CFD Assessment and HVAC System Design for Buildings

HVAC Building

Every good HVAC (Heating, Ventilation and Air Conditioning) system design must be able to ensure a proper air renovation inside the building it is installed at the lowest possible cost. This objetive is reached through the correct positioning and orientation of the supply and extraction air systems, and an adequate value of air changes per hour, also called air change rate. The use of CFD to assess and optimize the installation’s design is paramount in order to reach the best results.

There are different measurement parameters that allow the evaluation of the ventilation system quality:
First of all, the above-mentioned number of air changes per hour (ACPH) in the building. It is defined as the rate of the air volume added to a finite space in one hour divided by that spaces volume. If a perfectly mixed model is considered, the air renovations per hour parameter is a measure of how many times the air in this space is replaced each hour.

This parameter, despite being useful, may not represent completely how renewed is the air inside the space, since the disposal of the ventilation systems, as well as the elements inside the space, can generate flow patterns that make some areas be under ventilated. For this reason, it is also useful using the concept of age of air.

The local mean age of air is defined as the time that particles contained in a differential volume around a point (as is in the case in a cell of a CFD simulation) have been inside the space. 

If it is assumed that the age at the inlets is 0 (which equals to say that pumped air is completely new), this parameter evaluates the residence time that particles spend in the building from they enter until leaving.

Air Age Contours

This value does help assessing whether areas are renewing the air more frequently, since the residence time in them is lower. From this value, it is possible to obtain the mean age of air (MAA) in the building, which is calculated as the average of the local mean age for each point in the space.
The efficiency (ε) of the ventilation system is defined as the ratio between the minimum time that a particle spends in the space from the input to the output, and twice the mean age of air. This concept allows to seek a balance between a good air quality and a reduced air flow. 

HVAC Efficiency Formula

In addition, the evaluation of the minimum residence time enables the possibility of tracking for short-circuits of new air in the system. Along with this parameter and assuming a complete mixing model it can be stablished that:

  1. The optimal value is obtained at ε=50%
  2. For ε<50%, it can be considered that the space is lack of ventilation, or that the minimum residence time is too low, indicating a possible air short-circuit
  3. For ε>50%, it can be considered that the air Flow ratio is over dimensioned and it is advisable to decrease it

CFD techniques allow not only to determine the age of air in a room. It is also possible to introduce pollutant emission sources, chemical reactions and thermal effects in the models. This way, the pollutant concentration in the area of study can be analyzed. This measure can provide an even more valuable information than the age of air in terms of predicting more accurately which areas of the room are not being properly renovated, and optimize the HVAC installation from the pre-design stage. 

The concentration of a chemical compound in a large volume of air is often measured in milligrams of compound per cubic meter of air (in some cases per kilograms) giving a unit of particles per million (ppm).

To put this in perspective, let’s consider a Waste Water Treatment Plant (WWTP). These facilities are attached to a series of emissions of contaminants, perhaps the more characteristic being the hydrogen sulfide (H2S), produced during the decomposition of some amino acids, as well as the reduction of sulfates to sulphites by certain microorganisms.
Hydrogen sulfide is a colorless, flammable gas with an atomic weight of 34 g/mol and a density of about 1.5 kg/m3 at ambient conditions, being slightly heavier than air, tending to accumulate near the ground. Its odor is very unpleasant and it is associated to “rotten eggs”.

Contaminant Contours

The limit of perceptibility for human beings is 0.02 ppm, with some people being able to detect it at 0.0005 ppm. At higher concentrations can result toxic, produce metabolic changes and even cause death. In addition to this, it is one of the main causes of corrosion in this type of facilities, attacking in moist ambiences to iron and concrete with ease.
The dispersion effect that the air impulsion from the HVAC system can cause over these particles can be properly analyzed with a CFD model, along with the study of alternatives in the flow inputs, system location and effect of the different sources. The SDEA_Engineering team presents a broad experience in the HVAC and Computational Fluid Dynamics field and can help in the design assessment and study of the HVAC system.

Condenser retrofitting relying on CFD

Condenser CFD

Sometimes, pressure equipment life needs to be extended above the initial design requirements and that represents an engineering challenge. Even less frequently, the design point also changes, making a priority to understand if the original design can withsthand the new operational settings without eroding the equipment performance.  

This article provides a rough view of a project developed by SDEA where CFD modelling was used including a multi-phase solver and in-house developed functions to replicate the real behaviour of the condenser operational conditions. 

This condenser was part of a combined cycle power plant using a Rankine cycle. The aim of the low pressure equipment is to reduce the turbine back pressure and condensate the steam before it gets back into the boiler. A large number of tubes are placed inside the condenser, with a cooling fluid [brine in this case] running inside them. They are below the incoming steam dew point that takes energy away, condensing and separating by gravity the created droplets.

Condenser Geometry

Condenser geometry simplified for CFD

Most condensers have been designed using analytic methods, which usually are lumped models that can reproduce more or less the overall performance but can not provide detailed data on the insights of the process. The main drawback for this approach is that small details like the tube bundle geometry or undetected stagnant high-aged fluid are critical if an optimized condenser is aimed at. The main solution is to build laboratory prototypes and do physical test with the associated high costs.

SDEA engineering team developed a comprehensive CFD model to emulate the condensation process with a high level of detail. Nevertheless, the vessel has two tube bundles with 14726 tubes each, resulting in very expensive computational model if  every single tube was to be included in the final CFD model. 

Bearing in mind the previous information, the main objectives of this study were to predict the velocities, temperature, density and pressure drop distribution inside the condenser for different operating conditions: those for the original design and for the upgraded normal conditions.

Due to the time scales involved in this project a simplified strategy was carefully selected after an exhaustive literature review, in order to achieve a good balance between accuracy and model complexity. 

Condenser CFD Mesh

Condenser geometry CFD mesh

In this section the main assumptions will be explained and justified in order to provide a glimpse of not just SDEA’s CFD capabilities, also the physics understanding and the skills needed to carry on with projects like this and to complete all the goals stated in the project. 

The cooling system is composed by two different bundles with more than 14K tubes. Modelling each single tube is unaffordable, so a porous media approach was assumed for this CFD model. This method does not require solving the detailed flow and temperature fields around each tube.

The tube bundle is regarded as a distributed resistance region. The resistance is based in the theory developed by Patakar and Spalding. Many different models have adopted this approach and it proves to match experimental data with reasonable accuracy.

Porous Media CFD

Porous media submodelling

Porous media for the tube bundle region is modelled by the addition of a momentum source term to the standard fluid flow equations. The source added is a two terms expression including a viscous loss term and inertial loss contribution. Next equation details thematrix form included in the momentum transport balance for the tube bundle and the impingement rods region. 

Due to the obvious anisotropic resistance of the tube bundle an orthotropic [different behaviour on each of the three axes of a Cartesian reference frame aligned with the tube bundle] scheme was adopted. 

A local model was built for each direction and exposed to a wide range of velocities to obtain the equivalent pressure drop for each velocity field condition. The thermal resistance for the tubes has been calculated according to values and correlations obtained by several research papers. These include various terms in order to get the total value of the thermal resistance.

Thermal Resistance Condenser CFD

Thermal resistance vs cooling water velocity plot

The water side thermal resistance, Rw, is a function of the coolant water conductivity, flow characteristics and internal diameter of the tube.

The fouling thermal resistance, Rd, has been taken as an average value typically found in this type of condensers in previous research, with the usual value for the tubes heat transfer coefficient varying between 10000 and 35000 S.I. units depending on the position in the bundle.

Thermal resistances detailed above is an important issue. The heat transfer between the cooling tubes and the steam is important, in this matter different phenomena contributes on the overall energy transfer and the condensation flow rate depends due the mass transfer model used  strongly on that heat balance. 

The fluid is modelled as a single phase multi-component gas composed of steam; non-condensate was not included in the model for simplicity.  The condensation mass sink term for the steam fraction is then computed based on the local heat transfer in each control volume according to the frequently used thermal resistance mass transfer model. 

The thermal resistance mass transfer equation included T as the local gas phase temperature, L is the latent heat of condensation of the steam, A is the heat exchange area of the tubes in each control volume, V is the volume of the control volume and R the thermal resistance previously described.

A linear mean coolant temperature profile, Tw, is assumed along the tube bundle and is used to compute the heat transfer between the coolant and gas phase.

Final CFD results including plots and contours

After running different operational conditions cases, a comparison of velocities and temperatures was done to ensure the propper behavior of the condenser under the new set of conditions. Image above shows a glimpse on the type of results and insights you can get from an advanced CFD model like the one devoloped in this case.

Keep your curiosity in good shape.

CFD in HVAC for railway applications

CFD simulation for HVAC studies

When travelling, arriving to your destination on time is not the only concern. Comfort also plays a very important role on the travelling experience and may have an impact on your evaluation about the trip.

Railway transportation is one of the most used systems in the world, from high-speed trains allowing people to travel long distances quickly to underground railway facilitates mass transit for people movement inside a big city.

What all these types of trains have in common is their need for a Heating, Ventilation and Air Conditioning [HVAC] system to ensure environmental comfort inside the vehicle. In order to regulate these conditions, there is a strict normative related to comfort for both the passenger cars and driving cabins.

These regulations, specified in Standards such as EN-13129 or EN-14813, determine the comfort parameters inside the vehicles related to air temperature, humidity, air renovation and surface temperature in a wide range of locations, with the final objective of guaranteeing the maximum passenger comfort during the travel.

Computational Fluid Dynamics [CFD] is a powerful tool that can help assessment in the design phase of HVAC systems, since it can simulate the flow and temperature distribution inside the carriages and any environmental and input conditions even before the system is built. With these previous evaluations, it can be easier to determine optimal flow distribution, impulsion parameters, flow rates, poorly optimized areas and so on. CFD also counts with the advantage of not being restrained by any physical components, so data can be analysed in any spot of the geometry, giving even more additional information that can be useful to the designer than the parameters asked by the Standards.

Both EN-13129 and EN-14813 stablish the probing coordinates and sections during test conditions; so in the CFD model these locations can be fully monitored during the calculation to ensure the convergence and stability, and later on, these data can be compared to the requirements asked in the Standards.

Seating Measuring Points

HVAC systems have to meet an exhaustive process of evaluation in terms of thermal homogeneity in a large amount of different locations: seats, aisles, bathrooms, windows, walls, etc. in order to provide a comfortable feeling to the passengers along the ride. Air velocity is also regulated in order to prevent that any area is exposed to any jet stream or, on the contrary, be improperly conditioned. All these requirements involve a deep understanding of fluid dynamics and air behaviour.

SDEA Solutions provides a fully customized service using its knowledge on CFD HVAC modelling, simulation and expertise on Standards to fulfil any client’s request in HVAC systems assessment. Our services include the whole process of designing the geometry from drawings or modification of a given one to the creation of a high quality model to be simulated and a complete analysis of the results, giving the client the precise data they need to know in order to better understand and improve their system, with complete and exhaustive reports made to ensure that every condition and result is well interpreted.