahoo.com_______________________________________________
>
>     GSBN mailing list
>     GSBN at greenbuilder.com <mailto:GSBN at greenbuilder.com>
>     http://greenbuilder.com/mailman/listinfo/GSBN
>
>
>
>
> -- 
> John Swearingen
>
> Skillful Means
> www.skillful-means.com <http://www.skillful-means.com>
> ------------------------------------------------------------------------
>
> _______________________________________________
> GSBN mailing list
> GSBN at greenbuilder.com
> http://greenbuilder.com/mailman/listinfo/GSBN
>   