Read Steam condensation in annular gap steam injectors text version

Steam condensation in a 2D model of an annular gap steam injector

Andreas Nilsson

Department of Chemical Engineering, Lund Institute of Technology, May 2007 Abstract The Tetra VTIS steam injector is an annular gap steam injector used for sterilization purposes of liquid food products such as milk. The sound levels are high and knowledge of the condensation process is lacking. An experimental rig to study the condensation process is constructed. A test injector with sight glasses to make visual observations and to take photographs is used. The injector geometry is a two dimensional model of a basic annular gap steam injector. In the visualization injector steam and water is mixed with a two-phase jet as a result. The two-phase jet condenses in the water under different regimes. One regime is comparable to what can be seen in the real injector, oscillatory jet regime, with a jet oscillating back and forth in the condensation chamber. The condensation length measured increases as the level of heating increase. The length decreases with higher pressures due to density effects and higher subcooling. Higher product flows with increased mixing and turbulence as a result also decrease the condensation length.

Introduction Continuous direct steam injection is used to quickly raise the temperature of a process media, either for pure heating or for a sterilization process. The benefit of the direct contact condensation process is high heat transfer rates and low fouling compared with other methods such as heat exchangers. In the product portfolio of Tetra Pak the Tetra Therm Aseptic VTIS steam injector is an annular gap steam injector. It is used for sterilization purposes of liquid food products such as milk. The main function of the steam injector is to inject steam into the process media under high pressures to reach a desired sterilization temperature. The knowledge of the condensation process is lacking and sound levels in the steam injector are high. Therefore an investigation of the condensation process in a visualization injector is desired. The work in this study is aimed at an understanding of the steam injection process taking place in the Tetra VTIS steam injector. Tetra VTIS steam injector The Tetra VTIS (Vacuum Thermal Instant Sterilizer) is the process module where the preheating, sterilization and cooling takes place. In a direct UHT (Ultra high temperature) treatment the milk is pumped through a closed system where it is preheated, high temperature heated, cooled, homogenised and packed aseptically [1]. The Tetra VTIS steam injector is located after the preheating and it is where the high temperature sterilization occurs.

In the sterilization process the incoming product to the injector is preheated to approximately 80°C before entering the injector. In the injector the temperature is raised to about 140°C and pressures are held at >3.5 barg to prevent boiling. The product is held at this high temperature in approximately four seconds in the holding tube before it enters a expansion chamber where it is flash cooled. The flash cooling takes place in partial vacuum where the pressure is controlled so that the amount of vapour flashed off is equal to the amount of water added in the steam injection step [1]. The injector is of annular gap type (also called ring nozzle steam injector), Figure 1. The steam is injected coaxial in an angle into the product flow. The product flow is divided into two streams in the injector, one to the product gap and the other into the center and led out to the cone.

Figure 1. Tetra VTIS steam injector [2]

The product gap of the injector is adjusted with washers of different size at the back of the injector, to set the desired gap width. The adjusted gap width lets the injector work in wider area of capacities of product flow. There are four Tetra VTIS steam injectors available with capacities stretching from 800 to 35 000 L/h. Condensation regimes and jet length Previous work has reported regime maps to classify the how the condensation process takes place [3], [4]. Three distinct regimes are found in the literature; chugging, bubbling and jetting. The chugging regime is a periodic build-up and collapse of the steam cavity with large pressure fluctuations as a result. Bubbling is a build-up of steam cavity outside the nozzle that oscillates or even detaches some part of the steam. Jetting is considered to be the stable regime regarding hydrodynamic steam/water flow regime and noise levels [5]. Other regimes also found in literature are oscillatory interface regime where the steam mass flux is low; a clear interface between the steam and water can be seen. The water hammer regime which is when the steam flow rate is greater than the condensation rate and saturation temperature is reached with steam left in the condensation chamber. Most of the regime study and estimation of condensation length in literature is done in quiescent water under atmospheric pressures. In 1972 Kerney et al. [6] used a semi-empirical approach to set up relation for the steam penetration depth of a pure steam cavity. Since then this theory has been developed and used for estimation of cenodensation length in quiescent water. The conditions in these cases differ from the conditions in the Tetra VTIS where the flow of water mixes the steam cavity with improved heat transfer as a result. In the Tetra VTIS the level of heating is also higher where in the start of the jet the mixing is high with improved heat transfer. As the jet stretches out in the condensation chamber the velocities and subcooling decreases with less heat transfer as a result. Material & methods To make visual observations of the condensation process possible, a test injector had to be constructed, further called visualization injector. The injector, Figure 2, consists of a house, a cover, two inlets and an outlet. The flow through the injector is defined by two plastic inlays, formed to the desired geometry, inserted into the injector and kept in place by four bolts. The house and cover of the injector is hollow and glass plates, acting as channel walls, make observations of the process possible. The geometry of the channels used in the

experiments is similar to the ones used in the Tetra VTIS injector.

Figure 2. The visualization injector

In order to perform the experiments a surrounding system had to be set up. The experimental rig supplies the visualization injector with water and steam at temperatures, pressures and flowrates set by the experimenter. Previous experimental work at Tetra Pak with direct contact condensation has shown that there is a need for removal of non-condensable gases [7]. The solubility of nitrogen and oxygen is decreased as the temperature is raised in the steam injector. These gases are unwanted since they not only may be confused with steam but it lowers the partial pressure of the steam in the steam cavity. Therefore a deaeration branch is installed. The rig consists of injector, tank, vacuum vessel, pumps, heat exchangers and measurement devices, Figure 3. It is connected to a 5,5 bar g steam line and cooling water.

Figure 3. Process layout

Water is held in the tank at approximate 90°C. In order to heat the water in the tank steam is injected direct into the water well below surface. The water is the pumped through the first heat exchanger to the

injector. Steam is injected and the steam amount is controlled via a control valve. After the injector a control valve is used to keep the desired backpressure. Before the hot water enters the tank at atmospheric pressure a heat exchanger is used to prevent boiling. When the rig is in deaeration mode the vacuum vessel is used and a part of the water, and the air, is flashed off. The operation conditions in which the experiments are made is similar to the ones used in the Tetra VTIS process. The subcooling of the water in the experiments can be varied with the backpressure and inlet temperature. Inlet temperature in the experiments was set to 80°C, similar to the real injector, reducing the subcooling to be set by backpressure of the injector if saturation of the steam is assumed. The subcooling varies with the flow direction in the chamber where the condensation occurs. The operating conditions for the visualization injector are displayed in Table 1.

Table 1. Operating conditions for the visualization injector

The camera was set in a fixed position so that the image analysis and image processing is made much easier. The injector and plastic inlays are in the same position in each picture which allows one picture to be used to relate a distance measured in pixels to a distance in metric units, and to set reference positions such as jet start. Regimes In the experiments three classes of regimes are found: oscillatory interface regime, chugging regime and jetting regime. The latter, jetting regime is characterized by only one connected steam cavity length witch includes a wide area of jets, it is therefore divided into a number of sub regimes: Jet formation, oscillatory jet, oscillatory jet with pulsating upwards flow and jetting with jet length stretching outside the injector. A summary of the regimes found is seen in Figure 4.

Product flow Product velocity in gap Temp. inlet Steam flow Steam velocity in gap Heating Backpressure Initial subcooling

180-469 5-12 80 5-36 10-100 10-60 2.7-4.5 61-76

L/h m/s °C kg/h m/s °C barg °C

Figure 4. Summary of the regime classification

Image capturing device To capture images from the condensation process Sony XCD-X710 is used. It is a computer controlled industrial camera with the possibility to take pictures in series. The condensation process is fast so the camera is set to take pictures with a short shutter time so that the images are sharp and details can be seen. With a short shutter time the camera does not let enough light in to illuminate the object and the image becomes black. To illuminate the steam cavity both backlight and frontlight is used. The frontlight helps to bring out the details in the picture and it is possible to see the steam/water interfaces as a cloud like formation with light from front. The backlight is used to bring out the contours of steam cavity which is helpful when determine the surface area of the jet or the jet length. A camera software was developed in Visual basic, it is used for camera control, triggering the camera, triggering the lights and to process and save the image.

With low steam mass flux two regimes can be observed, oscillatory interface and chugging. An example of oscillatory interface regime can be seen in Figure 5. This regime is characterized by pure steam in the steam gap and an oscillatory steam/water interface direct at the steam exit.

Figure 5. Oscillatory interface regime

The other regime observed with low steam mass flux is chugging, Figure 6. In this regime the steam/water interface position is very irregular. The chugging regime can be described as a periodic build-up and collapse of the steam cavity. The steam cavity penetrates the water and condenses rapidly, as the steam cavity collapse a negative pressure is generated in the steam gap and water flows up in the steam channel. The steam then builds up pressure and pushes the water out of the nozzle to form a cavity again. When chugging occurred in the

experiments large pressure fluctuations in the system were observed, caused by the irregular pulses of steam build-up and collapse.

Figure 6. Chugging regime

As the steam mass flux is increased the penetration length of the steam condensation increases. The condensation zone starts to form a jetlike cloud, Figure 7.

The first jetting regime, jet formation, can also transit to another form of oscillatory jet regime, when the steam flux is increased. This regime is the oscillatory jet with pulsating upwards flow, Figure 9. This regime is also characterized by the same twophase jet stretching farther in the injector than the jet formation regime. There are also similarities with the oscillations from left to right as the first type of oscillatory jet regime, but in this regime there is a also a constant steam-escape causing an oscillatory up and down motion of the steam. The lower of the two steam cavities is more cloudlike than the upper. This is explained by the two phase jet formed when the water and steam is mixed, it appears as the water flow is not capable of mixing all the steam to a twophase jet, and some of the steam escapes. This regime was observed with low flow velocities of the water and at low pressures.

Figure 7. Jet formation regime

Figure 9. Oscillatory jet regime with upwards flow 1 and 2

With furthermore increasing steam mass flux the jet formation regime transits to the oscillatory jet regime, Figure 8. This regime is characterized by a two-phase jet with condensation lengths reaching longer than the mixing zone and a larger extent of jet length oscillation than the jet formation regime discussed above. The two phase jet occur in the lower part of the condensation zone of the injector. Further increase of the steam mass flux causes the jet to stretch further out in the condensation zone. Since this is the same type of fluctuating jet as discussed above this regime is classified as the same type of oscillatory jetting regime. The difference is the higher steam mass flux resulting in longer condensation length.

As the steam flow is further increased the condensation length stretches beyond the injector exit leaving a two phase flow in the entire condensation chamber, called outside jet regime. The full regime classification can be seen in [8].

Condensation length The jet lengths for the experiments were measured in Matlab. Figure 10 displays the comparison of the jet length with a constant flow velocity in the product gap of 8,5 m/s at different pressures. From the figure it can be seen that increased pressure leads to a shorter jet. This effect is due to the density effects (lower specific volume) and increasing subcooling. As the subcooling is increased, the water that is mixed and exist in the two phase jet can be heated to a higher temperature before it is mixed with surrounding water. This leads to that more steam can condensate per mixed water volume which enhances the heat transfer.

Figure 8. Oscillatory jet regime 1 and 2


2.7 barg


3.1 barg 4.0 barg 4.5 barg






0 0 10 20

Heating, T (°C)





comparable to what can be seen in the real injector. Increased level of heating cause the condensation to go from oscillatory interface, through jet formation, oscillatory jet regime and eventually the condensation is outside the injector. The condensation length measured increases as the level of heating is raised. There is a clear trend of higher pressures decreases the condensation length due to lower specific volume of the steam and subcooling. Higher product flows with increased mixing and turbulence as a result also decrease the condensation length. Acknowledgement This work was carried out at as a Master work at Tetra Pak with PhD Fredrik Innings as supervisor, who is gratefully acknowledged along with Prof. Stig Stenström, Dpt of Chemical Engineering LTH, who was examiner. References [1] Tetra Pak Processing Systems AB, Dairy processing handbook (1995). [2] Tetra Pak Media box at [3] Liang, K.-S.; Griffith, P. (1994) Experimental and analytical study of direct contact condensation of steam in water, Nuclear Engineering and Design, Vol. 147, 425-435. [4] Gulawani, S.S.; Joshi, J.B.; Shah, M.S.; RamaPrasad, C.S.; Shukla, D.S. (2006), CFD analysis of flow pattern and heat transfer in direct contact steam condensation, Chemical Engineering Science, vol. 61, 5204-5220.

Figure 10. Jet length comparison with constant product flow velocity (8,5 m/s) and increasing pressure

The same comparison but with a constant pressure of 3,5 bar g and increasing flow velocity of the product is displayed in Figure 11. It appears as the jets are shorter at higher product flow velocities. The higher velocities cause a better mixing and turbulence levels. This leads to increased heat transfer and therefore a shorter jet. The lowest flow velocity does not follow this trend, since it is in the upwards flow regime where the steam escapes upwards instead of stretches in the flow direction and therefore it is a shorter jet.


Jet length (mm)

5.0 m/s


6.8 m/s 10.3 m/s 12.0 m/s


Jet length (mm)





0 0 10 20 30 40 50 60

Heating, T (°C)

Figure 11. Jet length comparison with constant pressure (3,5 bar g) and increasing flow velocity

[5] Patel, R.; Varley, J.; Couch, M. (1996), Design of Stable Steam Injectors for Continuous Heating, Journal of Chemical Technology & Biotechnology, Vol. 66, 327-339. [6] Kerney, P.J.; Faeth, G.M.; Olson, D.R. (1972), Penetration characteristics of a submerged steam jet, AIChE Journal, vol. 18, 548-553. [7] Von Schenk, M. (1972), Studium av kondensation vid direktinsprutning av ånga i strömmande medier [8] Nilsson, A. (2007), Steam condensation in a 2D model of an annular gap steam injector, LTH, Lund

The heat transfer coefficient is estimated for the experiments with a measurable jet length. In the calculation of the coefficient the area of the jet is defined as the jet lengths multiplied with the depth of the channel due to the difficulties of determine jet area. In the experiments the estimated heat transfer coefficient varies from 0.2 to 2 MW/(m2 K), which is in the same order as in [4] and [5]. Conclusions The different regimes observed in the visualization injector can be divided into three branches, oscillatory interface regime, chugging regime and jetting regime which includes a wide area of jets. One of these, oscillatory jet regime 1 is


Steam condensation in annular gap steam injectors

5 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

Steam condensation in annular gap steam injectors
Time-Lapse Monitoring of Steam Assisted Gravity Drainage (SAGD) of Heavy Oil using Multi-Transient Electro-Magnetics (MTEM)
Chapter - Steam Distribution and Utilization.doc
Microsoft Word - 2009 PMO-FINAL.doc