Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
This is designed for introducing the displacement ventilation system and its related issues to colleagues and the person who might be interested in this techniques. This technique is getting more popular inside the company and naturally it comes to the point that it is wise to identify several crucial issues as well as technique itself for standardization. I strongly believe that it should be updated and evolved to the level that any new comer in this field can catch up with a right focus quickly.  Calculating Qoz , gravity currents, temperature gradients, stratification issue, considerations of model experiment, ventilation efficient, other important area are discussed in the later section.
Q: What is the phenomena behind the displacement ventilation?
Q: What is basically the differences between DV and mixing ventilation?
Q: What are the advantages of DV over mixing ventilation (MV)?
Q: What factors should be considered in DV design?
Q: In designing process, what is the main difference bwtn DV & MV?
Suppose there is a man in a open space, and he is exercising now. His hot skin  produces sensible heat and he is also sweating to the environment.  Also as he breathes, he produces pollutant.
If there is no buoyant force effect, what could happen to this man?
As he inhale and exhale, the CO2 is accumulated.  And that CO2 is diffusing to the environment but accumulating rate will exceed the dilution rate.
Therefor, this man unfortunately, will die unless he has no special equipment like fish.
Fortunately, since we are living in gravitational field generated by Earth, upwarding  plume is generated due to density difference coming from temperature difference between air around human body and environmental air.
The buoyant force is:                 
For air: from 32F to 100F:           is 0.1234
For water from 32F to 100F:         is 0.0064.
That is, in the air, buoyant force are 20 times stronger than that in the water per unit mass.
Look at the plume generated: 1. It acts as mechanism of transfer between   lower region and upper regions.  2. Due to entrainment, as it goes up, volumetric flow rate increases. However, since it is open space, it is not likely that the hotter and contaminated air is not going to return to him again.
Now, suppose that this person is in the enclosed space. The hotter and contaminated air start to accumulate in the upper region of the space.
This is the situation where more fresher and tempered air should be provided to this place.
There are two methodologies imaginable.
1. Provide fresher and dilute contaminant in the space by mixing and uniformize the filed parameters (temperature, contaminant, humidity, etc.)
2. If it is possible that the hotter and contaminant air can be entrapped in the upper region, the air had better be separated upper and lower region.
For 1, we need high velocity jet to enhance a mixing effect, therefore the location of the terminal should be away from human occupant.  This is mixing ventilation. For 2, we need the pool of the cool air at the floor level so that plume is generated naturally. This implies that location of diffuser should be close to floor level and since it is also close to occupants the air terminal velocity should be low. (gravity current).
Check: plume progress, entrainment, deflection
 This slide shows schematic conceptual drawing for mixing ventilation.
Supply air should be designed to absorb sensible and latent heat in the total conditioned space.
 All parameters are uniform throughout the space due to intense mixing by supply jet.
 CFM is usually obtained based on total sensible heat, however in case of presence of strong latent heat source, CFM based on latent heat should be considered and compared.
 The conditions of exhaust is assumed to be the same with the room condition.
 This slide shows a schematic drawing of displacement ventilation system.
Stratification level forms.
 Supply air should be designed to absorb sensible and latent heat in the occupied  space.
 All parameters are varying vertically throughout the space due to naturally generated convective plume.
 CFM is usually obtained based on total sensible heat only affecting the occupied zone. Here, strong convective plume energy and convective heat generated in the unoccupied zone are excluded.
 The conditions of exhaust is different from the room conditions.
Qoz = Q1+Q2+Q3
 Q1,2,3 are not easily, clearly estimated since it varies according to its vertical, horizontal location, its distribution and configuration of geometry.
 Q3 appears in the forms of convection from the floor and walls.
 There are two methodologies for Qoz
 Macroscopic approach: starts from Qtotal and try to make global weighting factors. It could work for similar geometry but not universal.
 Microscopic approach: starts from individual heat source and try to make local weighting factors. It could be more accurate but requires intensive investigation.
This slide shows the effect of heat source location on cooling load in displacement ventilation.
The central located small box produces heat and it is well shown that this heat source produces plume. Diffuser is located at the lower left corner and exhaust is right upper corner. The supply air temperature and flow rate and the amount of heat generated by heat source are the same in Case 1, 2, and 3. It is easily and clearly noticed that as the heat source is higher, the temperature fields are changing. Basically since the plume starts from the physical existence of the heat source, the lower-than-heat source region is pool of the supplied air (20C). This result conforms the fact that if the height of the heat source is higher than that of stratification level, practically its convection heat into the lower zone can be neglected for cooling load calculation unlike the conventional ventilation (mixing type). However radiation heat transfer from heat source should be considered. (This study does not include the radiation effect). Take a look at the shape of the “gravity currents” generated by low velocity diffuser from each case 1,2,3. In Case 1, we could see the gravity currents deflected downwards. And Case 2,3 we could not. Why is that? The answer to this can be found if we consider Archimedes number (Ar. No.). (seeDiffuser.ppt” in G server for the definition of Ar. No. and the behaviors of the gravity current according to different Ar. No.). The Ar. No. in each case are approximately 58, 30, and 10 respectively even though the supply temperature and flow rate are the same. That’s mostly due to the temperature difference between the supply temperature and environment temperature. This number act as a barometer for the deformation of the supply air flow. In  Case 3 (Ar. No. 10),  we don’t see any deflection of the supply air. In the other word, there is not enough buoyant force pushing supply air downward because the lower region is already filled with the same temperature.
Figure 6 shows cooling loads reduction when the source is getting higher. Note here that stratification temperature was not set up as a same constant. Rather, the supply temperature and flow rate are same and cooling loads are different so that the temperature at the stratification level are different. Therefore we have to be careful in interpretation of Fig. 6. This fact implies that in case3, if we set up the temperature @y=1.5m higher than 23C, we could reduce the flow rate to achieve cooling.
Gravity current
 Less entraintment, less mixing effect compared to high velocity jet. It results in relatively uniform volumetric flow rate. -- it causes higher velocity than diffuser discharge velocity at far field region as shown.
 It spreads more effectively than high momentum jet.
 The shape and behavior is characterized by Archimedes number (gDTd / TV2)
 Two factors
 Depth of near field, XD. (defined  by a distance where V=0.25m/s at 0.1m)
 Far field velocity, Vf.
Temperature gradient (no sharp stratification)
 It is known that the temperature gradient is non linear and dependent on supply airflow rate as well as supply air temperature.
 However, it can be represented with two linear temperature distributions without loss of any generality across the stratification level.
 It should be kept under 2~3K/m as a requirement.
 Pollutant gradient (usually sharp stratification)
 This is considered as one of most advantageous aspects of displacement ventilation, However, this is the case where pollution source ic coincident with heat source.
 It is reported that moisture follows the similar trend with pollutant. (by Halton).
Now we like to know Tf, Tf, air, Tc and temperature gradient in term of known quantity. Since we have three unknowns and we like to derive dimensionless temperature, we need to have two energy balance equations.
where hr and hf are the heat transfer coefficients for radiation (between ceiling and floor) and convection (between air and floor surface). And r is density of the supply air and V is volumetric flow rate and Cp is the specific heat.  Note we linearize radiation heat transfer, otherwise it is fourth power.
Based on the equations (1) and (2), we could drive dimensionless temperature,
In this model,  we discriminate the exhaust temperature, ceiling surface temperature, ceiling area air temperature. This is more realistic model since due to radiation heat exchange between ceiling and floor, actual surface temperatures are a slightly beyond at the floor and below at the ceiling. Even though the number of unknowns is five, this is considered as a four node model because Tc, air can be written in terms of Te.
Furthermore, we could still apply energy balance equations (1), (2) for the floor. Additional energy balance for the ceiling part is,
With the equations (1), (2), and (7), after a little modifying these equations, we get
Then, we got the other nodes’ temperature,
Stratification Level
 Analytically, the stratification level is formed based on mass conservation,
Qsupply = åQi = Q1+Q2
  Note that volumetric flow rate of the plume is varying (due to entraintment) according to heat source’s height. What it means is that the height of the heat source is very important factor to form the stratification level.
 What stratification implies here is that across this line momentum exchange is assumed to be minor, not zero. Some researchers consider this effect but too expensive.
 Once the CFM is calculated based on sensible heat, we need to check if this CFM gives a right stratification level.
This is the issue that must be considered before going further more since the presence of the stratification level is a must in displacement ventilation. In most cases in HVAC application, we take it granted that stratification is formed.
However, it is possible that there is no stratification. Several researchers (Akinchev, 1966; Ratter and Strizhenov, 1968) thought that there is just a single circulation zone in a space.
Then, what is the case that there is no stratification?
These equations imply that rarefaction (DPjet)  is strong enough to push heated air (by overcoming buoyant force, D(rgh)) to the lower zone.
There are two kind of ventilation efficiencies appeared in the literature. one for the ability of a system to exchange the air  in the room (1) and the other is for the ability of  a system to remove contaminants (2).
To properly design thermal displacement ventilation, we need to quantify several key parameters. Among these, thermal displacement cooling load, supply air flow rate, supply air temperature, temperature gradient, temperature at knee and at head level, exhaust temperature. It is then a first and most important step to quantify cooling load. And then we need a tool (model) to predict other parameters. There are several suggestions to approximate the cooling load in displacement ventilation but none of them are very promising.  As for the reason why it is difficult, ASHRAE Research Project -RP-949 (pp.15 and 65) and “Review of Design Procedures for Displacement Ventilation System”  (pp.5,6).  We will investigate more on this and compile the existing information for the next step. Here we will adapt the macroscopic approach. Next step is to model the space to predict other key parameters. Complex model is not necessary better that simple one since it requires more sophisticated input data which is rarely found analytically. We will investigate existing model and propose new model.   It is the purpose of this note that we compile all the pieces of information which is spread out in the papers but relevant to our goal and propose a new way and model and finally attempt to make one set of procedure for the displacement ventilation system design.