PRODUCT DESCRIPTION:



[pic]

FLUID PROPERTY TEST APPARATUS FOR HIGH HEAT FLUX COOLING SYSTEMS

Prepared For:

Isothermal Systems Research

By: Team ISoCool

Tony Crew

Adam Mattison

Carl Niggemyer

Ryan Watson

Lead Instructor: Steven Beyerlein

Technical Advisor: Steve Penoncello

Student Mentor: Mike Harper

May 5, 2006

Executive Summary

This report details the development of a test apparatus for high heat flux cooling systems. It is desirable to compare boiling curves for different fluids in Spraycooling applications to determine acceptable alternatives to the current fluid. The test apparatus takes data needed for the generation of a boiling curve for the fluid being tested. It features a spray chamber with a heat block, atomizer, condensing coil, fluid reservoir and pump with a bypass. The heat block controls the heat flux to the spray surface, in addition to measuring the surface temperature. A pressure transducer is located in the chamber to measure the saturation pressure. From that, the saturation temperature can be calculated. These three values (surface heat flux, surface and saturation temperature) are needed to generate the boiling curve. The uncertainty of surface flux measurement is +/-17.5 W/cm2 and the surface temperature has an uncertainty of +/- 3.137 K. The current version of the test apparatus experiences electrical interference that affects the collected data. Methods to improve this interference include changing the power supply to the heaters and sealing the heat block ceramic insulation. Other modifications for full fluid testing include non-reactive plumbing and fluid evacuation development. The material cost of the apparatus is $1715.

1. Project Background

Spraycool technology consists of a closed thermodynamic system that uses a fluid sprayed through an atomizer onto a heated surface, in particular computer chips. Once the fluid hits the surface, it vaporizes and the chamber fills with vapor. This vapor is taken out of the chamber and the heat is removed through a phase change using a heat exchanger. The fluid is then pumped back through the atomizer after being filtered. This spray causes the temperature of the computer chip to stay constant, thus increasing the life and operating capability of the chip. The system, shown in figure 1, operates in an environmental temperature range of -65C to 125C with a supply pressure of 15-25 psi with flow rates of 1-2 l/min. Isothermal Systems Research (ISR) currently uses a particular fluid, (PF-5060) in several Spraycool products. In the near future however, this fluid may be discontinued or its price will increase dramatically. Due to this possibility, it is in the company’s best interest to find alternative fluids to use in Spraycool applications. To accomplish this, a test apparatus must be designed that simulates this existing system to test alternative fluids and determine the heat transfer properties that are desired for their application in ISR products.

[pic]

Figure 1: Existing Spraycool system

To compare the heat flux capacities of different fluids, a boiling curve is generated for the test fluid. The boiling curve is the result of Newton’s Law of Cooling.

q’’ = h* (Tsurface – Tsat) [Eq. 1]

[pic]

Figure 2: General Boiling Curve

‘h’ is the heat transfer coefficient and is the slope of the boiling curve. Point 3 represents the critical heat flux for the fluid. This is the maximum value of heat that a fluid can transfer. Beyond this point, any increase in the flux results in a spike in surface temperature. For Spraycool applications, however, point 2, where the heat transfer coefficient is a maximum, is considered the optimal heat flux, since one does not wish to operate the system so close to the actual critical value. Once the boiling curve has been generated for the fluid, it can be compared to the curves of other test fluids. The fluid with the higher critical heat flux value indicates a fluid that may work better in a Spraycool product since it can transfer more heat.

2. Problem Definition

The test apparatus requires many qualities to accomplish the goal of testing any possible fluid. They include:

Manufacturing:

▪ Overall apparatus desktop size

▪ System includes a filter for the test fluid

▪ Non corrosive piping

▪ Non reactive component materials

System Operating Needs:

▪ Easy to change fluid

▪ Able to accommodate any kind of fluid, including dielectric, non-dielectric fluids, and reactive.

▪ Safety switch to turn system off before the critical heat flux is reached

▪ Adjustable distance between atomizer and heated surface

▪ Components able to operate in a wide range of operating conditions

▪ Easy to change atomizer

▪ Heat exchanger able to accommodate up to 500 watts/cm2

▪ Operation repeatable

▪ System data acquisition compatible with ISR/LabView 7

Data Acquisition (controls and measures):

▪ Chamber pressure

▪ Saturation temperature (calculated from chamber pressure)

▪ Controlled surface temperature

▪ Reservoir temperature

▪ Spray temperature

▪ Pressure across atomizer

▪ Surface heat flux output up to 150 watts/cm2

Data Reduction:

▪ Produce a boiling curve for the test fluid

▪ Determine heat transfer coefficient value

▪ Determine critical heat flux value

3. CONCEPTS CONSIDERED

To develop the overall design, it was essential that the system be broken down into parts. The main sub components focused on were the heating source, the ability to change the distance between the atomizer and the heated surface, plumbing system, heat rejection from the working fluid in the system and data reduction.

3.1 Heating Element

To generate the heat flux value of 150 W/cm2 requested by the client, we had two options, either buy an over the counter heating element or create our own. At first, the Peltier chip seemed like a viable option for the system because it offered several features that were needed in our design. Control over the heat flux of the chip and embedded thermocouples for temperature measurement were two features that excited the team about the Peltier chip. Further research on the chips showed that they were not capable of producing the required heat flux.

[pic]

Figure 3: Peltier Chip

The second option developed was manufacturing a heating element out of copper. The concept consisted of a copper block with a 1 cm2 top surface. From that surface there is a neck with the same dimensions attached to a base which is significantly wider. Heaters are installed in the base to produce sufficient heat flux. In the neck there are three holes drilled to the center where thermocouples are placed to measure the material temperature. Then the entire copper block component is well insulated (except the top surface which will be spray cooled) to ensure one-dimensional conduction though the copper block. This allows the use of Fourier’s Law for one-dimensional conduction.

q’’= - k*A*[ΔT/Δx] [Eq. 2]

This design has the advantages of the Peltier chip (control over the flux and imbedded thermocouples) in addition to the magnitude of flux generated being closer to the system’s needs. By varying the voltage into the heaters, we can change the heat flux and temperature of the spray surface.

3.2 Atomizer

Another major problem facing the design was developing a way to change the distance between the atomizer and heated surface. Initial design ideas to accomplish this task focused on moving the atomizer. Since there is a flow of fluid though the atomizer, the challenge is designing a way to move the location of the atomizer without having to re-plumb the whole system. This was further complicated by the fact that all plumbing needed to be corrosion resistant, and therefore needed to be stainless steel. The simplest way to accomplish the change in atomizer distance was to have multiple lengths of tubing. Having sized tubing for specified distances from the heating source has two major draw backs. First, the system would need to be drained in order to change atomizer location. Second, you would have to cut a tube for every possible test distance. Another design idea was to have flexible stainless steel tubing, like what is currently being used in shower heads and sinks. The major concern with this design is that the crevices could collect particle deposits and lead to corrosion. The deposits could also lead to fluid evacuation concerns. Other design ideas included telescoping piping and the use of angular displacements.

3.3 Plumbing

The next major design task is developing the plumbing system. The pump needs to provide a variable flow of approximately 200 ml/min and be corrosion resistant. Research showed that the best pump on the market for our application is one of the pumps in the GJ series produced by Micro Pump. The other major aspect of the pluming system is the pipe itself. For our application the pipes must be corrosion resistant, but don’t have to be able to handle any high stress or strain situations. Stainless steel piping was the main design idea; however, various copper and brass alloys with corrosion resistant coatings were researched.

3.4 Heat Rejection

Finally, the method of removing heat from the working fluid was researched. After being sprayed on the surface of the heating source, the fluid vaporizes and becomes either two-phase or all vapor. Since neither vapor nor two-phase flow can be run though the pump a way was needed to condense the fluid back to liquid, preferably sub-cooled. The first design called for the use of a Lytron air to liquid heat exchanger. The Lytron heat exchanger allows the user to control the amount of heat being removed from the working fluid by controlling the air flow over the heat exchanger with a Tarzan fan and features stainless steel tubing for the working fluid. This heat exchanger would be placed after the spray chamber and before the reservoir of working fluid. By controlling the fan, this system would have a direct control over the temperature in the reservoir. However, due to the location of the heat exchanger there would be very little user control over the pressure inside the spray chamber.

The next design used a shell in tube heat exchanger, which would be located in the same place as the air to liquid heat exchanger. The shell in tube design gives the user more control over the exiting temperature of the working fluid. A drawback to this concept is that it would require an addition of another cooling fluid with a pump.

Finally, a design to include condenser coils inside the spray chamber was considered. This system allows for direct control over chamber pressure. Since the condenser would require cold water to be pumped though the inside of the coils two designs were developed that would allow for the cold water circulation. The first is to simply run water from a tap or hose through the coil and exit into an existing water collecting facility (like a sink). Option two is to create a cooling system consisting of a variable flow rate pump, an air to water heat exchanger (similar to the Lytron models), and a cold water reservoir to serve as thermal inertia for the system. The cooling system would allow for control over the temperature of the water as it enters the condenser.

4. SYSTEM DESIGN

Our selected system concept simulates the working product that ISR uses for their applications. By choosing this setup, it gathers data at the closest conditions to their intended use. The copper heat block was chosen because of its ability to generate the necessary amount of flux to the surface in addition to its data reduction capabilities through the embedded thermocouples. An adjustable ring stand supporting the heat block was decided on since then the atomizer won’t have to be moved and it is much simpler to move the block since the atomizer will have a constant position within the spray chamber. For the initial system prototype plastic tubing was used for the plumbing system since water was the only fluid being used. This is to benchmark the prototype’s performance against known properties of water before testing more expensive fluids. The interior condensing coil was chosen because of its ability to control the pressure in the chamber. It was also favorable since it eliminated any two-phase flow into the pump.

[pic]

Figure 5: system design

5. PRODUCT DESCRIPTION

The test apparatus features a spray chamber with external plumbing associated with a pump, reservoir and bypass filter system.

5.1 Spray Chamber

The chamber body is made out of a 10-in diameter, 1/8-in. thick acrylic cylinder that is 15 in tall. The atomizer, condensing coil, heat block with spray surface and all data reduction sensors are contained within the spray chamber. Holes are drilled into the

Figure 6: Spray Chamber Design

chamber wall for the cooling water inlet and outlet, the pump reservoir inlet as well as the conduit for the thermocouple and plug heater wires. The chamber cover and base are made from a PVB plastic .5 in. thick 10.5 in. diameter. A lip is created on the cover by securing a second circle of 9.75 in. diameter to the larger diameter plastic piece. A seal is created by using a Vitron o-ring material around the smaller diameter lip. The atomizer housing is threaded into the center of the cover. There is also a threaded hole for the pressure transducer.

[pic]

Figure 7: Spray Chamber without cooling coil

5.2 Heat Block

The heat block is how the heat flux delivered to the spray surface is controlled. The heat block consists of a copper block machined to the shape shown in Fig. 6. Watlow cartridge plug heaters model #E1A53 are installed in the lower portion of the copper block. These plugs radiate heat in a radial direction with the wattage controlled using a variable voltage source. In the top, smaller diameter portion three sheathed, ungrounded type T thermocouples are inserted to the centerline of the copper 0.25 in. apart from the surface. From these temperature readings and the known distance between them, the amount of flux can be calculated in the copper between those points using Fourier’s Law for One Dimensional Conduction. This equation is valid because of

[pic]

Figure 8: Copper Heat Block

the insulation around the block and the flux loss factored in to the calculation. The insulation consists of a 4-inch diameter cast made using Rescor 750 Silica ceramic with the casting dye made out of aluminum flashing. The cover to the block has a 1 cm2 area hole cut in the center of it to expose the top of the copper block. This exposed copper is the spray surface. The insulation provides for almost purely 1-D conduction, but there are some heat losses from the block so pure 1-D is not achieved. Having multiple heat flux calculations from different sets of thermocouples in the block solves this problem since a real time heat flux loss to the surface can be calculated. This heat flux with a percentage loss is then extrapolated to calculate what the surface temperature is and is then inputted to the boiling curve.

5.3 Atomizer and Positioning

The heat block is placed on a .1” thick plastic ring with a 3” inside diameter and 6” outside diameter that is supported by three rods that are secured to the chamber base. The ring is secured to the rods with setscrews. These then allow the tester to adjust the height and angle of the heat block relative to the atomizer for the desired spray position.

[pic]

Figure 9: Atomizer positioning

The atomizer is press fit into a ½” diameter, ½” long housing and a female ½” – 20 threads. Inside there is an 1/8” milled hole down 0.47” and through hole reamed to 0.1”. This housing can then be screwed into a threaded cylinder which is threaded in the center of the chamber cover. This should cause the atomizer to spray exactly from the center of the chamber onto the heat block. Since the atomizer itself cannot be adjusted, the heat block should be moved to the desired position to get the best spray.

5.4 Condensing Coil

To condense the vapor in the chamber a 20’ long 3/8” O.D. ¼” I.D. copper tube is coiled and then placed around the outside edge inside the chamber. It is coiled to a 9.5” diameter with 10 coils. This coil is then connected to elbow joints that go through the chamber wall. These elbow joints are connected using two joints, one outside, and one inside, with o-ring material on both sides to provide the seal. The condensing coil is then secured to these elbows using hose clamps. Cold water is then pumped through the coil. Tap water will be acceptable for the system’s cooling needs. Connect the inlet to the tap and then the outlet can be dumped back into the same sink using appropriate tubing.

5.5 Sensors

The chamber features many sensors for boiling curve creation as well as system performance data. There are the thermocouples in the heat block to calculate the amount of flux at the surface as well as the surface temperature. There is an Omegadyne pressure transducer model #PX35D0-050AV extending from the chamber cover to measure the chamber pressure, which will also be the saturation pressure, and from that the correlated saturation temperature is calculated using fluid property models. These three things (flux, surface and saturation temperatures) generate the boiling curve.

5.6 Plumbing

There is a hole near the bottom of the chamber body that provides the refill fluid to the pump reservoir. The condensed fluid will accumulate in the bottom of the chamber body, which is sloped. At the end of the slope will be this drainage hole. Gravity will provide the force needed to move the condensed fluid out of the body and into the pump reservoir. See figure 7. The bottom of the reservoir will be the pump inlet hole, providing fluid to the pump. Then the fluid goes through an ISR developed filter and into a bypass. Enough working fluid necessary to the system is pumped through to the atomizer while any extra fluid is bypassed back into the pump reservoir using a valve.

Figure 10: Plumbing System Diagram

Just before the atomizer housing on top of the chamber cover is a bleed valve to remove excess air in the system before testing, ensuring that only fluid is going to the atomizer.

5.7 Data Reduction

The main purpose of the LabView program is to take real time data from the spray chamber and generate the boiling curve for the fluid being tested as well as to monitor key aspects of the system performance. For this section, please refer to the LabView block diagram and front panel in Appendix B. The boiling curve is generated by graphing the amount of superheat (Tsurface-Tsaturation) vs. heat flux (q’’). The heat flux is calculated using Fourier’s Law for One-Dimensional Conduction. We use the series of thermocouples in the copper block to measure the ΔT for that length of copper, which is constant for all flux calculations (0.25 in.). Our spray surface has an area of 1cm2. For copper the value of k is 385 W/K*m. The program measures the ΔT for the lower and middle thermocouple positions and calculates the flux for that section. It then calculates the flux in the next section, between the middle and top thermocouple positions. To define the flux in the top to surface section the amount of heat loss between the bottom to middle and the middle to top sections is assumed to be constant. This percentage loss, once factored into the middle to top section, is defined to be the flux at the top to surface section. Then, since the top temperature and the flux are known, the temperature at the surface can be extrapolated.

[pic]

Figure 11: Labview Logic Diagram

Tsur=(q’’/k)*dx+Ttop [Eq. 3]

The other component to the boiling curve is the saturation temperature. Since the chamber temperature is not necessarily the saturation temperature, we have to use the saturation pressure to find the saturation temperature. The pressure transducer placed in the chamber cover measures the pressure in the chamber, and this will be the saturation pressure. From corollaries for saturation conditions for the fluid being tested, the saturation temperature can be calculated. Currently the LabView program in Appendix B features a Sub-VI that contains these corollaries for pure water. Testers will be able to easily create a new Sub-VI for the fluid being tested and insert that new Sub-VI into the master program.

The LabView program also measures quantities that are of interest when monitoring system performance. The most important of which is the safety temperature. The safety temperature is measured by a sheathed, ungrounded type T thermocouple placed at the bottom of the copper block in between two of the plug heaters. This measurement is important because the system component most susceptible to damage from high temperatures is the heat plugs. The plug heaters could incur damage when the safety temperature reaches 200°C. Because of this, the tester must monitor the safety temperature as they increase the wattage to reach higher temperatures and fluxes at the spray surface. However, the spray surface will provide sufficient flux for boiling curve generation well below the maximum capacity of the heat plugs.

Other temperature readings that are taken include the inlet and outlet temperatures of the condensing coil. With this delta T and the known area of the coil, in addition to the mass flow rate of the cooling water, the amount of heat rejection from the chamber can be found. This will be useful to the tester to adjust the amount of heat rejected from the system as needed, whether through the inlet temperature or the mass flow rate. The pump reservoir temperature also features a thermocouple to measure the temperature of the fluid going into the atomizer. This information will be important to check if the amount of heat being rejected from the condensing coil is sufficient. The temperature of the fluid going into the atomizer is important because if the fluid that is being sprayed onto the copper surface gets hotter and hotter, then less heat is being taken from the surface, because the spray is at a higher energy level. If the temperature of the fluid continues to increase, steady state conditions will not be obtained.

6. Experimentation

6.1 Test Procedure

1. Position the heat block to the desired position using the setscrews on the ring stand.

2. Close and tightly seal the chamber cover. Check atomizer position to ensure proper heat block placement.

3. Prime the system by opening the top of the pump reservoir and filling it with the test fluid. Make sure the valve at the bottom of the reservoir is open. Fluid should flow down tube and through pump and up to the filter.

4. Start LabView program.

5. Start running cooling water through condensing coil.

6. Turn on heat plug power source. Begin at a voltage of 20 volts and run for 4 minutes to ensure heat block has an acceptable heat up period before turning on spray.

7. Once surface temperature nears boiling temperature, turn on pump power supply.

8. Close bypass valve, so all fluid is going to the atomizer.

9. Use the bleed valve to release air in the tubing before atomizer.

10. Once fluid has reached bleed valve, adjust bypass valve so that flow to the atomizer is acceptable.

11. Increase voltage to heaters incrementally until a new steady state has been reached for that voltage.

12. Increase voltage up to the point that either the inflection point of the boiling curve or the safety temperature have been reached.

13. Once sufficient boiling curve data has been obtained, turn off voltage source, but not the pump to allow the system to cool.

14. Before additional tests can be done, the fluid should be switched with fresh fluid, or the new fluid type. This is to ensure proper reservoir temperature.

6.2 Boiling Curve Determination

Because of the interference from the variac, a large number of our data sets contained significant electronic noise. To prove that our data reduction technique was viable, a data set was taken after the system was at maximum temperature, the voltage source was turned off, and the system was run for several minutes while boiling still occurred. Figure 12 is the resulting temperature profile for the three thermocouples in the heat block and the calculated surface temperature.

[pic]

Figure 12: Temperature Profile

Heat flux was estimated from temperature data as a function of time. This is shown in Figure 13. Heat flux began at 55 W/cm^2 and declined to 25 W/cm^2 during this experiment.

[pic]

Figure 13: Flux Change with Time

The lower end of the boiling curve for water was created by graphing the heat flux versus the superheat (surface temperature minus the saturation temperature). The results are shown in Figure 14. Several points from the published boiling curve for water are also shown on this graph. The boiling curve is within a factor of 2-3 of the published values for water.

Since the cartridge heaters used in this apparatus are only rated up to 200 ْC, we were not able to explore upper reaches of the boiling curve for water. The maximum surface temperature that we could achieve was 105 ْC and this had a superheat of approximately 3 ْC.

[pic]

Figure 14: Generated Boiling Curve

6.3 Uncertainty Analysis

In order to set a confidence level for the data, the uncertainty of each measurement was calculated. This was completed using the governing equations and known errors in the equipment. Each type T thermocouple has a 0.75% error on the given measurement. The holes in the neck of the heat block are positioned .25 in apart using a mill dial; therefore the assumed error in distances is .001 in. With these known errors, a partial derivative of Fourier’s Law was taken and calculated the error predicted by each measurement device. The error was calculated at a known temperature measurement. The results are shown in Table 1.

|Variable |Nominal Value |±Uncertainty |% of error in Qsurf |

|Qsurf |21.5 W/cm2 |17.5 W/cm2 |- |

|Δ x |.00625 in |.0000254 in |0.01% |

|k |385 W/m-k |0 |0.0% |

|Th |408.3 K |0.5412 K |22.65% |

|Tm |417.6 K |0.5784 K |66.24% |

|Tl |430.8 K |0.6312 K |11.09% |

|----------------------------------------------------------------------------------------------------------- |

|Variable |Nominal Value |±Uncertainty |% of error in Tsurf |

|Tsurf |404.8 K |3.137 K |- |

|Δ x |.00625 in |.0000254 in |0.02% |

|k |385 W/m-k |0 |0.0% |

|Th |408.3 K |0.5412 K |36.47% |

|Tm |417.6 K |0.5784 K |54.4% |

|Tl |430.8 K |0.6312 K |9.11% |

Table 1: Uncertainty Results

From the table, one can see that the surface temperature (Tsurf) is predicted within ±3K. This error comes primarily from the middle and top thermocouples because they are used many times in the governing equations. The uncertainty in the surface flux (Qsurf) is ± 17.5 W/cm2 which seems unacceptable, but at higher fluxes of 150 W/cm2 the predicted error is only 10%. .

7. Product Evaluation

Water was used as the test fluid for this apparatus because of the familiarity of it. It is a good way to benchmark the apparatus; if it will run with water, then it will run with anything. Because of this, some electrical interference problems arose that would not have occurred if flouinerts had been used. Another problem that testing with water caused was that it was very difficult to get enough superheat to fully develop the boiling curve. Since the boiling point of water is at 100ºC, a considerable amount of heat supplied from the heaters was needed. Again, if flourinert had been used, which boils at a considerably lower point, the amount of heat produced by the heat block would have been sufficient in developing the boiling curve further than the one that was generated using water.

There are four areas of needs that the test apparatus satisfies: data reduction, data acquisition, system operation, and manufacturing.

The test apparatus satisfies most of the data reduction needs. It gathers real time data needed to generate the boiling curve through the LabView program with real time heat flux measurements. The slope of the boiling curve will give the fluid’s heat transfer coefficient. The inflection of the boiling curve (refer to fig.2) is the critical heat flux for the fluid. In actuality, the critical heat flux occurs at the apex of the curve, but for Spraycooling applications, the inflection in the curve is called the critical heat flux since one does not wish to run the system so close to the actual critical value.

The data acquisition needs have also been met. The chamber pressure, spray temperature and saturation temperature are all measured with sensors, or calculated through LabView. The surface and reservoir temperature, and a heat flux of 150 watts/cm2 are all controlled. The surface temperature is controlled through the variable voltage source connected to the plug heaters, which also controls the amount of flux at the surface. The mass flow rate of the condensing coil can be controlled through the faucet of the sink being used, or if needed an additional pump can be easily inserted into the system to provide a greater mass flow. The reservoir temperature is regulated through the heat rejection from the condensing coil as well as the relatively large amount of fluid in the reservoir at the bottom of the chamber in addition to the pump reservoir compared to the amount of necessary working fluid needed by the atomizer. This disparity of volume allows the fluid temperature to come to an equilibrium associated with the steady state of the system.

All system operating needs have also been met. The current system is able to accommodate water. Except for the exterior tubing and some adhesives, all seals, joints and body materials are resistant to all fluids including flourinerts. The exterior tubing can easily be replaced by stainless steel piping to accommodate this need. Atomizer positioning is also a key element to operation. The atomizer is easy to change, since the housing for the atomizer can be removed by simply unscrewing the housing from the chamber cover. Then the new atomizer can be press fit into the housing once the previous one has been removed. The atomizer housing also provides for a consistent spray location in the center of the chamber. Then to get the desired spray distance and angle, the heat block can be adjusted through the ring stand that supports it. The ring that the block is on is secured by three setscrews on the rods. These can be loosened to adjust the ring stand height to the desired position for an optimal spray.

System manufacturing needs have been satisfied. The overall size of the test apparatus is desktop size with a stand made for it that is 36 inches long 30 inches tall. The pumping system features an inline filter developed by ISR to remove particles before they enter the atomizer. All apparatus materials are non corrosive to all types of fluids, except for the current exterior tubing.

8. Economic Analysis

Labor:

| |Hours |Rate |Total |

|Research |400 |20.00 |$8,000 |

| | | | |

|Manufacturing |500 |20.00 |$10,000 |

| | | | |

|Testing |400 |20.00 |$8,000 |

| | | | |

|Total Labor: |1300 |20.00 |$26,000 |

| | | | |

|Labor Costs |$26,000 |

|Material Cost |$1,715 |

|TOTAL |$27,715 |

9. Recommendations

The test apparatus described in this report has three aspects that should be further developed in order to ensure accurate, complete results for all possible test fluids, including electrical interference in data reduction, fluid evacuation, and plumbing modifications.

9.1 Electrical Interference

The first and most important aspect of any future work that needs to be done on the test apparatus will be to change the electrical input to the plug heaters. Currently the apparatus uses a Variac variable AC voltage source to power the heaters. This power supply causes interference in the thermocouple readings. This interference has been minimized through modifications of the prototype. These include using sheathed, ungrounded type T thermocouples so that the thermocouple themselves are least effected by the electrical interference. Type T thermocouples provide the most accurate results and are the least affected by the electrical interference. Also, a 10 MΩ resistor has been placed between the DACU’s high and low COM channels to limit the interference going to the 16-ch thermocouple FET MUX card. Another step taken was to surround the thermocouple junction inside the copper block with Arctic Silver 5 to ensure thermal, but not electric, conductivity. Even after these modifications, the temperature measurements still have a variance of +/- 4 degrees. Since a considerable amount of the calculations done in LabView are based off of these temperature measurements, any variance in voltage readings manifest themselves to a large degree of error in the boiling curve generation (see sec.6.3).

Another possible source of electrical interference comes from fluid infiltration of the ceramic insulation cast around the copper block. When breaking the cast on the first heat block made, a blue substance was discovered on the ceramic that was in contact with the copper. This indicates that water was coming into contact with and reacting with the copper. This could have contributed to an electrical field around the copper, affecting the thermocouple readings.

9.2 Fluid Evacuation

The second aspect that should be further developed is the fluid evacuation between tests. It is important that the previous test fluid be removed from the system before a new fluid is introduced since it could affect the results with the new fluid. An evacuation method needs to remove the fluid from the plumbing lines and the chamber reservoir. Removing the fluid from the exterior plumbing would be done by simply disconnecting the pipes and draining them. The reservoir is exterior as well, so draining that would be done in a similar manner. So, this evacuation method would have to be checked to ensure that this process of draining the plumbing would be sufficient in removing enough of the previous test fluid so as not to interfere with the next test.

9.3 Plumbing Modifications

The last aspect of any future recommendations involves preparing the test apparatus to accommodate any reactive type of fluid. The only modification to the test apparatus would involve changing the exterior plumbing from plastic tubing to stainless steel. This would ensure that the test fluid does not react in any way to the material.

Once these modifications are made to the data reduction portion of the apparatus, full boiling curve test data should be obtainable with minimal error involved for any fluid tested.

Appendices

A: Drawing Package

B: LabView Block Diagram and Front Panel

C: Bill of Materials/Cost

D: Uncertainty Analysis

-----------------------

Figure 4: heat block

Figure 1

Bleed valve

Valve

Bypass

Reservoir

Pump

Pump

Filter

Atomizer housing

Chamber

Spray

Boiling cruve

0

20

40

60

80

100

120

-35

-30

-25

-20

-15

-10

-5

0

5

10

Tsurface-Tsat

Flux

Flux

Series1

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download