Design and Selection of Materials Project UCF Formula Car Exhaust Headers

Luscuskie, Mark, [email protected] Wendrick, Nathan, [email protected]

No plagiarism was conducted in this project

_________________________________ Mark Luscuskie

_________________________________ Nathan Wendrick

Abstract The purpose of this project is to identify the best material and manufacturing process for a component. The component chosen is the headers of the SAE at UCF formula car. Acting as a path from the engine of the car to the muffler, they must exhibit several important properties for proper function. They must not be cost prohibitive for the team to use on this years car. A limit of $500 dollars was set for the finished product. Other requirements were to properly contain all exhaust gasses, be structurally sound at all times, and be non-toxic in all ways. Properties to be optimized are: flow of exhaust gasses, structural rigidity, and chemical stability. Properties to be minimized are: weight, cost, and mechanical degradation. Materials properties were analyzed and compared based on component operating conditions. Material indices were derived and respectively maximized and minimized. Shape factor was evaluated based on possible loading scenarios. After all material information was determined, possibilities were compared and contrasted and a final material of 304 stainless steel was chosen. It was found to have the best compromise of desirable and necessary characteristics. Once 304 SS was chosen as the material, manufacturing processes were researched and scrutinized to determine the best method. Constraints based on compatibility, availability, and cost were placed and weighed to determine the best method. Mandrel bending of tube and cutting and welding to shape was determined to be the best manufacturing process for the headers. The header material and process selection yielded the best results for the headers. A melting temperature of 2000 F (1100) degrees ensures that mechanical properties will be well maintained during operation. Chemical properties of stability ensure that at high temperatures very little oxidation or carbon precipitation will occur. From improved geometry, mass is calculated to be 2.75 lbs. down from 5.51 lbs. for the previous year. Flow is improved over the design from the previous car along with heat loss therefore energy. Cost also fits requirements totaling to $221.20. Overall, proper material selection and process selection are manifested in a header that is fits all criterion and will be well suited to its task. Table of contents I. II. III. IV. V. Purpose, Requirements, Objectives Material selection Process selection Conclusion References

Objectives and requirements The purpose of this project is to select the best material for the headers on the SAE at UCF formula car. The car is a student designed and built vehicle meant to compete in the SAE formula intercollegiate design series. The competition simulates a manufacturing company that has contracted a design team to develop a small Formula-style race car. The race car is to be evaluated for its potential as a production item. The target marketing group for the race car is the non-professional weekend autocross racer. Each student team designs, builds and tests a prototype based on a series of rules whose purpose is both to ensure onsite event operations and promote clever problem solving.(SAE) This means that the headers for the car must be effective, manufactureable, and economical.

The chosen design of the car is bent-tube space-frame construction. Then engine chosen to power the car is from a Suzuki GSXR-600. It is a gasoline powered, 600cc displacement, four stroke cycle, inline four configuration, engine. It has four exhaust ports all on the same side of the engine. The headers are used as an enclosed path from the engine exhaust ports to the muffler of the vehicle. They must contain and flow hot exhaust gases, they must be mechanically secure, and they must be safe when operating the vehicle. In order to contain hot exhaust gases, they must have a high working temperature and must have an even higher melting temperature. To be mechanically secure, they should be designed and constructed properly. The design needs to be strong enough to support their weight, both statically and dynamically, as no road surface is perfectly flat and smooth. For occupant safety, they must not be toxic to operate at any time. They must also not be toxic to manufacture, as students of the college will manufacture them in a basic machine shop. In designing the headers all above requirements must be met, several objectives will also be strived for. The first objective is to keep the weight as low as possible. Since the performance of the car carries significant scoring weight in the competition, vehicle weight is very important. An objective to maximize airflow of the headers is also important. Though a calculated backpressure needed for the engine is configured by length of the header, smooth internal flow is very important. Less restriction translates into better performance. For longevity, one objective is to minimize chemical and mechanical degradation of the headers. Decomposition and degradation decreases the mechanical properties of the headers, and negatively influences performance. The last objective is to keep the total cost under $500. The competition does take cost into account for scoring, but the main reason for the cost objective is that the organization tries to be savvy with its money. Material Selection To begin material selection, a market survey and analysis of possible materials were performed. Passenger vehicles were analyzed for production and aftermarket material choices of exhaust manifolds and headers. More detailed analysis was also performed of requirements, with respect to maximizing and minimizing certain material indices. The requirements needed to satisfy the design criteria are many but with no certain criteria standing out as the most important. Because of this no one material stands out as being perfect. It becomes obvious that each material has tradeoffs, but some have more benefits than downfalls. In considering materials and manufacturing, a brief look was taken at current headers in use. Low end production vehicles tend to use cast iron or cast steel. Higher end production vehicles are more readily seen with headers constructed out of hydro-formed stainless steel. Exact grade varies, but is generally a variation of stainless steel. The aftermarket for high performance vehicles generally utilizes steel in a bent-tube construction. The lower end headers use a coated mild steel, the mid-price headers use 304 stainless, while the high performance use 321 or titanium. Both stainless and titanium may be coated, but the coating is not as much of an aesthetic feature, like for mild steel, but is applied for performance. For construction of the headers for the formula car, weight effectively eliminates casting. Price requirements eliminate hydro-forming as well. The only reasonable construction method remaining is some type of bent-tube construction using minimal welding where needed or where bends exceed max bending capabilities. Having bent tube construction requires that the material be readily shaped into tube, probably round in shape. For a price nearly any material can be made into a tube, but with the cost limitations for this prototype car, it basically necessitates that the tube be already available for purchase. Tubes already in production include steel, stainless steel, titanium, aluminum, Inconel, GFRP, and

CFRP. Initial observations of cost and heat requirements yield choices of steel, stainless steel, and titanium. There is no loading done through the headers to hold the force of the car the headers they only have to support their own mass and possibly small additional mass from the added muffler. While driving, the headers may experience 2 gs of constant acceleration and peak gs or 3-3.5 gs over bumps. More extreme than the loading requirements is the very high operating temperature. The highest risk is a heat concentration on the outside of a bend. This can cause cracking and possibly a blow out if the temperature of the tube wall. This is caused by the material getting so close to the melting temperature that it weakens and blows out under the small internal pressure of about 7 psi of back pressure. The exhaust gas has been measured as high as 1650 F and can peak higher under some engine conditions. Material indices Because of the extremely high temperature required for this application, a Max Service Temp vs. Strength index will be applied to quickly narrow down the selection options and make analysis easier. The average service temp will be around 1350F (700C) so the service temp guide line will be placed here keeping in mind that the peak temp can be as high as 900-1000C. Another guide line is placed at 30 MPa because of the elevated temperature and the extremely thin wall tube that will be utilized for highest shape factor; it needs to have strength against buckling.

It can be seen from this that titanium is just barely usable, nickel alloys, and stainless steel alloys can be used. Ceramics qualify based on these guidelines. But, it is known that for the complex shape needed for the headers (seen below) will be difficult for ceramics. Ceramics will be shown to be too brittle with a material index later.

The fracture toughness has two guidelines. One is a fracture toughness of atleast 11 MPa.m1/2 and the other guide line is Kic/f. For Kic/f the material needs to be above and to the left of the guide line. We want the material to have a high Kic to protect against both thermal cracking and cracking due to sudden impulses and bumps. It can be seen that almost all ceramics are eliminated from the selection criteria.

The thermal expansion and thermal conductivity are important properties to consider because there will be such a large temperature difference between when the engine is running and when the engine is off. For performance the smaller thermal conductivity is better because the exhaust gases stay hotter and move at a more

consistent, higher, velocity through the header helping reduce pumping losses. For our material, L/a = 107. It can be seen of the main materials in consideration which are steel, stainless steels, titanium, nickel alloys; titanium and stainless steel have the lowest thermal conductivity and can be assumed the best for making horsepower. Another consideration that was thought of is using lightweight aluminum alloy for the tail pipe after the headers converge into one and the exhaust gasses cool some.

It was found that since stainless steel and aluminum are dissimilar materials on the galvanic scale this was not a good idea. It was also likely that the aluminum would weaken at the still elevated temperatures(~1000F) and not be very durable. The next index that will be considered is an index related to cost. This is one of the most important considerations of this selection and one of the most binding next to the temperature. Since the strength needed is quite small, the indicies used will be cost per unit volume. The cost per unit volume is the most important because the outside diameter needs to be a common inch size so it can be mated to other components of either 1.375 or 1.25 and the inside diameter as calculated by flow constraints needs to be 1.18. This yields either very thin (0.035) wall tube or thicker (0.0975) wall thickness tube. We will assume that all tubing can be purchased at 1.25 x 0.035 and the cost per unit vol will be the chart considered.

Here the derived material index is at 2x105. This is rather expensive, but we have such a small volume of material that the actual cost is not extremely expensive. This eliminates tungsten and almost eliminates titanium, but it will still be considered further at this point. Stainless steels, Nickel alloys, and CFRP and many ceramics. CFRP was eliminated earlier because the required service temperature is too high and ceramics were also eliminated based on the fact of too brittle and difficulty to form. At this point stainless steel is determined to be as effective in terms of temperature as nickel based alloys and cheaper so it is more viable and also is more available. At this point titanium, mild steel, and stainless steel seem to be the most viable choices. Mild steel has the problem of oxidation both during running and after shutdown. The problem occurs when operating the car in wet/rainy conditions and oils and paints burn off of the surface causing a loss of protective barrier. The wet conditions allow water to get on the metal can cause oxidation and rust very quickly and result in a very short lifespan for the material. Due to this only titanium and stainless steels will be considered from here. I made phone calls and requested quotes for both stainless steel and titanium tubing. The most common alloys of stainless steel and titanium were found to be Grade 2 for Ti and 304 and 321 stainless.
Titanium Grade 2 Stainless Steels 304 321

Quoted ($/ft) $ 21.80 $ 4.12 $ 19.95 This table is the quoted prices of the tubing. as can be seen, the price for titanium and 321 Stanless steel were very similar. The amount of tubing needed is 10 feet as calculated from solidworks, and there is a shipping charge of $30 associated with each. This cost is at the outer fringe of the allowed cost because a collector (about $120) and tail pipe (about $40) also need to be added. 321 Stainless Steel is out because of the high cost because it only offers small increase in expected life with no difference in weight or manufacturing. 304 Stainless Steel and Titanium will be considered from here.

As seen by the material indices the two general best materials will be steel and titanium. Below in figures 1 and 2 are material properties of 304 stainless and Grade 2 titanium. Titanium is nearly half

the weight of steel yet the yield strength is as good or greater. There are several drawbacks to titanium however. Price is the most obvious, but fabrication needs are more specialized than stainless steel as well.

Figure 1 304 stainless

Figure 2-Grade 2 Titanium

The durability of the titanium and stainless steel as well as the thermal conductivity has both been shown to be satisfactory. Now the manufacturing and weight savings need to be considered in relation to the cost of each. As mentioned before, the total cost of the entire exhaust cannot exceed $500 but it is beneficial if any money can be saved. This includes the $120 collector that takes each individual header into one solid tail pipe and the tail pipe itself, $40. This leaves $340 and for the titanium tube it will cost $248.00 just for the material leaving $92 for manufacturing costs, which are higher for titanium than stainless. The stainless tube costs only $71.20 with shipping leaving $268.80 excess. This project is real and budget is important so the excess money is of large importance for stainless steel. Next, using the model I made in SolidWorks, analysis was to determine if the added cost of the titanium would be seen in the weight savings. I modeled a brand new exhaust to use the minimum amount of tubing possible but still meeting the necessary geometry the final model looks like:

With this model mass analysis can be done. Due to the bending/forming or radii, the profile needs to be a round (tube) cross section. First, the old size and thickness of tube 1.375 x .065 was analyzed with steel to see how much can be gained by simply optimizing the geometric properties. Results are shown below:

It can be seen from the analysis that the headers by themselves using a thicker tube than needed weigh a whopping 5.51 lbs. The weight of the headers is important because it adds to the total weight of the car and makes acceleration done by the car more difficult. This weight needs to be reduced. In order to reduce the weight I optimized the shape factor next by using the thinnest available tube that can still be formed and joined. This is a tube with a 1.25 OD and a wall thickness of 0.035 this is the geometry that was used for the next two analyses. First Stainless steel was analyzed with the new, higher shape factor which can be seen on the left and then titanium was analyzed on the right, results can be seen below.

It was determined using the mass properties in SolidWorks that the headers made out of stainless steel would cost only 2.75 pounds whereas the titanium would weigh a mere 1.52 pounds. It was determined at this time, while the 4 pound loss from the lesser shape factor and the titanium, the cost and manufacturing difficulties (discussed more in-depth later) the titanium is not worth almost 4 times the cost of the stainless. Stainless steel is 3.77 times less expensive than the titanium making it a $261 saving with only a 1.25 pound penalty. Because of this 304 Stainless steel is the best choice for this component.

The shape of the headers could reasonably be anything that is necessary. Using a solid cross section is out of the question because one of the constraints of the headers it that exhaust gas must flow through them. Practicality constrains the shape to something that tube is already formed into. The four major shapes that tube comes in are: square, rectangle, circle, and symmetrical airfoil. The symmetrical airfoil is difficult to analyze, is very costly, and does not provide any benefits over the other three shapes. Square and rectangle tubing are very similar. Rectangular tubing however, does exhibit some undesirable features. Because it is not an even cross section, strength is not even. This means that during bending, special care must be taken to orient the tube properly. This is not a big issue, but many other factors negatively affect the choice of rectangle tubing. Square and rectangle tubing can only be bent in planes perpendicular to one of their flat faces. This presents a problem when trying to bend the header tubes into complex angles. Since the design of the tubes necessitates several angles of rotation of the planes of bending. Square tubing becomes severely impractical. When considering strength, square or rectangle tubing may have an advantage if loads were known and constant. However, loads on the headers vary immensely. Under static conditions, the headers must support the weight of themselves in the direction of gravity. Under cornering, the headers must support more than their full weight in a direction normal to the direction of travel of the car. The formula car is capable of a 1.6G turn. Also, because of the imperfect road surface, the car will experience a significant amount of force due to bumps

or dips. Altogether, the non-regular loading condition, along with the ease of bending, circular tube is the best for mechanical properties. Another problem with square or rectangle tubing is that when internal surface are of the tube is compared with volume, it is higher than that of a round tube. This decreases the ability of the headers to flow exhaust gasses. Combine this better flow with better mechanical properties and better fabrication properties and the choice of round tubing is obvious.

Process Selection Once material has been determined, manufacturing concerns must be addressed. The most important constraint of manufacturing is that the processes used must be compatible with the material. Very little will be accomplished if the process chosen will destroy or ruin the properties of the material used. Since the material will be an austenitic stainless steel several manufacturing options present themselves. The most economical solution will be to utilize a bent tube configuration. With this in mind, properties for making the tube properly, shaping the tube properly, and joining the tube properly must be made. The constraints are to use processes that are chemically compatible, and that will not decrease the mechanical properties. All processes performed on the premanufactured tube will be performed inhouse at the SAE shop. This manufacturing environment means that certain industry norms may be overlooked in favor of addressing concerns of the shop. Tube will be the most suitable for construction. Premanufactured tube is readily available, in both a rolled eletro-welded (REW) type and a drawn-over-mandrel type. When flat plate is cut, rolled into a tube and welded, the welding process creates a heat affected zone. This heat affected zone is more brittle than the parent metal, and is more likely to crack under extended vibrations. The weld also creates stress concentrations at the interface of the filler metal and the parent metal. These two decreases in mechanical properties make using rolled electro-welded tube undesirable. When manufactured by a drawn-over-mandrel (DOM) technique, the billet of material is forced over a cone and through dies to form the tube shape. Though this process is much more expensive than REW, the lack of weld is very desirable. The use of DOM tube is nearly half as expensive as using REW, but tube cost is still much less than other materials, not negating the choice of material. With a tube type of DOM chosen, the next step is choosing how to shape the tube. The most effective method will be to bend long lengths of tube to the desired configuration, and the join them. If construction geometry prohibits bending long lengths, a greater number of shorter lengths must be used and then joined. Several different types of benders exist, but the best type for bending thin wall tube without kinking the material is a mandrel bender. By drawing the tube around a specially formed mandrel, bends are achieved while keeping a constant cross-section. Some removing of material will also be required for shaping. For this, machining is the most suitable method. Water jet cutting could be suitable, but setup costs would be prohibitive. Laser cutting could also be used, but costs for setup would be high as well. Machining is the easiest for the students of the SAE club to accomplish in-house, so nothing would need to be outsourced. In a broad scale manufacturing environment, a water jet would probably be more economical, but for the construction of one set of headers, machining will be the best.

Since material shaping is decided, the last constraint to address is joining. Several compatible joining methods are possible for 304 stainless. The three broad processes are mechanical fastening, brazing, and welding. Mechanical fasteners are a fairly cheap solution, but properly sealing a very complex intersection of tube would be difficult. Brazing could be an economical solution, but brazing is used infrequently, so little is known about it at the SAE shop. Welding presents itself as the best option. More specifically tungsten inert gas (TIG) welding. Compared to other welding processes, TIG allows for the most precise control over the physical and chemical properties of the weld bead. Another consideration for a process is to apply a thermal ceramic coating to the headers. This helps flow of the headers by keeping the exhaust gas temperature high. Because of the high temperature, velocity of the gas is kept high. This high velocity improves the evacuation of the area when the next exhaust charge enters the header tube. This translates into higher horsepower for the engine which leads to better performance of the car. Done properly, the application of a ceramic coating also improves the surface finish of the tube. The improvement of surface finish decreases the stagnation at the surface of the tube. This also increases the flow of the headers, which increases horsepower. An ulterior motive for applying a ceramic coating is that it decreases the temperature of the engine bay. With less heat transferred from the header tube, less heat is transferred to the air in the engine bay. A cooler engine bay increases horsepower by lessening the work done by the water pump to circulate coolant through the radiator, thus allowing for a smaller water pump. Very little information is published Nearly all ceramic coatings are applied by physical vapor deposition. This is one of the few processes that is compatible with ceramic and leaves little room for choice of process. Most coatings, as witnessed by the name, are ceramic. Some variation of material occurs depending on the setup of the coating company. The material will probably be an alumina or zirconia derivative. The majority of ceramic coatings, for 4-cylinder motorcycle headers are in the $150, for coatings both inside and out, a worthwhile investment. Due to the nature of this project, manufacturing costs, after buying the rolled tube, can be neglected. Cost for this project for the SAE shop will total the $150 for the ceramic coating. Under a standard setting, for a production run of 5000, time to cnc mandrel bend tube is approximately .25 hours per set of headers. At a shop rate of $100/hour, bending equates to $25. Machining on the tube is estimated at .5 hours; with a rate of $100/hour machining will be $50. Welding the tube is estimated at approximately 1 hour, shop rates for TIG welding are approximately $50/hour. Total cost for manufacturing would be equivalent to about $125 per header assembly. However, with a run of 5000, a discount of 10% is approximated. The sum of manufacturing, including ceramic coating leaves a final manufacturing cost of $262.50.

Conclusion The choice of material of 304 Stainless Steel for the headers mirrors many real world systems. Stainless holds up to high temperatures much better than carbon steel tube. It is also much lighter than cast steel or cast iron which many production cars use. For example, cars in formula ford SCCA often used

carbon steel tube headers coated in ceramic. Other cars in the Formula SAE often use stainless, similar to the choice of our material of headers. The future of material selection for exhaust headers and manifolds is unlikely to change significantly any time soon. Most materials used have not made major improvements. Metallic choices may be finetuned slightly by varying the composition to achieve better results. Possible production exhaust manifolds could see a slight change in chemistry to improve cost or casting operations. The next foreseeable step would be to go to a carbon filament reinforced material. Due to ever growing demand for green vehicles, weight savings is a major concern. Carbon filament materials normally possess a low weight compared to metal counterparts. Before any of this can happen though, major improvements in the impregnating material of the carbon must be made. Though carbon itself can sustain the high exhaust gas temperatures, the current polymers are far below practical. References
http://www.suppliersonline.com/propertypages/304.asp#physical http://www.aedmotorsport.com/ http://www.matweb.com/ Phone interview with Mike Fine Aerogas http://www.matweb.com/search/datasheet.aspx?MatGUID=69814d8b0a4c43f9b116674067b3e154 http://students.sae.org/competitions/formulaseries/rules/2011fsaerules.pdf http://www.matweb.com/search/datasheet.aspx?MatGUID=1748ca73d11e4353b2aa700bfb119dfb&ckck=1