30 January, 2019
This blog will help you to get more out of your external modeling. We’ll explore the final steps of setting up and running an external model in the 6SigmaDCX software, including setting the outside environmental conditions, checking the grid, and viewing results after the solve. We’ll use an example model to demonstrate some of the new external model features of 6SigmaDCX Release 13 and to dig deeper into the easy-to-use External Model Wizard from previous releases.
Figure 1. Isometric View of the Building Compound that will be examined in this blog
The external model we will use for this example includes three buildings: two larger buildings with vents running along the sides and one smaller building with a chiller cluster on top, surrounded by generators.
Opening the Model and Setting External Conditions
We can set external conditions in two ways: either by changing properties of objects in the Model Tree window or by using the ‘Set External Conditions’ wizard on the ‘Build’ ribbon. While modifying object properties can provide a more extensive set of options to define the conditions, the Wizard provides an easier alternative that captures the most impactful properties.
Figure 2. "Set External Conditions" opens the External Model Wizard to help set up ambient environment condition
Click the ‘Set External Conditions’ button on the ‘Build’ ribbon to open the wizard. From here, we can include as much or as little information as we like on the Model Location, Wind Conditions, Solar Intensity, and Ground Roughness.
Information supplied in the Model Location section will be necessary for solar calculations. Similarly, it may prove useful for keeping track of which model is which, but you can leave this blank if the model is site agnostic.
The Wind Conditions section is the key element of external models. Here, you will be prompted to enter the velocity profile of the wind as well as the temperature of the model. It is easier to solve the model if the flat sides of the building are along the X and Z axes. It may be necessary to adjust the angle of the wind to accommodate for the change in building orientation.
Next is the Solar Intensity section. For most external models, solar intensity is not a primary concern; however, if it is vital to include, please contact firstname.lastname@example.org for help.
The last section pertains to the Ground Roughness of the model. Common values are presented in the lower left corner of the dialog, along with the environments you would expect to see those in. For most site agnostic models, it is safe to assume a value of 0.0024 m (0.0945 in.). Once you have supplied all the information you want to, click Finish to set the external conditions.
We can alter and add information to any of the external conditions later by changing an item’s property sheet in the Cooling tab of the Model Tree window.
Pre-simulation Grid Checks
One of the final steps before running a model should be a grid check to ensure that all inlets and outlets are gridded properly to produce accurate results. With Release 13, the grid rules that apply to external objects (the Chiller, Generator, and Cooling Tower objects) have been updated, so that the automatically generated grid should be sufficient to capture airflow around and through the objects.
Figure 3a. Automatically unstructured grid for Release 12 Figure 3b. Automatically unstructured grid in Release 13
The images above highlight the difference in the amount of grid generated using the same target cell count between Release 12 (left) and Release 13 (right). As can be seen, Release 13 generates a greater number of cells closer to the generator, and cell size increases at a much lower rate with distance from the generator when compared to Release 12. While having fewer grid cells generally leads to faster solve times, the transition from large grid cells to small grid cells can cause errors in the CFD Solver. This can lead to longer solve times and is ultimately not worth the tradeoff of having fewer cells.
Figure 4. Bird's eye view of the automatically-generated grid in Release 13
Examining the generators surrounding the smaller building, we see that an ample amount of grid has been generated and is consistent across each generator. It is vital that the Exhaust ports of the Generator objects be gridded properly. If they are not, the trajectory of contaminants can be skewed, and results may not be indicative of what is truly happening near your facilities.
Now that we have assessed the generators thoroughly, we will move to the rooftop to analyze the grid on the chillers and cooling towers. Often, chillers are clustered together on the roof of buildings. The main problem with this is recirculation between units as the airflow demand is much higher than what is being supplied. To properly analyze any recirculation, the Chiller objects need to have a sufficient amount of grid applied to the inlets and outlets. If there is not enough grid, Grid Control objects may be required.
Figure 5. The automatically-generated grid for chillers in Release 13
For this blog, we will assume that the automatically generated grid is sufficient to track the outflow of our chillers without Grid Control objects, as it appears to be in the above image.
Other objects that should be examined for having proper grid spacing include any ducts that have airflow through them, vents of high importance, and some buildings (depending on their level of detail and the complexity of their geometry).
Once we have verified that all objects are gridded accordingly, we can go ahead and solve the model.
Now that the model is solved, we can explore results using a variety of old and new features.
Release 13 introduces result plots that are unique to external models. These include the Building Plot and the Plant Plot, located on the ‘Results’ ribbon.
Figure 6. The red boxes highlight the object-specific result plots and box in blue is the Result Cloud object plots, which can display any variable measured in the solver window
The Building Plot will highlight buildings based on which result is selected to view, including Vent Temperature, Max Temperature In, Heat Rejection, and more. The Plant Plot can display the same results as the Building Plots, but will also display results for Chiller, Generator, and Cooling Tower objects. These result plots will make reporting results much easier when determining overheating units and potential spots of recirculation.
For tracking recirculation, the Result Cloud Environment Plot is the unsung hero of 6SigmaDCX external models. The Result Cloud can display the results for variables with values above, below, or within certain thresholds.
Figure 7. Isometric view of one of the buildings focused on in this external model, with a Result Cloud displaying any locations with Hydrocarbon concentrations above 5 PPM
In the image above, I have displayed the Hydrocarbon Mix contaminant, showing values only greater than 5 PPM. Using this, we can see that there are no vents in danger of contamination. The Result Cloud can display results for any variables tracked. Potential uses for the Result Cloud include tracking high velocities to view recirculation patterns, displaying EP to show where your model could use more grid cells, or adjusting temperatures to see if generator exhaust flow is approaching building vents.
External Modeling: Key Features
In this blog, we learned how to use the External Model Wizard to set up environmental conditions for the model, outlined the improvements of grid rules in Release 13, examined the mesh generated by the software, and evaluated the results of the model using both new and old result plots to analyze how effective our facility is.
Blog written by: Max Coscullela, Applications Engineer
Other Recent Posts
CFD Simulation Should Be a Data Center Best Practice
Over the years that I have been involved in the data center industry, I have read numerous articles…
Model Server Virtualization in the Data Center Digital Twin
A recent study by Uptime Institute surveyed 250 C-level executives and data center/ IT managers at…