Transfer line, cool-down

In our previous work with horizontal liquid transfer systems, we developed a simplified line cool-down model (Fig. 3) which permitted us to predict line cool-down time [3]. 

Since the line cool-down model enabled us to predict cool-down times successfully for our test system, we concluded that heat transfer rates across the nitrogen film did not limit line cooling. However, we did obtain heat transfer coefficient data at several locations in the piping system. We estimated the coefficients from line wall and flowing fluid temperature data. Data from horizontal... 

Fig. 5. Heat transfer coefficients line cool-down of 4 In. system.

At the start of line cool-down, high values of film temperature gradient are obtained. Since the line fluid is primarily gaseous at the beginning of cool-down, the data points obtained represent gas film coefficients (at velocities less than 100 ft/sec) and are relatively low. As line cooling continues, the temperature gradient across the film decreases, and a simultaneous increase in the amount of liquid present in the coolingfluid stream causes an increase in the heat transfer coefficient. 

Line cool-down also accounts for a considerable loss of liquid during transfer operations. This is especially important in large test operations. In one case a 13 loss of liquid oxygen due to boil-off was reported in transferring the liquid from railroad tank cars to the test site 30 miles away. The cool-down loss was 30 of the test-stand loss at this site [10]. Losses due to liquid-hydrogen transfer and storage are mentioned in (A-4) and (A-6). 

Solution. Equation (7.98) can be used to estimate the time required to cool down the transfer line. This will necessitate the determination of rhg, hg, p s, Wss> P/> and Ui. The transfer line cooled with saturated liquid oxygen under a pressure of 0.405 MPa will attain a final temperature of 105.9 K. The mean gas temperature at the exit is... 

LNG was stored on site in a large insulated tank (see Fig. 4). Prior to a test the tank was either self-pressurized or pressurized with nitrogen gas, and some liquid is forced into 53 m of 25-cm-diameter pipe before the control valve. A small by-pass line around the control valve permitted LNG to enter the final 41 m of transfer piping in order to lower its temperature. Following the cool-down, the control valve was opened to allow liquid to flow and spill onto the surface of a small pond. The horizontal... 

Two-phase flow is always involved in the cooldown of a transfer line. Since this process is a transient one, several different types of two-phase flow will exist simultaneously along the inlet of the transfer line. Severe pressure and flow oscillations occur as the cold liquid comes in contact with successive warm sections of the line. Such instability continues until the entire transfer line is cooled down and filled with liquid cryogen. 

The diluted sulphophenol is cooled down to 35°C before being transferred by compressed air to a sulphophenol tMik, heated by a heating coil. The pipe line through which it passes should also be heated up. 

However, rheokinetic effects cause the develproducts accumulate on the walls and in the axial zone the flow is accelerated, i.e. the feed rate of the reactants increases. As a result, the Vf = U,. equilibrium is violated, the front line is distmted and its central part is displaced towards the ouq ut. Consequently, the temperatiure becomes lower, the rate of combustion dr<, and the feed—combustion equilibrium is violated still more. Also, the frcmt region is cooled down and is transferred out of the tube. Therefore, for a rheokinetic liquid (polymerizing medium with a sharp viscosity growth), a low-temperature condition for the process is the only steady-state solution. The polymerization front normal to the flow can exist only as an unsteady state and this solution is unstable. 

Due to the continuous transfer of heat to the fluid along the length of the test section, liquid was vaporized as it flowed down the test section, and the slug flow phenomenon at the exit of the test section was not observed until a later period. By this time, the line had cooled down considerably so that the temperature oscillations (between vapor and liquid) were less pronounced and liquid residence times could not be determined as readily. However, from the vapor fractions calculated for the inlet to the test section, those at the outlet can be computed from energy balances. 

Fig. 24.2 Transfer of entropy 5t in a reversible cycle from a cold to a warm reservoir. Changes of volume are indicated by arrows (initial state contour line solid, final state contour line dashed). More heat Q flows off with the entropy St than in Qout > l2m. even though the body completely reverts to its initial state after every cycle and does not cool down at all. This means that energy is emitted as heat, which was not present in that form before but is generated. The question remains what phase of the process does this happen in and how ...

In missile-loading and certain other transfer systems, large quantities of cryogenic liquids must be transferred in a relatively short time. Such systems must be cooled to the point at which transfer is accomplished as a liquid rather than as a gas in a fraction of a minute, or, at most, in a few nqiinutes. Four significant factors influence the cool-down time of a transfer system (1) transfer pressure (2) heat leak to the fluid from the environment (3) line flow impedance and (4) quantity of system mass that requires cooling. These factors have been studied experimentally by Burke et al., who have developed an equation for predicting cool-down times of horizontal lines. 

We applied this model to the new transfer line configuration and found that cool-down times could be predicted with at least 15 accuracy. Some of our results are shown in Fig. 4 and Table I. 

We note that the model is based on the assumption that cool-down is limited by the rate at which gas is vented through the line end restriction. For purposes of the model, heat transfer coefficients across the fluid film next to the inside pipe wall are considered infinite. 

Heat transfer coefficients as high as 300 Btu/hr-ft - F and average heat transfer rates as high as 30,000 Btu/hr-ft were observed during cool-down of our experimental transfer line. The equivalent cool-down rates are limited by the capacity of the system for venting cool-down gas rather than by the heat transfer. It is possible that heat transfer rates could limit the cool-down rate during even a faster cool-down. 

J. Burke, W. Byrnes, A. Post, and F. E. Ruccia, "Pressurized Cool-down of Cryogenic Transfer Line," Advances in Cryogenic Engineering, Vol. 4, K. D. Timmerhaus, (ed.). Plenum Press, Inc., New York (1960). 

Other cryogenic liquid losses in transfer operations result from liquid trapped in the system after loading and unloading. At the Edwards Air Force Rocket Engine Test Laboratory this liquid loss originally amounted to approximately 20 of the test-stand losses [10]. Trapped liquid left in the system may be reduced by the use of auxiliary line and pump sections in the transfer system, permitting return of unused liquid to the storage tank. This may precipitate additional cool-down losses if liquid remains in the test tanks after a run. Accurate control of test liquid requirements will help to minimize these losses. 

It should be mentioned that the problems associated with cooling down the transfer line and establishing single-phase flow require further attention. Analytical or experimental studies should be made to determine whether singlephase flow can always be attained by throttling or whether special venting must be provided to bypass the boil-off gas from the system. 

A brief analytical study will be presented to show how flexibility for cryogenic lines may be obtained by the use of expansion loops and expansion joints that satisfy both the requirements for uniform thermal cool-down and bowing considerations. Finally, the liquid-oxygen transfer piping developed for the Atlas missile launching complex will be presented and special features of this system will be explained. 

For uniform thermal cool-down, a 12-in.-diameter stainless steel thin-walled liquid-oxygen transfer line in a U-bend configuration with a value of Yl equal to 360 in. requires a length/f of 140or 150in., depending on whether allowable piping... 

---