Expansion Valves


The purpose of the expansion valve is to control the flow of refrigerant from the high-pressure condensing side of the system into the low-pressure evaporator. In most cases, the pressure reduction is achieved through a variable flow orifice, either modulating or two-position. Expansion valves may be classified according to the method of control.


Direct expansion circuits must be designed and installed so that there is no risk of liquid refrigerant returning to the compressor. To ensure this state, heat exchange surface in the evaporator is used to heat the dry saturated gas so that it becomes superheated. The amount of superheat is usually of the order of 5 K.

Thermostatic expansion valves (TEVs) for such circuits embody a mechanism which will detect the superheat of the gas leaving the evaporator (Fig. 1). Refrigerant boils in the evaporator at Te and pe , until it is all vapour, point A and then superheats to a condition Ts , pe , at which it passes to the suction line, point B. A separate container of the same refrigerant at temperature Ts would have pressure ps , and the difference ps − pe represented by C–B in Fig. 1 is a signal directly related to the amount of superheat.

Figure 1 Superheat sensor on direct expansion circuit

The basic thermostatic expansion valve (Fig. 2) has a detector and power element, charged with the same refrigerant as in the circuit. The pressure ps generated in the phial by the superheated gas equalizes through the capillary tube to the top of the diaphragm. An adjustable spring provides the balance of ps − pe at the diaphragm, and the valve stem is attached at the center. Should the superheat fall for any reason, there will be a risk of liquid reaching the compressor. Ts will decrease with a corresponding drop in ps . The forces on the diaphragm are now out of balance and the spring will start to close the valve.

p1/p2 = T1/T2

Figure 2 Thermostatic expansion valve. (a) Circuit, (b) cross-section (Danfoss).

Conversely if the load on the evaporator increases, refrigerant will evaporate earlier and there will be more superheat at the phial position. Then ps will increase and open the valve wider to meet the new demand.

The phial must be larger in capacity than the rest of the power element or the charge within it may all pass into the valve capsule and tube, if these are colder. If this happened, the phial at Ts would contain only vapour and would not respond to a position Ts , ps on the T − p curve.

Use can be made of this latter effect. The power element can be limit charged so that all the refrigerant within it has vaporized by a predetermined temperature (commonly 0°C). Above this point, the pressure within it will cease of follow the boiling point curve but will follow the gas laws as shown in Fig. 3; and the valve will remain closed. This is done to limit the evaporator pressure when first starting a warm system, which might overload the drive motor. This is termed limit charging or maximum operating pressure. Such valves must be installed so that the phial is the coldest part.

Figure 3 Detector pressure for limit charged valve.

The slope of the T − p curve is not constant, so that a fixed spring pressure will result in greater superheat at a higher operating temperature range. To allow for this and provide a valve which can be used through a wide range of applications, the phial may be charged with a mixture of two or more volatile fluids to modify the characteristic curve.

Some manufacturers use the principle of the adsorption of a gas by a porous material such as silica gel or charcoal. Since the adsorbent is a solid and cannot migrate from the phial, these valves cannot suffer reversal of charge.


The simple thermostatic expansion valve relies on the pressure under the diaphragm being approximately the same as that at the coil outlet, and small coil pressure drops can be accommodated by adjustments to the spring setting.

Where an evaporator coil is divided into a number of parallel passes, a distribution device with a small pressure loss is used to ensure equal flow through each pass. Pressure drops of 1–2 bar are common. There will now be a much larger finite difference between the pressure under the diaphragm and that at the coil inlet. To correct this, the body of the valve is modified to accommodate a middle chamber and an equalizing connection which is taken to the coil outlet, close to the phial position. Most thermostatic expansion valves have provision for an external equalizer connection (see Fig. 4).

Figure 4 Thermostatic expansion valve with external equalizer

The thermostatic expansion valve is substantially an undamped proportional control and hunts continuously, although the amplitude of this swing can be limited by correct selection and installation, and if the valve always works within its design range of mass flow. Difficulties arise when compressors are run at reduced load and the refrigerant mass flow falls below the valve design range. It is helpful to keep the condensing pressure steady, although it does not have to be constant and can usually be allowed to fall in colder weather to save compressor power. Valves on small systems may be seen to fully close and fully open at times. Excessive hunting of the thermostatic expansion valve means that the evaporator surface has an irregular refrigerant feed with a resulting slight loss of heat transfer effectiveness. If the hunting is caused by a time lag between the change of valve position and the effect at the evaporator outlet, a solution can be to increase the mass of the sensor phial which will increase damping. Over-sized valves and incorrect phial position can also give rise to hunting. The phial should always he located on the horizontal outlet, as close to the evaporator as possible and not on the underside of the pipe.


The electronic expansion valve offers a finer degree of control and system protection. The benefits can be summarised as follows:

  1. Precise flow control over a wide range of capacities.
  2. Rapid response to load changes.
  3. Better control at low superheats so that less evaporator surface is required for superheat. More surface for evaporation results in higher evaporating temperature and better efficiency.
  4. Electrical connection between components offers greater flexibility in system layout, which is important for compact systems.
  5. The valve can close when the system shuts down, which eliminates the need for an additional shut off solenoid valve.

Types of electronic valve in use include a continuous flow type in which the orifice size is varied by a stepper motor, and a pulse width modulated (PWM) type. In each case a controller is used in conjunction with the valve. The controller is pre-configured for the refrigerant and valve type and it receives the information from sensors, for example, pressure and temperature at the evaporator outlet. This enables the superheat to be determined. The output signal to the valve initiates the orifice adjustment. In the case of the PWM valve it is the relationship between the opening and closing which determines the capacity of the valve. The valve is either open or closed and each time interval of a few seconds will include an opening period depending on the signal.

There is a third type of valve that combines both features. A modulating voltage is sent to the actuator, and as the voltage increases the pressure in the actuator’s container increases, resulting in an increased valve opening during an ‘on cycle’ of fixed duration.

In each case the control can be configured so that the valve remains closed in the event of power loss. Under partial load condition or floating condensing pressure, which happens at low ambient temperature, the condensing pressure decreases. Thermostatic expansion valves tend to hunt, but systems with electronic components operate at partial load in exactly the same and stable manner as at full load.

A continuous flow type valve is shown in Fig. 5. The valve seat and slider are made out of solid ceramic. The form of the valve slide provides for a highly linear capacity characteristic between 10 and 100%. Depending on the controller and its configuration, a single control valve can be used for different control tasks. Possible uses include: expansion valve for superheat control, suction pressure control for capacity control, liquid injection for de-superheating of compressor, condensing pressure control and hot gas bypass control to compensate excess compressor capacity and to ensure evaporating pressure does not go below a set point.

Figure 5 Electronic expansion valve. (a) Outside view, (b) sectional view, (c) sliding orifice (Emerson Climate Technologies).


The variable orifice of the expansion valve can be replaced, in small systems, by a long thin tube. This is a non-modulating device and has certain limitations, but will give reasonably effective control over a wide range of conditions if correctly selected and applied. Mass flow is a function of pressure difference and the degree of liquid subcooling on entry. The capillary tube is used almost exclusively in small air conditioning systems and is self-regulating within certain parameters. Increasing ambient temperature results in increasing load on the conditioned space and the condensing pressure will rise, forcing more refrigerant flow.

Tube bores of 0.8–2 mm with lengths of 1–4 m are common. The capillary tube is only fitted on factory-built and tested equipment, with exact refrigerant charges. It is not applicable to field-installed systems.

The restrictor expansion device overcomes some of the limitations of the capillary tube. The orifice can be precision-drilled whereas capillary tubes can suffer from variations in internal diameter over their length which results in changes to predicted performance. Fig. 6 shows how the device is applied in a reversible air conditioner. In Fig. 6a the device is shown in normal cooling mode. A bullet which is free to move horizontally by a small amount is pressed against a seat-forcing the refrigerant through the central restriction which acts as an expansion device. When the flow reverses, Fig. 6b, the bullet moves back to the other seat, but grooving allows flow around the outside as well as through it, so that the restriction is very small.

Figure 6 Restrictor expansion device.

It is normally fitted at the outlet of the condenser rather than at the evaporator inlet. This means that instead of a liquid line to the evaporator, the pipe contains liquid and flash gas and must be insulated. Although heat pick up is detrimental to performance, the pressure drop, which is used to drive the fluid, would normally have occurred in the expansion valve anyway. Liquid lines to remote evaporators on split systems can be quite lengthy and in a high-pressure liquid line of the type more usually used, the pressure drop can result in an increased condenser pressure and tendency to form bubbles. Also the restrictor can be delivered as part of the condensing unit and is removable, allowing changes to be made to give optimal performance.


Flooded evaporators require a constant liquid level, so that the tubes remain wetted. A simple float valve suffices, but must be located with the float outside the evaporator shell, since the surface of the boiling liquid is agitated and the constant movement would cause excessive wear in the mechanism. The float is therefore contained within a separate chamber, coupled with balance lines to the shell (see Fig. 7).

Figure 7 Low-pressure float valve on flooded cooler.

Such a valve is a metering device and may not provide positive shut-off when the compressor is stopped. Under these circumstances, refrigerant will continue to leak into the evaporator until pressures have equalized, and the liquid level might rise too close to the suction outlet. To provide this shutoff, a solenoid valve is needed in the liquid line.

Since the low-pressure float needs a solenoid valve for tight closure, this valve can be used as an on–off control in conjunction with a pre-set orifice and controlled by a float switch (Fig. 8).

Figure 8 Low-pressure float switch.

The commonest form of level detector is a metallic float carrying an iron core which rises and falls within a sealing sleeve. An induction coil surrounds the sleeve and is used to detect the position of the core. The resulting signal is amplified to switch the solenoid valve, and can be adjusted for level and sensitivity. A throttle valve is fitted to provide the pressure-reducing device.

Should a float control fail, the level in the shell may rise and liquid pass into the compressor suction. To warn of this, a second float switch is usually fitted at a higher level, to operate an alarm and cut-out.

Where a flooded coil is located in a liquid tank, the refrigerant level will be within the tank, making it difficult to position the level control. In such cases, a gas trap or siphon can be formed in the lower balance pipe to give an indirect level in the float chamber. Siphons or traps can also be arranged to contain a non-volatile fluid such as oil, so that the balance pipes remain free from frost.


On a single-evaporator flooded system, a float valve can be fitted which will pass any drained liquid from the condenser direct to the evaporator. The action is the same as that of a steam trap. The float chamber is at condenser pressure and the control is termed a high-pressure float (Fig. 9).

Figure 9 High-pressure float valve circuit.

The high-pressure float switch keeps the condenser drained without the need for a high-pressure receiver. The level in the evaporator is fixed by the system charge. Low charge systems using shell and plate heat exchangers and spray chillers are possible with this method. The type of float valve in Fig. 10 can work with ammonia or carbon dioxide refrigerants. Economizer circuits with the float switch expanding the liquid to an intermediate flash expansion vessel are used for low-temperature applications. This control cannot feed more than one evaporator, since it cannot detect the needs of either.

Figure 10 Sectioned view of Witt high-pressure float valve (Titan).

The difficulty of the critical charge can be overcome by allowing any surplus liquid refrigerant leaving the evaporator to spill over into a receiver or accumulator in the suction line, and boiling this off with the warm liquid leaving the condenser. In this system, the low-pressure receiver circuit, liquid is drained from the condenser through the high-pressure float, but the final step of pressure drop takes place in a secondary expansion valve after the warm liquid has passed through coils within the receiver. In this way, heat is available to boil off surplus liquid leaving the evaporator (see Fig. 11).

Figure 11 Low-pressure receiver circuit.

Two heat exchangers carry the warm liquid from the condenser within this vessel. The first coil is in the upper part of the receiver, and provides enough superheat to ensure that gas enters the compressor in a dry condition. The lower coil boils off surplus liquid, leaving the evaporator itself. With this method of refrigerant feed, the evaporator has a better internal wetted surface, with an improvement in heat transfer.


If a small heater element is placed at the required liquid level of a flooded evaporator, together with a heat-sensing element, then the latter will detect a greater temperature if liquid refrigerant is not present. This signal can be used to operate a solenoid valve.

The thermostatic or electronic expansion valve can also be used to maintain a liquid level. The phial and a heater element are both clamped to a bulb at the required liquid level. If liquid is not present, the heater warms the phial to a superheat condition and the valve opens to admit more liquid.

How useful was this post?

Click on a star to rate it!

Average rating / 5. Vote count:

No votes so far! Be the first to rate this post.

We are sorry that this post was not useful for you!

Let us improve this post!

Tell us how we can improve this post?