Revision history [back]

So, this can get kind of messy and complicated depending on how you build your model.

The basics are that each of the discharge coefficient, width factor, and height factor settings controls how those parameters will scale according to the window opening factor. So, for example, if you have the width factor set to equal 0 when the opening factor is 0, and 1 when the opening factor is 1, then when the opening factor is 0.5 the window width factor will also be 0.5.

It is the product of the discharge coefficient, width factor, and height factor that defines the actual effective window opening area. This can result in some interesting scaling if you allow all three of them to vary at once. In your example you have the discharge coefficient varying from 0.001 to 0.5, and the height and width factors both varying from 0 to 0.5. This means that the maximum opening area of the window will be equal to 0.5x0.5 = 25% of the window area, with a coefficient of 0.5. If your opening factor is only 0.5, then the effective opening area will not be half the maximum - instead it will be 0.5 x 0.5 x 0.5 = 12.5% of maximum effective area.

If you want the window area to scale in a more intuitive linear fashion, it may be better to fix both the discharge coefficient and the width factor and only vary the height. That way, when the window is 50% open the effective opening area will also be 50% of the maximum.

Below, for example, is a typical window definition in one of my models using this approach. Here I have set the maximum height factor (i.e. the height when the opening factor = 1) to 0.2 / 20%. This was estimated by using this window opening calculator from the UK. I have the air moss flow coefficient when opening is closed input set very low because I didn't want to be modelling the infiltration using the AFN in this case.

AirflowNetwork:MultiZone:Component:DetailedOpening,
ExteriorWindowAwning_0.2,    !- Name
0.0000001,               !- Air Mass Flow Coefficient When Opening is Closed {kg/s-m}
0.65,                    !- Air Mass Flow Exponent When Opening is Closed {dimensionless}
NonPivoted,              !- Type of Rectangular Large Vertical Opening (LVO)
0,                       !- Extra Crack Length or Height of Pivoting Axis {m}
2,                       !- Number of Sets of Opening Factor Data
0,                       !- Opening Factor 1 {dimensionless}
0.62,                   !- Discharge Coefficient for Opening Factor 1 {dimensionless}
1,                       !- Width Factor for Opening Factor 1 {dimensionless}
0,                       !- Height Factor for Opening Factor 1 {dimensionless}
0,                       !- Start Height Factor for Opening Factor 1 {dimensionless}
1,                       !- Opening Factor 2 {dimensionless}
0.62,                    !- Discharge Coefficient for Opening Factor 2 {dimensionless}
1,                       !- Width Factor for Opening Factor 2 {dimensionless}
0.2,                       !- Height Factor for Opening Factor 2 {dimensionless}
0;                       !- Start Height Factor for Opening Factor 2 {dimensionless}

Note however that letting having the window opening area scale in a simple linear fashion like this make result in overventilation in many cases. I have it set up like this because I control the window opening factor using my own EMS scripts which are designed to deal with that problem, which EnergyPlus's AFN control systems are not very good at (discussed here if you want more detail).

If you do let all three factors scale like that though then it's important to be aware of how that affects the overall relationship of your window opening area to the opening factor, since this will then interact with the maximum opening factor setting of the individual openings. For example, if you tried to set the maximum opening factor to 0.5, and allowed the discharge coefficient, width, and height factors to all vary, then your actual maximum effective opening area would be 0.5^3 or 12.5% (an error I have made in the past, and eventually discovered that the windows I thought had been set to have a maximum opening area of 20% actually effectively had an maximum opening area of 0.8%).

So, this can get kind of messy and complicated depending on how you build your model.

The basics are that each of the discharge coefficient, width factor, and height factor settings controls how those parameters will scale according to the window opening factor. So, for example, if you have the width factor set to equal 0 when the opening factor is 0, and 1 when the opening factor is 1, then when the opening factor is 0.5 the window width factor will also be 0.5.

It is the product of the discharge coefficient, width factor, and height factor that defines the actual effective window opening area. This can result in some interesting scaling if you allow all three of them to vary at once. In your example you have the discharge coefficient varying from 0.001 to 0.5, and the height and width factors both varying from 0 to 0.5. This means that the maximum opening area of the window will be equal to 0.5x0.5 = 25% of the window area, with a coefficient of 0.5. If your opening factor is only 0.5, then the effective opening area will not be half the maximum - instead it will be 0.5 x 0.5 x 0.5 = 12.5% of maximum effective area. area (i.e. 12.5% of 25% of the window area).

If you want the window area to scale in a more intuitive linear fashion, it may be better to fix both the discharge coefficient and the width factor and only vary the height. That way, when the window is 50% open the effective opening area will also be 50% of the maximum.

Below, for example, is a typical window definition in one of my models using this approach. Here I have set the maximum height factor (i.e. the height when the opening factor = 1) to 0.2 / 20%. This was estimated by using this window opening calculator from the UK. I have the air moss flow coefficient when opening is closed input set very low because I didn't want to be modelling the infiltration using the AFN in this case.

AirflowNetwork:MultiZone:Component:DetailedOpening,
ExteriorWindowAwning_0.2,    !- Name
0.0000001,               !- Air Mass Flow Coefficient When Opening is Closed {kg/s-m}
0.65,                    !- Air Mass Flow Exponent When Opening is Closed {dimensionless}
NonPivoted,              !- Type of Rectangular Large Vertical Opening (LVO)
0,                       !- Extra Crack Length or Height of Pivoting Axis {m}
2,                       !- Number of Sets of Opening Factor Data
0,                       !- Opening Factor 1 {dimensionless}
0.62,                   !- Discharge Coefficient for Opening Factor 1 {dimensionless}
1,                       !- Width Factor for Opening Factor 1 {dimensionless}
0,                       !- Height Factor for Opening Factor 1 {dimensionless}
0,                       !- Start Height Factor for Opening Factor 1 {dimensionless}
1,                       !- Opening Factor 2 {dimensionless}
0.62,                    !- Discharge Coefficient for Opening Factor 2 {dimensionless}
1,                       !- Width Factor for Opening Factor 2 {dimensionless}
0.2,                       !- Height Factor for Opening Factor 2 {dimensionless}
0;                       !- Start Height Factor for Opening Factor 2 {dimensionless}

Note however that letting having the window opening area scale in a simple linear fashion like this make result in overventilation in many cases. I have it set up like this because I control the window opening factor using my own EMS scripts which are designed to deal with that problem, which EnergyPlus's AFN control systems are not very good at (discussed here if you want more detail).

If you do let all three factors scale like that though then it's important to be aware of how that affects the overall relationship of your window opening area to the opening factor, since this will then interact with the maximum opening factor setting of the individual openings. For example, if you tried to set the maximum opening factor to 0.5, and allowed the discharge coefficient, width, and height factors to all vary, then your actual maximum effective opening area would be 0.5^3 or 12.5% (an error I have made in the past, and eventually discovered that the windows I thought had been set to have a maximum opening area of 20% actually effectively had an maximum opening area of 0.8%).