Hello
I am trying to model a wall that is only 30mm from a neighbouring building. Has anyone any guidance on the best way to simulate the external conditions one might expect in this situation?
One idea is to model the neighbouring wall and enter the 30mm space as a ventilated cavity and add a moisture source to reflect some rain.
Alternatively, I could alter the external climate to reflect the conditions in the 30mm space. This would however involve some large assumptions. I presume you would also turn off radiation absorption and emission in this instance.
Any guidance on setting up this situation would be much appreciated.
Simulating external conditions between buildings
-
Madonna Stewart
- WUFI User
- Posts: 5
- Joined: Mon May 06, 2019 3:36 pm -1100
- Location: Australia
Re: Simulating external conditions between buildings
Hi,
if the "air gap" boundary condition is not the primary focus of the simulation, it will probably be sufficient to use a more or less strongly simplified model of the situation.
Assuming that the neigbouring wall surface has approximately the same temperature and humidity content as the surface of the simulated wall, and that the thin air gap is only poorly ventilated, one option may be to assume that no appreciable heat and moisture flows are exchanged through this component surface at all. In this case, instead of modelling the air gap, an "adiabatic" boundary condition may be used: Set the heat transfer coefficient to zero (or equivalently the heat resistance to a large number, depending on your preferred settings) in order to suppress any heat transfer, and set the sd-value of the surface to a high number in order to suppress any vapour transfer.
You mention rain; if you think that such a narrow gap will intercept enough rain to be relevant for the simulation, you can place a moisture source into the wall, close to the surface. WUFI provides the option to have the hourly moisture source strength derived as a user-specified fraction of the rain.
In fact, if you wish to "soften" the adiabatic condition a bit and if you have some idea about any residual heat flow, you can even put a heat source next to the moisture source and control the heat source strength appropriately.
Another option would be to model the air gap as one of WUFI's air layers, as you describe. WUFI assumes the (closed) air layers contain stagnant air, but placing an air change source into the layer and connecting it with the outdoor climate will simulate some air exchange (I don't know what an appropriate air exchange rate would be). If you wish to take incident rain into account, the moisture source should be placed in the wall, not in the air gap.
For the wall "on the other side", it may be sufficent to include a few centimeters of that wall in the assembly and to cut off the rest with an adiabatic surface condition (i.e. suppress any heat and moisture flows as mentioned above). In this way that wall reacts to the conditions in the air gap but not to the neighbour's indoor conditions. If you happen to know those indoor conditions, you can choose to model the entire neighbouring wall and apply the indoor conditions.
Finally, if you happen to know the conditions in the air gap (from measurements, or from a simulation with some sophisticated fluid dynamics software), you may simply apply those data as the boundary conditions via a climate file. This file would contain temperature and humidity, no (or only negligible) solar radiation, no (or only negligible) wind, and some rain (if you wish so). Surface transfer coefficients would be closer to values for indoor than outdoor coefficients.
It may be helpful to try a few options and to play with the settings for which only a guess is available. If the resulting variation of the results is small enough so that your conclusions drawn from the simulations remain the same, the precise choice of model is not important. If your conclusions depend on the model choice, more thought and/or research will be nesessary.
Regards,
Thomas
if the "air gap" boundary condition is not the primary focus of the simulation, it will probably be sufficient to use a more or less strongly simplified model of the situation.
Assuming that the neigbouring wall surface has approximately the same temperature and humidity content as the surface of the simulated wall, and that the thin air gap is only poorly ventilated, one option may be to assume that no appreciable heat and moisture flows are exchanged through this component surface at all. In this case, instead of modelling the air gap, an "adiabatic" boundary condition may be used: Set the heat transfer coefficient to zero (or equivalently the heat resistance to a large number, depending on your preferred settings) in order to suppress any heat transfer, and set the sd-value of the surface to a high number in order to suppress any vapour transfer.
You mention rain; if you think that such a narrow gap will intercept enough rain to be relevant for the simulation, you can place a moisture source into the wall, close to the surface. WUFI provides the option to have the hourly moisture source strength derived as a user-specified fraction of the rain.
In fact, if you wish to "soften" the adiabatic condition a bit and if you have some idea about any residual heat flow, you can even put a heat source next to the moisture source and control the heat source strength appropriately.
Another option would be to model the air gap as one of WUFI's air layers, as you describe. WUFI assumes the (closed) air layers contain stagnant air, but placing an air change source into the layer and connecting it with the outdoor climate will simulate some air exchange (I don't know what an appropriate air exchange rate would be). If you wish to take incident rain into account, the moisture source should be placed in the wall, not in the air gap.
For the wall "on the other side", it may be sufficent to include a few centimeters of that wall in the assembly and to cut off the rest with an adiabatic surface condition (i.e. suppress any heat and moisture flows as mentioned above). In this way that wall reacts to the conditions in the air gap but not to the neighbour's indoor conditions. If you happen to know those indoor conditions, you can choose to model the entire neighbouring wall and apply the indoor conditions.
Finally, if you happen to know the conditions in the air gap (from measurements, or from a simulation with some sophisticated fluid dynamics software), you may simply apply those data as the boundary conditions via a climate file. This file would contain temperature and humidity, no (or only negligible) solar radiation, no (or only negligible) wind, and some rain (if you wish so). Surface transfer coefficients would be closer to values for indoor than outdoor coefficients.
It may be helpful to try a few options and to play with the settings for which only a guess is available. If the resulting variation of the results is small enough so that your conclusions drawn from the simulations remain the same, the precise choice of model is not important. If your conclusions depend on the model choice, more thought and/or research will be nesessary.
Regards,
Thomas
-
Madonna Stewart
- WUFI User
- Posts: 5
- Joined: Mon May 06, 2019 3:36 pm -1100
- Location: Australia
Re: Simulating external conditions between buildings
Hi Thomas
Thank you very much for your suggestions, it was extremely helpful. I tried a couple of the options you suggested and through testing multiple variables was able to reach a qualitative conclusion due to the results not varying too much at all. Thanks again for your comprehensive advice.
All the best, Madonna.
Thank you very much for your suggestions, it was extremely helpful. I tried a couple of the options you suggested and through testing multiple variables was able to reach a qualitative conclusion due to the results not varying too much at all. Thanks again for your comprehensive advice.
All the best, Madonna.
Re: Simulating external conditions between buildings
You are welcome, Madonna 
Best regards,
Thomas
Best regards,
Thomas