WSU Tree Fruit Research & Extension Center

Postharvest Information Network

Sunday, March 24, 2019

WSU-TFREC/Postharvest Information Network/Thermosyphon Oil Cooling Demonstration Project

Thermosyphon Oil Cooling Demonstration Project


The mission of Chelan County PUD Energy Services is to encourage energy efficiency through a variety of services. One service is an annual demonstration project to show the viability of a technology that is new or unfamiliar to local industry. Thermosyphon oil cooling for screw compressors is one such technology that is commonly used in the refrigeration industry, but is not common in this region. Although the technique is standard for the industry, as of January 1996, there were no thermosyphon systems in Chelan County.

In the last quarter of 1996, a thermosyphon system was installed on an existing screw compressor at Trout/Blue Chelan in Chelan, Washington. Savings verification lasted over a month and was completed in February 1997. This report details the technology, the project and measured savings.

Chelan County PUD Energy Services has a long history of promoting energy efficiency in the refrigeration industry. A majority of the efforts have focused on use of computer controls and strategies that can be implemented with these controls. Controls provide the first major step to comprehensive refrigeration plant efficiency.

Controls alone though are not the pinnacle of efficiency for this energy intensive industry. Numerous technologies are used by the industry, which increase the efficiency of refrigeration plants. Many of these technologies are not used in Chelan County due to the high cost of implementation and relatively inexpensive power. One such technology is thermosyphon oil cooling for rotary screw compressors. According to one supplier of new screw compressor packages, 70 to 80 percent of the systems sold in the Northwest, outside of the apple industry, are equipped with thermosyphon oil cooling. A smaller percentage use liquid injection and the remainder use other means, predominately water-to-oil heat exchangers.

Thermosyphon oil cooling is a passive means of cooling compressor oil using refrigerant condensate returning from the condenser. There are no power penalties associated with this method since there are no pumps and no loss of compressor capacity. Due to these facts, thermosyphon is a reliable, energy efficient means of oil cooling. Energy savings from thermosyphon are due to elimination of liquid injection and the ability to lower the pressure differential between suction and discharge. Both of these issues are discussed in the context of this report.

The demonstration project approach is simple. The project is a three-way partnership between the PUD, facility and contractor. The cost of the project is split three ways with the PUD paying a majority of the cost, the facility paying a smaller amount and the contractor providing some goods or services at no cost. In May 1996, selected refrigeration contractors from the Northwest were approached to determine interest, expertise and availability. From this process, two contractors were selected to provide bids for a site yet to be determined.

In June, a letter was sent to all industrial customers with refrigeration facilities to solicit potential participants in the project. Only one customer responded to this request. Although response was minimal, the District felt that this project was worth pursuing and in September received proposals from the two contractors. The bids were reviewed by the facility's refrigeration personnel, local trade allies and Energy Services staff. The work was awarded to PermaCold Engineering of Portland, Oregon.

Work began in October and was completed in November. Energy savings verification started on December 16, 1996 and continued through February 10, 1997. The remainder of this report will describe in detail screw compressor oil cooling technologies, thermosyphon system components and operation, typical system cost and potential energy savings.

Screw Compressor Oil Cooling

The process of compressing a refrigerant gas produces large quantities of heat which must be removed from the system. Some of the heat is transferred to the refrigerant and rejected to the atmosphere through the condenser. Some of the heat is simply transferred into the space. The rest of the heat is transferred to the oil and must be removed by external means. The three primary types of oil cooling systems for screw compressors are:

  • Liquid injection
  • Indirect with water or glycol
  • Indirect with refrigerant condensate (thermosyphon)

Liquid Injection

In the Northwest apple industry, liquid injection is the preferred oil cooling method. The main reasons stated for using liquid injection are low power cost and limited annual hours of screw compressor operation. The most common liquid injection oil cooling method is the direct injection of liquid refrigerant into the compression process. The liquid refrigerant cools the compressor and therefore the oil by absorbing the heat through a phase change as it does in an evaporator. The refrigerant vapor is then compressed with refrigerant returning from the evaporators and is condensed in the condenser. This refrigerant performs no process cooling, but requires compressor energy to perform the function of oil cooling. Aside from the lowest installation cost, the only benefit that liquid injection provides is a low discharge temperature. A basic liquid injection system is shown in Figure 1.

There are other techniques that use liquid injection to cool the oil. One such technique is to use a refrigerant pump to provide the liquid pressure. Another is to inject liquid into the discharge line. Direct liquid injection into the compressor using system pressure, is the only method considered for comparison in this report since it is the predominate method used in the region. The advantages of liquid injection are:

  • Low initial cost
  • Low discharge temperatures
  • Minimal maintenance

The disadvantages have a dramatic impact on the operation of the compressor. The disadvantages include:

  • Reduction in compressor capacity
  • Higher operational cost

The capacity of the screw compressor is decreased due to the increased volume of gas that must be compressed. There is no argument in the industry that compressors using liquid injection cost more to operate than compressors using thermosyphon oil cooling. The additional cost is due to the requirement to compress the vapor used to cool the oil. Compression requires energy, which comes at a cost. Both of these issues are supported by the data listed in Table 1.

Figure 1

Table 1

Indirect Cooling with Water or Glycol

This cooling method utilizes a shell and tube heat exchanger and a source of cool liquid from an external cooling tower or closed loop evaporative cooler. Once-through water can be used but results in high water usage. An indirect cooling system uses a pump to circulate the cooling medium and a cooling tower or evaporative cooler to reject heat from the cooling medium.

The advantages of water or glycol systems are:

  • No reduction in compressor capacity
  • Lower compressor operating cost than with liquid injection

The capacity of the compressor is not reduced in this method since the oil is cooled outside of the compressor. Since there are no compressor inefficiencies, operational cost are lower, although there are operational costs associated with the circulating pump and cooling tower fans. The potential savings from reduced condensing pressure will offset these small additional costs.

The disadvantages are:

  • Additional cooling equiplment
  • A circulating pump is required
  • Increased maintenance of cooling system components
  • Potential tube fouling

As discussed above, this system uses additional cooling equipment, namely a heat exchanger, circulating pump and a cooling tower. Some of these components use energy to operate. There is increased maintenance with this system due to these additional components. If a cooling tower is used, there is a chance for fouling of the water side of the heat exchanger.

Thermosyphon Oil Cooling

Thermosyphon oil cooling is an indirect cooling method which uses refrigerant condensate as the cooling medium. The heat exchange process takes place in an external heat exchanger, using passive circulation of refrigerant from a high pressure receiver. The high pressure receiver is physically located above the heat exchanger and liquid refrigerant is fed to the heat exchanger by gravity. In the heat exchanger, the refrigerant is "boiled off" as it absorbs heat from the hot oil. The vapor and some liquid are returned to the high pressure receiver. The hotter liquid returns since it is less dense than the supply refrigerant. The gas is vented to the inlet of the condenser where the oil heat load is rejected. A basic thermosyphon system is shown in Figure 2.

During periods of low condensing temperatures below 75 ° F, a temperature regulating valve (Amot Valve) is used to maintain proper oil temperature. If the oil temperature goes too low, inadequate bearing lubrication could result and refrigerant and oil could mix. Adequate oil temperatures insure no refrigerant remains entrained in the oil. The advantages of thermosyphon are:

  • Reduction in compressor power and increase in capacity as compared to liquid injection systems
  • Reduction in energy requirements compared to water cooled systems incorporating a circulation pump and cooling tower
  • Reduction in water cost compared to water cooled systems using once-through water
  • Reduced maintenance cost compared to water cooled systems since tubes do not foul

Compressor efficiency is increased since all the work of the compressor goes to the cooling process. No capacity is utilized for oil cooling. Overall energy usage is less than other indirect systems since there are no energy consuming components in the system. There are no water costs associated with the system so additional operational savings are realized. The heat exchanger is not subject to fouling so maintenance costs are minimal.

There are some disadvantages of thermosyphon systems. The most notable are:

  • Higher initial cost over liquid injection due to additional vessels and piping
  • Higher initial cost over water cooled systems since oil cooler and piping must be designed for 300 PSI
  • Additional design cost over the other systems since proper design is vital to proper operation
Thermosyphon systems contain more vessels and components than liquid injection systems, making the first cost more expensive. Heat exchangers and vessels must be rated for 300 PSI, increasing the cost of the components. Piping and vessel layout must be carefully designed to ensure proper operation of the system. These few disadvantages are easily overcome if design guidelines are followed and installation is properly performed.

Figure 2

Compression Efficiency

With a thermosyphon system, no refrigerant is injected into the compressor. This allows the compressor to perform more work by increasing the flow of gas performing process cooling. Table 1 shows compressor operating parameters with and without liquid injection for the listed conditions.

Discharge Temperature

Discharge temperatures are higher with thermosyphon since no cooling takes place until the condenser. There are no downsides to higher discharge temperatures. Discharge temperatures with and without thermosyphon are shown below:

  • Discharge temperatures with thermosyphon, 140 °F to 180 °F
  • Discharge temperature with liquid injection, 120 F to 130 F

Oil Cooling Method: Impact on Compressor Life

One of the long term goals of the demonstration project is to determine impacts of oil cooling methods on compressor component life. This topic is debated in the industry with no definitive answers as to whether one oil cooling method is better for a compressor over another. At issue is whether or not the injection of raw liquid into the compressor reduces the life of the compressor or its components.

There is a "gut feeling" of some industry professionals that liquid injection will lead to reducing the life of bearings and rotors. One manufacturer has even gone as far as stating increased compressor life from elimination of injection of raw liquid. Some of the rationale to support increased life includes the possibility of gas bubbles in the oil affecting bearing life and rotor tip wear at the point of injection. Neither of these problems are supported by studies or by manufacturer's literature. Discussions with compressor repair professionals also found no trends of compressor problems directly related to liquid injection.

Based on the lack of information available and the experience of the repair professionals, no statement can be made on the detrimental effect of liquid injection. This is not to say though that elimination of liquid injection will not extend the life of the compressor due to another benefit.

The benefit is the reduction in the work the compressor must perform based on a reduction in condensing pressure. When condensing pressure is reduced, the efficiency of the compressor increases. Therefore, the compressor can perform more work for the same horsepower. Simply stated, this increase in efficiency translates to less strain on the compressor. Less strain should result in increased compressor life.

It is difficult to use this assumption to quantify the economic benefit of thermosyphon oil cooling. The benefit can be justified by energy savings alone, which can be quantified and is discussed in the next section.

Energy Implications

The main emphasis of the demonstration project is to show the energy savings of a thermosyphon system. There is little argument in the industry regarding the impact of liquid injection. Manufacturers will provide information on the bhp/ton efficiency for compressors with liquid injection and with external oil cooling. The efficiency gain depends on the operating conditions so most general information is just that, very general (taken from a Vilter publication). Table 2 (below) shows BHP/TON penalties due to liquid injection. Operating parameters are not stated.

Pressure ratio implications
The BHP/TON penalty varies based on the pressure ratio. High pressure ratio compressors have a greater penalty, whereas lower pressure ratio compressors have lower penalties. The higher penalties are due to the decrease of the compression efficiency at higher pressure ratios. In addition to the decrease in compression efficiency is the increase in liquid refrigerant flow required to provide oil cooling. The actual penalty is based on the compressor and the operating parameters of the system.

Testing for minimal energy savings
A simple test can be performed to determine the liquid injection penalty associated with the actual compressor system. The test involves measuring the motor amperage with and without liquid injection. With the compressor operating fully loaded, place a clamp-on amp meter on a motor lead and record the amps in a steady state condition. Next, close the valve feeding liquid refrigerant to the compressor. Again, record the amp reading. Be sure to open the valve as soon as the second amperage reading is taken to prevent over-heating the compressor. Calculate the percent reduction. This is a rough estimate of the minimum full load energy that can be saved simply by eliminating liquid injection.

The "real" energy savings
Eliminating liquid injection is only a small part of the compressor energy that can be saved. The majority of savings is from the ability to lower condensing pressure. With liquid injection, the condensing pressure must be high enough to maintain an adequate pressure differential across the thermal expansion valve feeding liquid into the compressor. The industry standard is to maintain a minimum 100 PSI differential between suction and discharge. The actual discharge is often higher for a variety of reasons.

Efforts have been made to convince operators to raise the suction pressure of their systems to increase the compressor capacity and to reduce fruit weight loss and defrost requirements. Though, based on the use of a 100 PSI differential, raising the suction pressure results in an increase in the discharge pressure, a far worse situation than operating with lower suction.

Using a thermosyphon system eliminates the problem since discharge pressure no longer has a bearing on cooling the compressor. This is where the real savings come from. Table 3 shows the information from a compressor manufacturer for two conditions. Column A shows the information for a system operating with liquid injection and low suction and high discharge. Column B shows the same compressor with external oil cooling, higher suction and lower discharge pressures. Column C shows the increase in efficiency for this compressor. Values are based on full load conditions.

The ability to lower the condensing pressure is dependent on adequate condenser capacity. In general, fruit industry facilities have ample condenser capacity to produce low condensing pressures.

Table 2

Table 3

Trout Case Study

To prove the viability and the savings of a thermosyphon system in a real application, a demonstration was needed. Trout-Blue Chelan agreed to be the site for the demonstration. Trout's site fit the criteria for the demonstration. The compressor had to be a low-vi machine over 150 horsepower and currently using liquid injection. In addition, the compressor had to be a base load machine on-line at least 9 months a year. Based on early estimates of cost and savings, this is the minimum application that would result in adequate savings to justify the facility's portion of the cost of the demonstration project, based on a five year payback. Of equal importance, but rejected as a requirement, were the current operating pressures. The system was installed in November 1996. Verification test ran from December 16, 1996, through February 10, 1997.

Table 4

Kilowatt Reduction

The kilowatt requirement of the compressor is drastically different for the two sets of parameters. Table 5 provides information on the average hourly peak demand for each test period.

Energy usage depends on the operating pressures, and to a slight degree, the outside air temperature. Table 6 lists the average daily values for each of these parameters for both test periods.

The refrigeration plant used in this study provides cooling for controlled atmosphere rooms. The cooling load in these rooms is not greatly affected by the range of outside air temperatures experienced during the test periods. Due to this fact, no corrections to compressor energy use were necessary due to differences in outside air temperatures.

Table 5

Table 6

Annual Energy Savings

Annual energy savings is determined by extrapolating the measured values out over the course of the year. Since the operating parameters used during the test period are not used all year, adjustments are made to the measured data.

The computer modeling tool used to estimate energy usage has been proven to provide reliable results when compared to actual systems. Due to this reliability, the model is used to predict annual energy usage and therefore savings. The model was levelized to match actual measured data. The measured data includes suction and discharge temperatures and compressor horsepower. Cooling load is predicted based on these values using manufacturer's performance data. Once the model is levelized, the only parameter that needs to be changed is the cooling load. This may not be true for all cases, but was a reasonable assumption for this site. Predicted annual energy based on measured data is summarized in Table 7. Project costs for the installation only are shown in Table 8.

Table 7

Table 8

Minimum Specifications for Thermosyphon Systems

The following is a list of the minimum requirements, recommendations and considerations for ammonia screw compressor applications in which the thermosyphon method of oil cooling is utilized:

  • Thermosyphon receiver should be mounted high enough above the oil cooler vessel or vessels so that the weight of the liquid refrigerant column (liquid head) is great enough to overcome the pressure drop through the liquid supply piping, valves and oil cooler itself. Most manufacturers in the industry now suggest that this be a minimum of six feet.
  • Pipe sizing is critical to proper operation to achieve a true thermosyphon effect. Experience has shown that lines can be undersized or oversized. A good engineering review of the pipe sizing is strongly suggested.
  • Thermosyphon receiver should be sized based upon oil heat rejection at maximum load conditions, including all compressors connected to receiver, and assumption of a 4:1 overfeed recirculation rate (liquid supply to each oil cooler of four times the amount of ammonia evaporated to remove the oil heat load). Receiver should be designed such that a minimum liquid reserve contained in the vessel at all times is equal to a 5-minute supply of the quantity that is evaporated at maximum load. Receiver shall also be designed with proper cross-sectional area to allow liquid-gas separation of wet gas returning to vessel and with proper venting to condensers.
  • The thermosyphon vessels, including receiver and oil cooler heat exchangers, shall be ASME certified vessels properly designed and coded for the high side pressure involved (250 psig standard minimum for ammonia high side). Safety valves are required in full accordance with code requirements, including shell side of these vessels. Pressure safety relief provision shall also be provided on the tube side of oil coolers which are provided with inlet and outlet valves. Relief valve factory settings shall be no greater than the ASME certified pressure rating of the vessel which they protect.
  • Generally, use angle type service isolation valves in lieu of straight-through type valves for thermosyphon pipelines because of their typically lower pressure drop rating.
  • Service isolation valves are recommended for all such applications, but particularly for systems with multiple compressors. Again, use angle valves where possible.
  • Drain valves are strongly recommended on the oil cooler head, receiver and other areas where oil may be trapped during operation of the system.
  • An automatic temperature regulating valve should be used to maintain proper set-point oil temperature. At this time, the industry has standardized on a three-way valve that modulates the flow of oil through or around the oil cooler heat exchanger. The particular valve used by most manufacturers is manufactured by AMOT and is often referred to as the "AMOT valve".
  • In the case of evaporative condensers which are found in most ammonia systems, the liquid draining piping from the condenser to the thermosyphon receiver must be installed in full accordance with condenser manufacturer's liquid drain standards, so as to avoid the possibility of inadequate draining from the condenser. In most cases, proper piping of the condenser or condensers in a system requires trapping of the individual liquid drains from each condenser coil.
  • If converting an existing liquid injection screw compressor package to thermosyphon oil cooling, the control arrangement will need to be checked. It is necessary to verify set-point ranges, control logic and capability with the original manufacturer to insure compatibility with thermosyphon operation. In some instances it has been found that a new electronic control board or EPROM is required because of the difference in compressor operation with thermosyphon operation. At time of start-up, all compressor operational parameters, including discharge temperature and oil temperature, should be checked to verify that they are within the stated tolerance ranges of the compressor manufacturer for thermosyphon operation. (Specifications provided by Ward Ristau of PermaCold Engineering.)

    Recommendations for Other Systems

    Thermosyphon oil cooling is not for all screw compressor applications. The applicability depends on the size of the compressor, the annual hours of operation and the operating parameters. In the Northwest fruit industry, this equates to 200 horsepower or greater, base-loaded compressors. The compressors should operate a minimum of 9 to 11 months of the year.

    To realize the savings discussed earlier, the refrigeration engineer must be able and willing to operate with a discharge pressure below 120 psig. If the discharge pressure cannot be lowered, the full savings will not be realized. There are not very many good reasons to operate with discharge pressures higher than 120 psig once liquid injection is eliminated.

    The additional cost of thermosyphon systems does not vary greatly with compressor horsepower. The cost to retrofit systems is between $13,000 and $15,000. If new compressor systems are specified with thermosyphon, the cost is reduced to between $10,00 and $12,000.

    For the base compressor system noted above, the payback for a new system will be just over three years, and for a retrofit system, the payback will be just over four years, using Chelan County PUD's industrial rate of $0.01115/kWhr. Higher electric rates will have shorter payback periods.


    The purpose of the demonstration project is to show the viability of a technology that is new or unfamiliar to local industry. It may also show that the technology is not viable. In this demonstration project, thermosyphon oil cooling was shown to be a technology that performs as expected.

    The main purpose is to prove that the technology works. This was done. The secondary purpose is to show a level of energy savings that supports the installation of the system. The savings for this system were shown. Not all facilities will experience the level of savings measured at this site, but the savings potential exists at all similar facilities. This system now operates with a discharge pressure of approximately 120 psig. This is not meant to be an indicator of a low limit, but it is as low as the operator is willing to go at this time. Since discharge pressure is not critical to compressor cooling, discharge pressure can be lowered even further.

    The savings were verified and the operation of the thermosyphon oil cooling system have both been proven. This is a viable technology that should be considered for any new installation meeting the earlier noted minimum criteria and reviewed for use in existing systems. There are no downsides to a properly designed and installed thermosyphon oil cooling system.

    The refrigeration industry outside of the apple industry is using thermosyphon on a large majority of systems. The apple industry has not done so due to the nature of the annual cooling loads and the misconception that savings from eliminating liquid injection are minimal. Warehouses are consolidating and operating large screw compressors greater hours than ever before. It is vital that the real savings related to lowering discharge pressure be considered when deciding on which oil cooling system to use.

Shaun M. Seaman

Chelan County PUD Energy Services
P.O. Box 1231, Wenatchee, WA 98807-1231

13th Annual Postharvest Conference
March 1997

Tree Fruit Research & Extension Center, 1100 N Western Ave, Washington State University, Wenatchee WA 98801, 509-663-8181, Contact Us