Table of Contentsweb.cecs.pdx.edu/~far/Past Capstone Projects/2011/FSAE... · Web view(using Figure 11.14 from Fundamentals of Heat and Mass Transfer 6th by Incropera and DeWitt)

FSAE Cooling

Progress Report – Winter 2011

Group Members

Craig Mclain

Reuben Ness

Riki Hopkins

Portland State University Advisor

Dr. Lemmy Meekisho

Table of Contents

Table of Contents...............................................................................................2

Introduction.......................................................................................................3

Mission Statement.............................................................................................4

Project Plan........................................................................................................4

PDS Summary.....................................................................................................5

External Search.................................................................................................. 6

Internal Search...................................................................................................7

Concept Evaluation............................................................................................7

Detail Design Progress........................................................................................9

Conclusion:...................................................................................................... 10

Appendix A:..................................................................................................... 11

Appendix B:......................................................................................................13

Appendix C:......................................................................................................14

Appendix D...................................................................................................... 17

Progress Report FSAE Cooling Page | 2

Introduction

Formula Society of Automotive Engineers (FSAE) is an international engineering competition where students design, build, and test small-scale autocross racing vehicles. The given rules and constraints create a real-world engineering challenge. Approximately half of the FSAE teams have cooling-related issues during competition, making it such that they could not finish all the events or adversely affecting their scores. From catastrophic engine destruction and head gasket failure, to “lesser” issues of hard starting and performance reduction can be explained by insufficient cooling. This happens because most teams address cooling as an afterthought and then use something off the shelf instead of an engineered solution.

Portland State University’s (PSU) FSAE team in 2010 fell into this trap and did not have sufficient cooling. The three problems we had with the 2010 car due to insufficient cooling:

The car ran hot during testing and competition. The car would overheat when run at an extended time at idle. The car had hard hot starting due to excessive temperatures.

Running hot during testing and competition

The 2010 car had inadequate cooling capacity due to undersized radiator and fan, and improper airflow due to poor radiator positioning. Since the radiator was an afterthought for this team, not enough time was allocated in analyzing the necessary cooling loads. The radiator design (sizing) relied heavily on broad assumptions that were not well understood. The fan sizing followed the flawed assumptions used for sizing the radiator, proving inadequate. Radiator placement and orientation were packaging constraints without considering fluid dynamics (airflow). Since the team did not have many resources, ducting and fan placement were neglected.

There were two major problems with the cooling design of the 2010 car. The first is that the importance of the cooling system was underrated. The second is that during the design and build process, inadequate resources were allocated to the project.

Overheating at idle

During most testing sessions and driver’s training, extended idle periods are normal and should be expected. The periods of idle should not be of concern if the cooling system is engineered properly because the cooling system load is significantly less.

For the 2010 car, the heat load of the engine was unknown, and the assumptions were obviously inadequate. Also the fan placement was poor in that it only pulled air through a small area of the radiator. The fan sizing was dependent on the assumptions of the heat load as well.


Hard hot starting at excessive temperatures

A race car is expected to start every time. Therefore, it should always be in a state for easy starting. To be in this ideal range, many factors need to be considered, and cooling must not be a problem. For the 2010 car, the engine was too hot to start during competition, causing the team to fail an event.

For the coolant temperature to always be within the ideal operating range, all factors need to be considered. The cooling system must be able to handle all loading conditions. These include: different drivers, different tracks or events, ambient air temperature, and loading patterns.

Mission Statement

The FSAE Cooling capstone team will design a new solution for the cooling of the 2011 FSAE car. Our goal is to produce a solution with correct engineering methods and to understand the physics, math and engineering behind those methods. The final design will be prototyped and documented, with all of its performance characteristics quantified.

Project Plan

The dates in the following table 1 are goals and deadlines for completion of project milestones. Dates other than due dates are subject to change dependent on project requirements.

Table1: Project plan.

Project MilestonesTask Start Date Finish Date Due Date Status

Initial Brainstorming June 22 Sep 12 CompletedResearch Possible Solutions June 22 Sep 12 Completed

Initial Design Sep 5 Nov 18 CompletedPDS Report Jan 3 Jan 31 Feb 9 Completed

PDS Report Presentation Jan 24 Jan 31 Jan 31 CompletedDesign Evaluation Feb 7 Feb 25 Completed

External Search Presentation Feb 7 Feb 14 Feb 14 CompletedDesign Nov 18 Jan 17 Completed

Progress Report Presentation Feb 21 Feb 28 Feb 28 CompletedProgress Report Feb 21 March 7 March 7 CompletedPrototype & Test Jan 17 June 13 In progress

Redesign Jan 24 April 29 In progressManufacture Feb 7 Mar 22 March 22 In progress

Assemble/Install/Test March 22 June 13 Not started


Repeat Prototype & Test April 29 June 13 Not startedAssemble/Install/Test March 22 June 13 Not started

The corresponding Gantt chart is shown in Fig. 1.

Figure1: Gantt chart displaying the project plan.

PDS Summary

In the PDS document, customers, design requirements, and potential risks, involving the progress and design of the product, were specified. Customers for the project were identified as follows:

Internal Customers Viking Motorsports Portland State University Capstone Dr. Etesami (Mechanical Engineering Capstone coordinator) Dr. Meekisho (Project Advisor) for internal customers.

External Customers FSAE regulations Judges of the Business Presentation Weekend Auto Crossers (End Users/Consumers) Sponsors of Viking Motorsports Mac's Radiator


The most important requirement for the product is reliability. It is crucial that the car with the new cooling system consistently maintains a top tank temperature of under 210°F in order to prevent difficulties encountered in the 2010 car; such as the loss in performance and difficulty starting the car due to excess temperatures, or overheating after idling for an extended time. Some other important design criterions include a heat rejection performance of 30hp, 1 year life in service, not to raise the center of gravity of the overall car by more than 0.5 inches, while meeting FASE regulations and keeping to a target budget of under $200. For the complete product design specification, see Appendix.

Major risks involving the design process of the product that could severely compromise the team’s success in meeting the aforementioned design criteria (or meeting them in a timely manner) are identified as follows:

The product requirements, such as the minimum heat transfer requirements or the maximum allowable power consumption of the product, are significantly under estimated.

An erroneous analysis that leads to a failure in meeting the specified requirements for the product. Unexpected failure of a part/component of the product. Ordered parts not arriving on time.

External Search

The engine used in the car is a Honda CBR600F4i 4-cylinder motorcycle engine, and is specified by Viking Motorsports. Therefore, cooling technology similar to those used for motorcycles with similar engines are considered and studied. The following lists a few existing technologies that were considered.

The first option is to use the radiator from the CBR600F4i. From further research, however, it was found to be less than half of the required capacity. Motorcycles designed for the street are designed with a much higher average speed range than the car used by Viking motorsports, which run at an average 30mph in competition. Furthermore, the time in which the engine is at full throttle is much greater for a race car than a motorcycle on the street. Therefore, these motorcycles radiators are not designed to reject as much heat at the lower air flow rate compared to the car, even though they use the same engine. Honda also has motorcycles with larger displacement in that series. A radiator from CBR900RR was also studied, which was closer to the size required by the product, but was found to be difficult to package. The radiators are also difficult and expensive to obtain. In search of the existing radiator with a larger size, car radiators were also considered. However, even the smallest car radiator found, which was a radiator from a Honda Civic manufactured around 1995, was too large in order for practical packaging. Radiators from several racing series that also use motorcycle engines, such as D-sports Racers, Mini Sprints, and Midgets, were studied. They were all found to be too large and geometrically impractical for packaging. The last option explored is a custom-made radiator, which is expensive, but has the distinct advantage being able to be sized perfectly and packaged in a way that the team specifies.


Internal Search

Ideas on what are needed and how to make the custom radiator work from the project were brainstormed. This work overlaps into design work as well. The first idea was to use two radiators, to provide a symmetric look for improved esthetics and balanced drag profile. However, two radiators with the same cooling capacity as one large one would weigh more, and was concluded to not yield a net merit to the car. Similarly, an external oil cooler was explored, but it was determined that it added little cooling, and was much more efficient to use the oil-to-water cooler that the stock motorcycle used to transfer the cooling load to the water.

Much of the brainstorming phase was done in Solidworks, examining how the radiator would sit and how the hoses would look. One idea was to use a dual-pass radiator, to make the hoses shorter and easier to package, but it would make the radiator wider, and had efficiency consequences. Though it really is a design decision, the size of the radiator was discussed, as packaging is a heavy constraint. Also the placement angle and height was explored in attempt to balance the height of the center of gravity and radiator performance. Various duct ideas were also tossed around, with two main ideas: external ducts and internal ducts. External ducts help with aesthetics and provide low external drag. Internal ducts improve cooling efficiency and provide mounting for the fan. From the esthetics aspect, the external duct was a must-have, while the internal duct may happen later if time allows. Fan sizing and placement were also considered, with final decisions being made in design.

Sizing analysis for the radiator is done in Appendix A, while fan sizing analysis is done in Appendix B. The multiple iterations of solid model design are in Appendix C.

Concept Evaluation

Leading into the concept evaluation portion of the project, the team had several different ideas for solutions to cooling. Table 1 shows a decision matrix created by the team outlining these ideas and weighing them against customer requirements. Using this tool, the team was able to make an informed decision. This was an important step in the design process. Using two radiators versus one radiator would dramatically influence the design.


Table1: Radiator decision matrix

One radiatorDual

RadiatorsExternal Oil Cooler (&

One Radiator)External Oil Cooler (& Two Radiators)

Weight 10 7 9 6Efficiency 8 9 8.2 9.2

Drag 8 6 7.5 5Cost 9 4 8.5 3.5

Complexity 10 7 9 5Maintenance 9 7 8.5 5

Installation 9 5 9 3Packaging (Fit on Car) 10 6 9.5 0

Customer Requirements 10 8 10 5Totals: 83 59 79.2 41.7

Once all concepts had been evaluated and detail design began, vendors for the team’s parts had to be selected. Table 2 shows the decision matrix used to chose vendors for the radiator and fan selected. The team evaluated two custom radiator shops and an off the shelf solution for the radiator. The fan had to be off the shelf, but evaluation of the possible vendors was still important. Solution is shown in Appendix D.

Table 2: Radiator and Fan Vendor Decision matrix

Radiator Mac'sMac's Scaled

Other Custom fabrication

Other Custom fabrication scaled

Off the shelf

Off the shelf scaled

Cost $315 7 >$300? 4.5 Unknown 4Timeline Fits 10 Unknown 4 Unknown 4Capstone Fits 10 Unknown 4 Doesn’t fit 0

Excellent 10 Good 9 Unknown 4Interface Excellent 10 Good 9 Unknown 4Totals: 47 30.5 16

Fan Mac'sMac's scaled Jegs Jegs scaled

Cost $100 7 $60 9Timeline Fits 10 Fits 10Capstone Fits 10 Fits 10Quality Excellent 10 Questionable 5

Interface Good 8 Good 8Actual

performance Excellent 10 Questionable/poor 4

Totals: 55 46


Detail Design Progress

Design and Manufacture:

The first step in this project was to determine the boundary conditions for cooling loads. With no published data available, some assumptions had to be asserted. These assumptions, broad but firm, lead to the ability to calculate the cooling load a system would see in this application. Once the system load was calculated, sizing calculations could be done. These two portions are the most critical part of the entire project, and because of this, precautions were taken in calculation. Within the assumptions there is a "factor of safety" built in. This "factor of safety" was added by calculating the maximum theoretical load the cooling system could ever see under ideal conditions. After this "factor of safety" was implemented and a radiator size calculated, the radiator was up scaled a small amount to add (~17%) capacity to the cooling system.The next step was packaging. Working with Viking Motorsports, the geometry of the radiator was designed and redesigned to fit their criteria. Several iterations of radiator (SolidWorks) models were presented to Viking Motorsports, and inputs on the design exchanged. Once it was certain that the radiator itself would not have to change in design, the model was sent to be produced by Mac's Radiator, and is shown in Appendix D.

Remaining Detail Design:

Radiator Mounts

The radiator must be rigidly mounted to the chassis of the 2011 FormulaSAE car. To do so, mounting points were added to the model of the radiator so brackets can be attached. The position of the radiator with respect to the chassis has been somewhat dynamic over the course of this project. Therefore, detail design of radiator mounts is problematic. To account for this, the mounts are being designed such that, once produced, they can be used in more than one radiator position.

Solid Hose

Rather than using conventional rubber heater hose, the capstone team and Viking Motorsports have decided to use solid aluminum tubes for coolant flow to and from the radiator. This solution is lighter and more efficient that rubber hoses, but harder to implement. Detail design has been done for these solid hoses, but it is also dynamic. The hoses cannot be bent until the radiator position is finalized.


Conclusion:

The capstone team has completed a PDS document detailing the requirements of this project in terms of customer expectations. The team has also completed internal and external research to compare possible solutions to the PDS requirements. Detail design and manufacture has been completed for the largest portion of the project and verification through testing can begin. The team is on track for the completion deadline of June 13. Detail design is being developed for smaller, less complex, parts of this project, and will be completed on time.


Appendix A:

Radiator Sizing Analysis

The object of radiator sizing analysis is to determine an effective fin area for the radiator so that core dimensions can be established. A proper core size is critical for effective cooling of the car's engine. If the cooling is insufficient, the engine's temperature will increase above the 210°F stated in the PDS and the engine will fail. There is a point at which there can be too much cooling of the engine, but this would require such a large radiator, that it would not fit the PDS criteria. The radiator must be as light, efficient, reliable and inexpensive as possible. The expected result of this analysis is an effective area in square feet.

Result: Aeff=723.9 ft2

This area represents an effective fin area for the radiator. This area is assuming straight (non louvered) fins and laminar flow through the core. Both of these assumptions drive up the area needed for cooling. The only way to correlate this effective area with the true needed area is through testing. There are general guidelines that can be used to steer in the right direction from this point onward, but the final determining factor is testing.

Given: Average Engine RPM = 7377 (from test data), 210°F Engine/Top tank temperature, Water Flow Rate: 8gpm @ 6000rpm & 10.4gpm @ 7400rpm, Heat Rejection: Q=mCpΔT, 1HP=5.28gpm*°F, Cp,air=1.0kJ/kg*K, densityair=0.08018lb/ft3

Assumptions: Constant properties, Steady state, steady flow, No losses, 80HP engine (shaft power), 30HP heat to cooling system, 10-15% humidity

Average Shaft PowerMax Shaft Power

∗100=%heat ¿cooling

From track simulation -> 30% cooling load

From track data -> 34.4% cooling load

Find: Determine an effective (fin) area for a radiator to be used under the given conditions.


Solution:

Using the Ɛ-NTU method:

ε=Ch¿¿

Water: Cp=(0.2388 Btu/lb*°F)*(4.184J/g*K)=1Btu/lb*°F)

Cmax=3984 Btu/h*°F

CminCmax

=0.639 A=NTUCmin

Uqmax=Cmin(T h ,i−T c ,i)

Th,i=210°F, Tc,i=100°F, Cmin=2547.6 Btu/h*°F

qmax=2547.6Btu /h∗° F(210° F−100° F)=110.0HP

ε= qqmax

= 60110

=54.5%

ε=C c¿¿ -> Tc,out=100°F (check, good)

C c=ChT h ,∈¿−T h,out

Tc , out−T c,∈¿=3984210−170160−100

=3984 4060

=2656 Btu/h∗° F ¿¿

CminCmax

=0.667

(using Figure 11.14 from Fundamentals of Heat and Mass Transfer 6th by Incropera and DeWitt)

NTU≈1.2

NTU=U h AnCmin

U h Ah=1.2∗2656=3187.2Btu/h∗° F

U=25~50 W/m2K 1 W/m2K=0.17611 Btu/h ft 2° F

Assuming; U=25 W/m2K≈4.403Btu/h ft 2° F

A≈723.9ft2


Appendix B:

Fan sizing analysis

The objective of the analysis is to find the correct size and rating of the radiator fan required to meet the PDS criteria. The applicable PDS criteria are to achieve the minimum heat rejection, maintain 210°F radiator temperature, and to be reliable, light and inexpensive. The expected result of the analysis is a rating in cubic feet per minute (cfm).

Results: 1014cfm

For most fan manufacturers this relates to a 12 inch diameter fan, which on inspection of the many pictures we got from competition seem to be about the same size. This is also about four times the size of the fan last year that was miserably undersized.

Given: 30hp heat rejection (from engine); fluid properties of air: cp=1.0kJ/kgK=0.241BTU/lb°F, ρ=0.08018lb/ft3, Tambient=100°F

Find: Airflow required to remove 30hp of heat.

Assumptions: Steady-state, constant properties, no losses, reasonable air temperature increase of 65°F

Solution:

Q=m×cp×∆T

m= Qc p×∆T

m=30hp×42.44 BTU /min

hp

0.241 BTUlb℉ ×0.08018 lb

ft3×65℉

=1014 cfm

All numbers for fluid properties and conversion coefficients came from Fundamentals of Heat and Mass Transfer 6th Edition by Incropera and DeWitt.


Appendix C:

Design Iterations

Figure 2 is a screen shot of the first design iteration produced by the capstone team. This design is a dual pass radiator, making it easier to package the coolant hoses to and from the engine. The radiator's fins are suppressed in this model for ease of use in SolidWorks. The core size of this radiator is 12.5-inch wide, 12.5-inches tall, 1-inch thick with 1in tanks on each side. For mounting, there are C-channels on both the top and bottom of the radiator. The dual pass design is generally more efficient, but was determined to be less efficient for the Formula SAE application. Therefore the design changed dramatically from the first iteration.

Figure 2: SolidWorks model, design number 1.


Figure 3 is another screen shot from SolidWorks showing the second design iteration of the radiator. The changes from the first design iteration to this one are significant. The design changed from a dual pass system to a single pass and the geometry of the core changed. The core remained 1-inch thick because of standard core sizing, but the dimensions of the core changed because of packaging. This radiator is 16.5-inches tall and 11.5-inches wide with 1-inch tanks on either side. For mounting, there are two pins welded onto the top and bottom tanks in one corner.

Figure 3: SolidWorks model, design number 2


Figure 4 is a SolidWorks screen shot of the final design iteration of the radiator. There are very few changes from the previous design iteration to this one. The geometry remained the same, but the mounting changed. The top mount remains a pin connection, while the bottom was changed to a plate for rigidity reasons. Without the bottom mount being a plate, the radiator had the potential to pivot about the mounting pins. This would cause the airflow to no longer be perpendicular to the radiator, causing a drop in efficiency. A radiator cap was also added to this model.

Figure 4: SolidWorks model, design number 3. Final Design.


Appendix D

Current Parts

The radiator is back from manufacturing, and is shown in Figure 5. It differs a little from the final solid model shown in Figure 4 above. The tanks ended up less than the 1-inch specified, and the C-channel was turned the wrong way. These discrepancies were deemed inconsequential.

Figure 5: First prototype radiator, in shop.


The fan was purchased based on the decisions in Table 2 and the sizing calculations in Appendix B. The fan chosen is a Spal 12-inch medium-profile fan with a rated flow of 1230 cfm, and is shown in Figure 6.

Figure 6: Final fan solution, Spal 12”, in shop.


Documents

Table of Contentsweb.cecs.pdx.edu/~far/Past Capstone Projects/2011/FSAE... · Web view(using Figure 11.14 from Fundamentals of Heat and Mass Transfer 6th by Incropera and DeWitt)