U-values for internal elements (floors and partitions)

[GeoKor]

Hi folks,

I am not sure if this query is even valid, but lately I have been receiving feedback from various sources, regarding u-values of internal elements. When modelling, for compliance or overheating purposes, I have been reading (and hearing) that u-values of internal partitions and ceilings/floors should not be considered. So instead of applying the specified u-values for internal elements they should be kept at their default u-values.

While in my head, this doesn't make sense, it does change things dramatically(obviously). For example, when running an overheating analysis for communal corridors in a residential block, the walls of the corridors are considered to be sheltered walls. Therefore their u-value should be that of the specified sheltered walls (e.g. 0.17). This results in internal gains getting trapped inside the corridors, and that does not work in my favor at all. On the other hand, if internal u-values are left as the default values (e.g. 1.7) things are significantly better.

Can anyone advice? It might be too obvious, but once the seed of doubt is planted, then you just need to know for sure.

Thanks in advance, cheers.

[farahghanem]

Not sure why you would call it a seed of doubt, you have already tested and proved that the partition's u-values make a difference. However, if the adjacent spaces are being conditioned to the same temperature, then the transfer between them would be more or less adiabatic and hence people often ignore the u-values of interior construction.

For compliance purposes however, if you are leaving the u-values as default, then you're doing so for both the proposed and baseline, and hence setting both models at equal terms in internal heat transfer rates. I think this is also why most of the time rating authorities and modelers in general would say it is an acceptable practice.