Document 159grgn33qrnkLj05Yeb2rJdo

788 CHAPTER 72 1962 Guide And Dafa Book other part of the system. It may be drained into a run down receiver which is disconnected from the suction pressure when the level has increased to a preset point and connected to the discharge pressure which serves to push the liquid back to a point where liquid is wanted, such as the low temperature evaporator, the receiver or liquid line. In some cases a heater is installed-to evaporate the liquid. The heat might be provided by a coil, n*nng the warm con denser liquid, if the quantity of liquid from the knock-out drum is not too great as compared to the quantity of warm liquid available. If the refrigerant is boiled out of the liquid which has drained into the knock-out drum, there is a probability of a quantity of oil remaining which must be returned to the com pressor, by injecting it into the suction line or, if relative pressures permit, directly to the crankcase. It is good practice to heat the oil or otherwise insure that it is above the tem perature corresponding to tire saturation pressure at the crankcase before it is returned. This heat can be, and often is, supplied by warm liquid by means of a heat exchanger. Some times heat is supplied from another source, such as electricity. This is not as efficient but its convenience may make it pref erable. Horisontal knock-out drums are also sometimes used. They may have baffles or mesh eliminators to remove the liquid from tire gas flow. Pipe Sizing PIPING liquid lines are generally deed for Emits of friction loss and possibly to avoid excessive velocities, particularly where quick shutoff valves such as solenoid valves are used. The discharge and suction gas lines are generally sized for rea sonable friction losses in the case of horizontal or downward flow lines. Other treatment may be necessary where the lines result in an upward flow. This case is discussed in a later paragraph on arrangement and size of piping for oil return. Pipe friction data are available for Refrigerants 12,.22 and 717 in Chapters 70 and 71 of this book, and in a number of other sources. Similar data for Refrigerants 50, 170, 290, 600, 601 or 1270 are available in petroleum industry handbooks. For .Refrigerant 13 and some of the newer refrigerants, there is not as yet much prepared data available. Friction losses can be calculated from basic data such as the Moody curves when Reyholds number and relative roughness can be deter mined, but this takes more time than is desired in many cases. It is often assumed that friction for Refrigerant 13 is reasonably close to that of the other halocarbon refrigerants under conditions where gases at similar specific volumes are oompared. Results obtained from such an assumption will probably be useful but it should be always recognized that it is an approximation and consideration must be given to the effect of deviations of several percent from the true values. Arrangement and Size of Piping for Oil Return Oil return in the suction line of multi-stage systems en counters tire same difficulties as in other systems as long as the oil remains reasonably fluid. For appEcations where the oil is at such a low temperature that its viscosity is high enough so that it will not run, it cannot be expected to return with the gas stream and must be separated, warmed up, and then returned to the compressor through separate means. Oil which is low enough in viscosity so that it will flow will Table 1 .... Sizing Data for Oil Return in Discharge or Suction Lines With Row Vertically Upward*** Soforofroa Temp, F Una Siza 2 in. and ten Abov* 2 At SO and above 0 -50 0.25 psi/100 ft 0.35 psi/100 ft 0.45 psi/100 ft 0.15 psi/100 ft 0.20 psi/100 ft 0.25 psi/100 ft thes*eFvreichtieiosn. ratoe*. U minimum operating flow rate, should aot be ten then * From SiitUot Datign and Imtctlafien Manuel, Worthington Cwmitj^ encounter no problem in returning with the gas stream through horizontal lines or lines which are pitching down ward in the direction of flow. Where the oil must flow up in on upwardly pitched or vertical line, it is necessary to main tain a certain minimum gas flow velocity which will exert sufficient friction drag on the oil at the pipe wall to cause it to move in the same direction. There are curves available for the minimum upward velocities for Refrigerants 12 and 22 (Chapter 70) but these are not generally available for other refrigerants. ' To cause oil to flow upward along the walls of a pipe re quires a certain minimum drag of the gas flow agahist the pipe. This is represented by some minimum friction gradient of.the gas flow in the pipe line. Table 1 shows values for mini. mum friction gradients which have been found to be satisfac tory in practice. Velocities which cause these friction gradients are very similar to those given in Chapter 70 on refrigerant piping systems for Refrigerants 12, 22 and 500. To illustrate the use of Table 1, assume that the refrigerant flowing upward in a vertical line is at a pressure correspond ing to zero F. 11 the line is likely to be larger than 2 in. diame ter, oil can be expected to flow in the direction of. gas flow if the line is sized to result in a friction of at least 0.20 psi per 100 ft. CONTROL Liquid Feed Control In addition to the discussion in Chapter 36 of the 1961 Guide Arm Data Book the following comments apply, particularly to low temperature or low absolute pressure applications. If Uquid flow is controlled by means of low-dde float valves, having their floats directly in the chamber where the level is being controlled, the low pressures and temperatures have no appreciable effect on their operation. External float cham bers, however, must be thoroughly insulated to prevent best influx which might cause boiling and an unstable level affecting the float response. Equalizing lines to external float chambers, particularly the upper line, must be gen erously sized so that liquid can reach the float chamber and the gas which results from any evaporation can return to the vessel with no appreciable pressure loss. (A loss of 0.1 P51 corresponds to about a 2 in. column of Refrigerants 12, 22, or 13 and a 4 in. column of Refrigerant 717.) Constant pressure expansion valves can be used if the sys tem has only one evaporator, but it must be recognized that at low absolute pressures very small differences in presure will have significant effects on the saturation temperature of the refrigerant. For the direct-expansion type of evaporator the superheat controlled expansion valve (thermal or thermostatic) is most `(!efr>geront Piping Systems for Multi-Stage Applications, 789 rfcely to be used. It is important to realize that this type of --ive actually operates on a pressure difference between that > the bulb which is responsive to the suction temperature j below the diaphragm which is actually the suction pressure. Thus a Refrigerant 12 valve which operates with 10 F deg superheat at suction pressures corresponding to 40 F. ^ouid require about 16 F deg superheat at.an.evaporating pressure corresponding to 0 F. Since it is often the case that 0e fluid being cooled is less than 16 F deg warmer than the evaporating temperature, there would be no possibUity of providing 16 F deg of superheat to make the .valve operate. ^There are two of avoiding this problem. One is to certain cross charges (charged with a fluid other than the gystem refrigerant) in the power elements for the superheat control valve, for which one is referred to the manufacturer's data. Another solution is to obtain the necessary superheat by providing a Uquid to suction heat exchanger at the outlet of; the evaporator. The expansion valve bulb is then placed on the compressor side of the heat exchanger so that the warm liquid being supplied to the expansion valve will give up some of its heat to the suction gas and thus provide superheat for the valve. Within the Emits of satisfactory oil return as previously mentioned, this liquid-suction heat exchanger should be pitched slightly upward in the direction of flow. Hie gas flow cross section in this heat exchanger should be approximately the same as that of the main suction line. Double-pipe heat exchangers and some other types satisfying tfroeA requirements have proven satisfactory for this service. Shell-end-coil or shell-and-tube heat exchangers with liquid in the coil or tubes are usually unsatisfactory. With a properly sized and designed liquid-suction heat exchanger it is practicable to use a liquid-charged type of superheat controlled valve to rather low temperatures. This is desirable in that the same valve is satisfactory over a con siderable temperature range. There is sometimes excessive hunting with this control, particularly if the exchanger is not of the best type. A part capacity bypass around the expansion valve or a constant source of heat (such as electric heat) to the suction line pre ceding the bulb, reduces the hunting tendency. Sizing and selection is critical and adjustment is usually required. Although the thermal expansion or superheat controlled valve is designed to maintain a preset superheat in the suc tion gas, it must be kept in mind that the pressure sensing part of the system responds almost immediately to any change in conditions whereas the temperature sensing bulb must overcome a definite amount of thermal inertia before its effect is felt on the power element of the valve. For this rea son when sudden compressor capacity increases occur, it is' possible for the evaporator to boil over and the expansion valve to continue overfeeding for a few seconds before the bulb has wjvH the presence of liquid in the suction line and cut down on the feed. It is for this particular reason that it is necessary to use knock-out drums, which have previously been discussed, on direct-expansion systems, especially low temperature systems. Controlling Load During Pull Down Since a compressor can handle a greatly increased mass flow of refrigerant at the higher absolute pressures correspond ing to the starting conditions than it can handle at the low absolute pressures in the final operating conditions, some degree of capacity control must be included if very large motors and very large condensers are not desired. If the motore nnH condensers are sized for the maximum rate of flow that ran be expected during all stages of the pulldown pe riods, they will be considerably oversized for the final low temperature operating condition. In addition, if the com pressor is allowed to operate at its full capacity during the puli down, the resultant rapid removal of the refrigerant gas from tiie evaporator will result in seriously increased prob abilities of suction linfi glop over. One method of avoiding these problems is to use compressor.. capacity control or compressor cylinder unloading. Properly controlled, this will reduce the capacity of the compressor to stay within the capacities of the motor and the.condenser. Another method is the use of a hold-back valve installed in the suction line which will throttle the flow as required to limit the suction pressure at the compressor to a predeter mined maximum pressure. This would.cause the compressor to operate at a controlled maximum suction pressure through out the pull down period. Use of the hold-back valve in multi-stage systems to pre vent overloading during the pull down period has one distinct disadvantage. Such a valve will impose some friction resis tance in the suction line even when the valve is fully opened. Auxiliary controls may be used so that the pressure drop will be only that required to pass through the fully open valve and not that required to hold the valve in the open position. Even this amount of pressure drop can be significant in re ducing a compressor capacity on very low absolute pressure systems. If pull down is required only occasionally the cost of the condenser and driving equipment can hardly be justified for the amount of time which must be invested for the pull down. In that case a hold-back valve which does not impose too much pressure drop in the suction line or un loading of the compressor becomes the proper answer. With proper pilot connections to a large size hold-back valve, which is simply a pilot controlled piston valve, it is possible to use t.hia same valve as a back pressure regulator which controls the evaporating pressure after the system has pulled down to the operating level. It is also possible by put-* ing a solenoid valve in the pilot line to use the hold-back valve as a suction line stop valve. Operation at Varying Loads and Temperatures The means of controlling compressor capacity and evap orator capacity is about the same with multi-stage systems as with single-stage systems. This may be by compressor unloaders, hot gas bypasses, back pressure regulators, etc. One effect of low pressure on control systems is that the more significant increase in specific volumes of the refrigerant gas with respect to pressure drop results in a much greater capacity reduction when a decrease in load is accompanied by a decrease in evaporator pressure. Therefore a small pressure reduction may cause a very large percentage capacity reduc tion. These controls may be accompanied by hot gas by passes into the evaporator or by the use of balanced loaders. It is important to realize that when the capacity on one part of a cascade system is reduced by any means the load on the upper cascade will be also reduced and means must be made available to take care of tbe system control prob lems which this might create. MOISTURE IN THE REFRIGERANT CIRCUIT The subject of moisture in the refrigerant circuit and de scriptions of the testing and drying procedures for refrigerant systems are discussed in Chapters 59 and 61 of the 1961 \; ' j;] 'h i ! ll