Therm 7.4.3 – Use for Interstitial Condensation Assessment. Is my approach correct?

[MC]

Hi,

I am trying to use Therm 7.4.3 to find out if there will be interstitial condensations inside a terrace.

I used Therm to calculate temperature isotherms and I compared it to moisture condensation chart provided by an insulation manufacturer (page 3 of ManufacturerCalcs.pdf). The temperatures within the deck, as calculated in Therm, are high enough to not allow condensation inside the terrace. (The attached file Terrace Full Detail.THM).

The manufacturer of insulation is advising that there is a risk of condensation. Though, in their calculations they do not consider effects of thermal bridges and the terrace geometry. They consider the terrace energy and moisture flow only in 1 direction (ManufacturerCalcs.pdf).

When I build the detail in Therm, following the manufacturer approach, I received similar results to the manufacturer temperatures distribution (attached file: Terrace Only.THM).

I consulted this with my sustainability engineers and they advise me that they don't believe that Therm can be used for interstitial condensation assessment and they think that my approach is not correct.

Could you please advise, if my approach to interstitial condensation assessment, using Therm, is correct, i.e. comparing temperature distribution inside terrace to moisture condensation chart?

Thank you.

Maciej

[Toby Cambray]

Hi Maciej

Condensation is a process that can't really be properly represented in Therm. As well as the temperature (which could give you the saturation vapour pressre), you would need to calculate the actual vapour pressure, which is dependent on the internal and external conditions and the vapour resistance of the materials. This can be done with other pieces of software. Depending on the detail, you may get some useful qualitative information from Therm, e.g. consider likely paths for vapour movement and think about whether significant amounts of vapour will reach a location of less than say 12 degrees (roughly the dew point of 20degree C, 50%RH air.) An example would be the typical timber frame intermediate floor detail we use in the UK, where a band of air tightness membrane wraps around the floor plate, where the insulation is either thinner or the AVCL ends up part way through the insulation.

The Glaser method of predicting interstitial condensation is codified in EN 13788, but is really intended for one dmiensional siuations. HTFlux impliments this in 2D, which may do what you need. However, I'd caution that 13788 has significant limitations; it cannot take into account moisture storage (in masonry, wood, gypsum or other hygroscopic materials), or the transport of liquid water due to capilary action and surface diffusion. If you have any materials that support these effects, even HTFlux may not give you a sensible idea of what's going on. If this is the case you would need to look at simulaiton in Wufi 2D which is somehting we could help with.

Toby

[yal...@gmail.com]

Hi.

The glaser and dew point calculation method is used for 1-dimensional assemblies. It is basically finding the temperature at each surface of the construction layers and the partial vapor pressure based on the vapor resistance of the layers. You can use the glaser and similar methods at the center of assembly where flow is one-dimensional or temperature lines are parallel.

There's really no way to model your condition with Glaser method accurately. You could use a transient hygrothermal program like WUFI 2D but would likely take weeks to solve this condition and will likely have a lot of convergence errors as much of the materials are low permeability and could have misleading results. You can estimate the partial pressure based on the effective distances and temperatures but would be an arduous task.

Conservatively you can use Therm to evaluate the detail. Assuming that the interior dew point temperature is when condensation will occur you can use that to evaluate if there would be condensation within the assembly and take appropriate measures.

Keep in mind that you have metal deck and rigid insulation which have low vapor permeability. The metal deck is a vapor barrier and isn't in the manufacturer calculations.

I suggest reading ASHRAE Handbook of Fundamentals Chapters 25 through 27 for a better understanding of heat and moisture transfer in exterior walls. It will also help understand what boundary conditions to use. Another document to look at is ASHRAE research project RP-1365 Thermal Performance of Building Envelope Details. This has good explanations of the temperature factor and ways to assess performance of building envelopes.

Last of all you need to consider historical performance of the assemblies. Roofs have a membrane that is impermeable and a vapor barrier which is basically on the wrong side of the assembly for cold climates. However these have historically been used successfully.

Yalin

[MC]

Hi Toby

Thank you for your response. I am looking more for a crude assessment, than an exact solution, as there can be many details that need to be assessed, for every project.

I used Therm to model temperature distribution, assuming that if temperature will be 12ºC below slab, there will be condensation. For the sample terrace, considering that 12ºC is above the slab, the risk of condensation should be low. It is because on the top of the slab is a vapour barrier. Additionally, the insulation below the slab and the slab vapour resistance reduces the amount of moisture that can reach locations where the temperature is 12ºC. I also added a vapour barrier below the insulation that is fixed to underside of the slab.

My question would be, if Therm can model temperatures distribution accurately.

I downloaded the HTFlux software and I will test the detail.

Maciej

[MC]

Hi Yalin,

Thank you for your response. You noted that it is possible to use Therm to conservatively evaluate the detail.

Do you think that Therm isotherms are sufficiency accurately calculated and can be used to be compared against the dew point temperature?

My thinking was that if dew point is 12ºC degree below slab in the manufacturer calculations than if the 12ºC isotherm will be above the slab, then there should be no condensation, when looking at the dew graph (calculated in one direction). Additionally, that there is a vapour barrier, on the top of the slab, which performance is guaranteed by the manufacturer and moisture will not be able to get above the slab. I will take a look at the ASHRAE research project RP-1365.

Maciej

[yal...@gmail.com]

THERM is a tested and validated software. Maybe the creators can get into more detail but as far as I know with the correct inputs it can model isothermal lines correctly.

If you are in doubt you can use the validation models in ISO 10077 to check the accuracy.

Yalin