Expander cycle

Fig. 1. One-expand cycle with external refrigeration for high ethane recovery in the hydrocarbon Hquid product.

The main advantage of a cascade over an expander cycle is presumed to be the lower power requirement. Since power is a large percentage of the total cost in large capacity plants, the possible... 

Becker also analyzed a similar cycle in which the expansion valve was replaced by a turbine and obtained an efficiency of 48.1%. Analysis of Becker s single-expander cycle gave an efficiency of about 40%. 

Published analyses of cascade cycles by means of energy balances under conditions comparable to those used in analyzing expander cycles are very scarce. Longwell and Kruse tabulated computer-calculated compressor and expander powers for a C3Hg-C2Fl4-CF[4 cascade. For a feed of 1,566 Ib-moles/hr at 515 psia and 60°F, 490 lb-moles of LNG and two gaseous products were produced. 

Since only 31% of the feed was converted to LNG, this cascade is not comparable with the expander cycles considered in which all the feed was converted to liquid. When the efficiency was calculated by Equation 3-1 it turned out to be only 2.15%. 

Figure 1-5. Flow diagram of the expander cycle for liquefying natural gas (1) pretreatment (mol.sieve), (2) heat exchanger, (3) turboexpander.

Natural gas expander cycle 2 Refrigerant expander cycle 3 Cascade cycle 4 Mixed refrigerant cycle... 

Natural gas expander cycles can be utilized at places with a high-pressure main gas pipeline and a low-pressure distribution line. The natural gas is expanded in a turbine, thereby cooling the exit gas. Dependent on the inlet conditions, the exit temperature will be so low that a part of the gas is condensed, typically around 12 to 30 %. The liquid is separated from the gas flow in a liquid-vapour separator as shown in Figure 11. 

Figure 21 Natural gas expander cycle with CO2 freeze out included.

The expander cycle and its corresponding temperature-entropy diagram are shown in Figure 4. Notice that the expander cycle is similar to the simplified Joule-Thompson cycle, except that the Joule-Thompson expansion valve has been replaced by an expansion turbine. The performance of this cycle differs in several ways. The expansion of the gas is no longer isenthalpic, but with the expansion turbine, it is isentropic that is, there is a change in... 

A prerequisite to the design of the gas liquefaction unit is to establish the cycle which is to be used. The choice of cycle is dependent on several factors including the capacity of the unit and weight and space limitations incurred by land-based or shipboard installation. For the purpose of this discussion, liquefaction cycles are classified into two categories the cascade cycle and the expander cycle. 

In the expander cycle (Fig. 5), the gas to be liqueiied is raised to some elevated pressure, say 1500 psi. The refrigerant in this cycle is the exhaust from an expansion engine. Methane refrigerant is compressed to about 1500 psi and cooled by ammonia or propane refrigeration. It is then expanded isentropically... 

A comparison of the cascade and expander cycle units when liquefying stripped natural gas received at 700 psig reveals the following ... 

Using the expander cycle permits the design of a simple plant with fewer items of equipment. A well-designed expander unit would require 830 installed brake horsepower with 100 bhp developed by the expander. About 80 of the sweet gas charged to the plant will be liquefied, with the remainder going to fuel. 

In an area where fuel is cheap it may appear offhand that a relatively inefficient cycle would be justified. However, the more inefficient the cycle, the more horsepower is required, which requires a correspondingly larger investment. Therefore, for a large-scale liquefaction unit the cheapest plant is the most efficient one. For small-scale liquefaction units, the expander cycle is actually the cheaper because of the lower capital charges. 

Fig. 5. General flow diagram for liquefaction of natural gas by means of expander cycle.

Comment by P. M. Schuftan, British Oxygen Engineering Co. I am surprised that there should be such a big difference in efficiencies between the cascade and expander cycles, especially since the raw product is essentially CH4 with few higher hydrocarbons. Perhaps this is due to the fact that it had been assumed for both cases that the natural gas is available at 700 psig. This may not be the optimum for the expander cycle. 

Answer by Author As a design basis, natural gas was assumed available at 700 psig for both cycles. It is true that this may not be optimum for the expander cycle. 

---