REDESIGN OF TURBULATOR MANUFACTURING PROCESS

Transcription

1 Proceedings of the 2004/2005 Spring Multi-Disciplinary Engineering Design Conference Kate Gleason College of Engineering Rochester Institute of Technology Rochester, New York May 13, 2005 Project Number: REDESIGN OF TURBULATOR MANUFACTURING PROCESS Chris Mabry/Mechanical Engineering Sukhdeep Sohal/ Mechanical Engineering Phil Oakes/ Mechanical Engineering Matt Romanishan/ Mechanical Engineering ABSTRACT A simple method of producing industrial turbulators involves sending a length of metal through two offset rollers to create a twist. The result is known as a twisted tape turbulator. Modifications to the end product include punch hole and end tabs that make the turbulator idea for long and/or vertical tubes. This project improves on a method developed by a previous RIT design team. The project specifically developed a new method of shearing the trubulators after they have been run through the indexer that creates a hole and a tabbed end. Automated, the system can production time for turbulator by up to 65%. INTRODUCTION Turbulators are used in numerous industrial applications, particularly in boilers and heat exchangers. The turbulator typically consists of a length metal twisted or bent in such a way that an air flow over the turbulator is disrupted. The resulting non-laminar flow allows for an increased rate of heat transfer through the walls of the boiler or heat exchanger. Figure 1- A twisted tape turbulator Fuel Efficiency LLC has been producing twisted tape turbulators for several years. This style of turbulator is created by twisting a length of metal along its entire length. The company offers several materials, length, and diameters for the turbulators. Fuel Efficiency offers a modified turbulator that is ideal for use in vertical or extremely long tubes. This turbulator has a punched 3/16 hole on one end, and welded piece of metal aligned perpendicular to the length at the other end. The hole allows the turbulator to be fished through a long tube, while the tab allows the turbulator to hang in a vertical pipe. These modifications are added to nearly ever turbulator the company produces. A previous RIT design team developed the method used by Fuel Efficiency to produce turbulators. The team developed a device known as the indexer that contains two set of rollers aligned at differing angles. The material for the turbulator is fed through the 2005 Rochester Institute of Technology

2 Proceedings of KGCOE 2005 Multi-Disciplinary Engineering Design Conference Page 2 rollers, and the twisted shape is created. After the turbulator is created and cut to length, an operator punches a hole and welds a tab on each one. The intent of our project is to improve Fuel Efficiency s current production scheme. The current indexer can create turbulator of a maximum 1.5 diameter. Fuel Efficiency would like to increase this to 3. The company would also like a revised method of shearing the turbulator after they have been through the indexer. The current method using a standard shear block tends to deform the end of the turbulator, producing an unclean and undesirable cut. Figure 2- Current Indexer and Shear Setup INITIAL CONSIDERATIONS The team initially focused on shearing the turbulators. The deformed end is a result of the twist in the material. The shear block is designed to cut flat material, and thus has a flat surface on either side of the blade to support the material as the blade engages. The turbulator has no flat surface area, but rather a sinusoidal shaped profile. As the blade engages, it attempts to flatten the profile of the turbulator before it begins to shear. This results in the deformed end that the company feels is unacceptable for the finished product. In considering a solution to this problem, the team also noticed that considerable time is spent adding the punch and the tab after the turbulator has been produced. A time study of a typical production run (650 turbulators, 12 feet long) showed that the turbulators could be produced in 1 to 1.5 days, while the punching and welding took an additional 4 to 5 days. The team realized that the company could greatly improve speed and efficiency of production if the tab and punch were created when the turbulator was sheared. DESIGN PROPOSAL The team sought a way to create a shearing method that would shear the turbulator, punch the hole, and create a tab. The obvious obstacle was the shape of the turbulator. With a profile that did not lend itself well to a standard shear; the new complex geometry would be nearly impossible. The modified shear would be possible, but only if there were a flat length a material to cut. With Fuel Efficiency s approval, the team decided to develop a system where the indexer would create a 2 long flat spot in the turbulator at the required point of shear. This spot would allow the turbulator to be supported as it was sheared, and would allow enough room for a tab and a hole to be added to end product. The finished turbulator would have a 1 flat section on each end. The design would entail creating a shear blade that could shear, punch, and tab at the same time. For the convenience of the company, each of these processes would have to be optional, such that a turbulator could be produced with only a shear, a shear and a punch, or all three. A new indexer would have to be created that could be rotated while the material was in the rollers. This would allow the indexer to twist the material for the turbulator, rotate to a neutral position while some material is fed through, then rotate back to its initial angle to create the rest of the twist. The indexer would also be revised to accommodate material up to 3 wide. INDEXER DESIGN The new indexer is redesigned to allow turbulators to be made with 3 material. Nearly all dimensions had to be increased to allow for this. The thru hole on the indexer is now 3.5 in diameter. Several options were considered to automate the rotation of the indexer. In the end, the team decided to use a motor to drive a worm gear attached to the indexer. The worm gear is advantageous because it provides a large torque increase, thus reducing the requirements on the motor. The worm gear will also lock the indexer in place once it is rotated, since the torque produced on the indexer by the material in the rollers is not enough to drive the worm. The worm gear is placed on the back of the indexer to allow space for the worm and connecting shafts. This is a change to the previous indexer, in which the wheel rested inside the indexer support block. The wheel was not fastened to the indexer in any way. Our design requires the wheel and the gear, which are on opposite sides of the support block, to be connected. Hence, a Paper Number 05110

3 Proceedings of the Winter KGCOE Multi-Disciplinary Engineering Design Conference Page 3 center ring was added that travels through the indexer support block. The wheel and the gear are both fastened to this ring Torque required to twist width = 1 inches With the previous indexer design, the indexer was intended to be rotated to its desired angle and locked in that position for the duration of the production. Our designed called for the indexer to be rotated twice for each turbulator produced. This will cause more wear on the indexer time. The revised design calls for the center ring to rotate on four bearings instead of the previous singe bushing. This will reduce the contact friction between the center ring and the indexer support block and will increase the life of the turbulator. Torque (ft-lb) P = 1 P = 2 P = 3 P = Steel Grade Chart 1- Torque Required to Twist To ensure proper contact between the center ring and the bearings, the top of the indexer support block is removable. In between the top and bottom parts of the support blocks there are several rubber bushings. The space contact between the center ring and the bushings can be adjusted by loosening or tightening the bolts connecting the two parts accordingly. MOTOR SELECTION Several calculations were performed to determine the appropriate motor to drive the indexer. The first step in selecting the motor was to determine the maximum amount of torque required to rotate the indexer with the turbulator in it. Based on design requirements, the worst-case properties of the turbulator are: Width-3 Material- Stainless Steel Thickness- 16 gauge (.0598 ) Pitch (Inches per 360º of twist) - 2 The formula used to determine torque required to twist the material is: Where: T 2 c wt P G 3 2 (1) c 2 = coefficient based on ratio of w/t. Here,.333 w = width of material t = thickness of material G = Shear modulus of material P = Pitch A parametric analysis of the formula produces an estimated maximum torque of 625 ft-lbs. The worm gear was selected based on the internal hub diameter. The diameter had to be at least 3.5 to accommodate the thru hole for the turbulator. The only gear available that met this requirement had a 10 outer diameter, and 100 teeth. Speed and torque reduction of worm gear system is derived directly from the number of teeth on the gear, 100:1 in this case. We desire the indexer to rotate the full range (140º) in approximately 3 seconds. If the indexer rotates faster than this it may become too difficult to control the stopping position. This requirement equates to 7.77 RPM desired for the indexer, or 777 RPM desired for the worm. The horsepower required for the motor was found by multiplying the torque required at the worm by the angular speed desired of the worm. This resulted in a required motor power rating of 1Hp. After an extensive catalog search and being unable to find a motor with the above specifications, it was apparent that a gearbox would be needed in order to use an off-the-shelf motor reduce the speed to the desired RPM, while still producing the required torque of at least 6.25 ft-lb. Initially it seemed appropriate that a gearbox with a ratio of 2:1 would be feasible, since motors with 1 horsepower and a rating of 1750 RPM were easily available. In this ideal scenario the required RPM would be met, but unfortunately the gearboxes of ratio 2:1 available on the market were not capable of accepting maximum input of more than 0.33 horsepower. The next phase involved searching for a gearbox with input rating of at least 1 horsepower. The gearbox that met this requirement had a ratio of 5:1, which was acceptable because this in turn required purchasing a motor with rating of 3450 RPM. Motors with this rating along with 1hp rating were easily available off Copyright 2005 by Rochester Institute of Technology

4 Proceedings of KGCOE 2005 Multi-Disciplinary Engineering Design Conference Page 4 the market. With the 1hp, 3450 RPM motor and 5:1 gearbox ratio, the angular speed of the indexer wheel was reduced down to 0.72 rad/sec (41 degrees/sec) as compared to the original speed of 0.82 rad/s (47 degrees/sec). Although the speed was cut down slightly, it still was acceptable since the maximum travel the indexer will ever undergo is 140, and this is accomplished within approximately 3.4 seconds. The gearbox selected is a Right-Angle Worm Gear Speed Reducer from McMaster with shaft input, and left hand output with the ratio of 5:1. The motor selected is a NEMA Single-Phase AC motor from McMaster with 1 hp, 3450 RPM rating. SHEAR SYSTEM The shearing system of the project is designed to shear, punch, and create a tab all in one press action. The entire process is housed between two steel plates of a die set. The die set consists of four 1.5 steel posts pressed into the bottom plate and four ball bearing bushings mounted to the top plate with a distance of 12 between plates to allow enough space for the upper and lower shear blocks. The four posts are necessary to prevent the uneven forces created by the shearing process on the plates to misalign the upper and lower block sets. The ball bearing bushings allow for ease of travel throughout the necessary range along the four posts. The upper shear block set consists of a 2 x 7 square block of A2 tool steel, a 5 x7 x ½ thick plate of D2 tool steel with a 15 degree angle, dual surfaced blade, and a ¼ punch / stripper mechanism. A2 tool steel is used for the 2 square block to prevent any deformation that may occur due to horizontal forces acting at the point of shear. The blade itself is made of D2 tool steel. D2 is a high carbon, high chrome alloy that is more resistant to wear than most other tool steels, and is therefore ideal for a shear blade. Through observation and some estimation calculations, an angle of 15 degrees was determined as an effective angle to shear both the 14 and 16 gauge stainless steels that this system is expected to encounter. The blade edge itself is precision ground to ensure sharpness and has a dual surface. This surface creates two shear edges, or essentially two separate blades slightly offset. By doing this, the material being cut will be sheared completely through on one side while the offset side will not cut completely, leaving a small amount of material on the end. The small amount of material connects the sheared off material to the end of the turbulator and can then be bent around and used as a tab. The punch / stripper mechanism consists of a ¼ blank punch of hardened steel protruding from the bottom of a 1 x 2 x 4 block of A2 tool steel along with two ¼ diameter stripper bolts housed within the A2 block connected by a ¼ x 1 x 4 mild steel plate resting two inches below the bottom of the block. Two 3 springs with a rate of 40lb/in also protrude from the bottom of the block and press against the steel plate. This mechanism is designed for use with three inch wide material of 16 gauge or less. Four 3/8 machine screws hold the 2 square block to the top plate while two 3/8 screws hold the stripper mechanism. Two 3/8 bolts travel through the stripper, blade, and block, while four (two on each side) hold the blade and the block tight. The lower shear block set consists of two 3 x 3 x 7 blocks of A2 tool steel, each with a ¼ x 1 x 7 blade of D2 tool steel bolted into grooves on the inside edges of each block. To space them apart there is a smaller 1 x ½ x 7 bar of A2 tool steel. The first block has a large 45 degree chamfer cut out of it starting ¼ behind the blade and is held to the lower plate by six 3/8 stainless steel machine screws. The last block is similar to the first except there is a ½ diameter, taper reamed, oil hardened bushing of 01 drill rod located 1/8 behind the blade. This bushing is hardened to Rockwell C 63.5 and is designed to act as a die and is aligned with the hardened steel blank punch and is precision ground to ensure sharpness. Directly underneath this bushing is a hole which allows the slug of material to escape. 1/8 beyond its edge is a 45 degree chamfer. This block is also held into place with six 3/8 stainless steel machine screws. There are four ½ stainless steel bolts that run through the entire setup, through the ½ spacer bar to help hold the assembly together. The entire assembly, when in motion, will come down with the stripper plate engaging first to hold the material in place. The blades will begin to shear, first with the through blade and then with the tab section. Halfway through the punch will perforate the material through a hole in the stripper plate. Once the shear has gone completely through, it will stop and begin to draw back. The blade slowly pulls up and the punch is pulled from the die while the stripper springs and plate hold the material firmly to the lower assembly. Once the system is fully disengaged, the end of a Turbulator will be sheared and tabbed, and the beginning of the next will have a ¼ punch. A lifter setup can be implemented to pull the tab out of the space between the lower shear blades if necessary by attaching it to the upper assembly. This way when the upper plate retreats it will pull the tab out automatically. This system is designed to be able to work without the stripper mechanism, the first lower block, or both in order to create any combination of shear, tab, and punch. It is also able to act as a straight shear for cutting flat stainless steel up to 6 wide and 14 gauge. Paper Number 05110

5 Proceedings of the Winter KGCOE Multi-Disciplinary Engineering Design Conference Page 5 BLADE ANGLE SELECTION Observation has taught us that the angle of the shear blade affects the force required to shear through a material. We attempted to determine the force needed to shear through the turbulator based upon the angle of the cutting blade. Through extensive research, we have determined that there is currently no direct way to relate the force required to shear and the angle of the shear blade. We instead sought to determine the most likely angle-force relationship and use this to determine an optimal angle for the cutting blade. In general, the force required to shear a material is a function of the ultimate shear strength of the material and the area being cut. Angling the blade reduces the area being sheared and thus reduced the total force required. The area under the blade can be considered in two different ways. We could consider as the area directly beneath the blade and below the applied force (Figure 3) or we could consider it the area under the blade and perpendicular to the contact line (Figure 4). result is incorrect, we have at least incorporated a factor of safety into our calculations. Determining the area under the blade is a geometrical problem, which can be expressed as a function of the blade angle: t 2 2 tan( ) tan(90 ) A (2) Where: A = Effective Cutting Area t = Material Thickness θ = Blade Angle The force required then becomes: F A* (3) S su Where: F = Force Required to Shear A = Effective Shear Area S su = Ultimate Shear Strength of Material An analysis of this function reveals the following plot. Applied Force Force Required to Shear Figure 3 Material Force To Shear (lbs) Blade Angle (degrees) Applied Force Material Plot 1 The plot indicates that the optimal blade angle for shear is 45º. The team feels that this angle would be unfeasible since the blade would have to travel an excessive distance to shear through all of the material. We chose instead to use a cutting angle of 15º, which does not add an excessive travel distance for the blade and but still greatly reduces the force required to shear. Figure 4 We opted to consider the second scenario, as this gives a greater estimate for the effective shear area. If the CURRENT STATUS At the time writing, revised indexer block has been completed. New roller bearings are currently in development. The team will provide Fuel Efficiency with several different roller bearings that accommodate material widths from.5 to 3. Tailoring the size of the roller bearing for the specific Copyright 2005 by Rochester Institute of Technology

6 Proceedings of KGCOE 2005 Multi-Disciplinary Engineering Design Conference Page 6 production will ensure that the roller bearings are not overstressed. A 1 width in a 3 roller bearing would apply considerable torque near the center of the roller bearing that may prove too great for the bearing to handle. The motor drive system for the indexer is complete. The single phase AC motor is capable of turning the indexer at approximately 8 RPM. Currently the electrical leads must be reversed to allow the indexer to travel in the opposite direction. We are currently developing an inverter that will allow the motor to rotate in either direction without physically changing the electrical connections. The shear system is still in production. All components have been completed, but some material hardening is necessary to ensure a robust system. All blades and cutting components will be oil hardened, then re-machined to the proper dimensions. FURTHER CONSIDERATIONS At completion of this project, Fuel Efficiency will be in possession of a modified indexer capable of twisting material up to 3 wide and rotating automatically in either direction. They will have a press that can shear a turbulator and create a hole and tab. An operator will have to feed the material through the indexer to the proper length, rotate the indexer to a neutral position, feed it 2 inches, rotate the indexer back, then run the feeder to the shear point. The operator must then activate the shear and dump the completed turbulator in a shipping crate. The system does not currently have a method of activating the shear block. Initially, the company wanted to pursue a pneumatic system to power the shear. There are several concerns that may prove this idea unfeasible. Plot 1 shows a general relationship between the shear angle and force required of the power system. The force indicated, however, is inaccurate, as it does not account for the tabbed side of the shear blade cutting and the punch. The actual force required will likely be much greater. The team cannot presently provide a function to determine the force required to shear, as our assumptions were made only in an attempt to find a general relationship. We have estimated the force required to shear the turbulator at 5000 lbs. This number will be verified upon completion of the shear block using a Tinius- Olsen Tension-Compression Testing Apparatus. We expect our estimate to be within 2000 lbs. The force likely needed to shear the turbulator will be more than can be obtained with a conventional pneumatic cylinder. Any pneumatic cylinder that can apply that much force will likely be uneconomical and hence undesirable for Fuel Efficiency. The force could be obtained using several cylinders connected in parallel, but this too would drive cost beyond what is reasonable. Furthermore, a pneumatic cylinder would cause a concern in regards to the compressibility of the air in the piston. Upon shearing through the turbulator, the cylinder may have accumulated considerable potential energy that will be released. With nothing to dampen this force the shear blade could cause considerable damage and present a danger to the operator. Implementing a dampening system would again prove uneconomical. The team explored the use of hydro-pneumatic cylinders, which would solve several problems. Hydro-pneumatic cylinders are able to provide considerable force, but are driven off an air line. They are economical, easy to use, and easy to obtain. They also provide the same damping benefit as a hydraulic piston, since the work is done by the incompressible oil. Investigation, however, has shown that hydropneumatic generally have a stroke of only 1 to 2. The minimum stroke required for our project is 6. The team recommends that Fuel Efficiency consider powering the shear block with a hydraulic cylinder. Hydraulic cylinders can provide more than enough force for a reasonable cost, and provide a damping factor to prevent damage and potential danger. The downside to a hydraulic system is the extra components that must be used, notably a pump, and the mess that tends to accompany it. The benefits outweigh these concerns, however. The team also recommends automating the production system with the use of a programmable logic controller. An auxiliary PLC can be set to run with the current feed controller to run the turbulator to a set length, rotate the indexer, and activate the shear. This would free up the operator for other tasks, and provide a more cost-effective production system. ACKNOWLEDGMENTS We would like to acknowledge our sponsor, Fuel Efficiency LLC for their help in developing and implementing this project. In particular, Joe Connelly and Lee Krohn have been a significant asset to the team. Our mentor and faculty co-coordinator is Dr. Alan Nye. Dr. Nye has been extremely helpful in pointing us in the right direction and keeping our focus on the important issues. Paper Number 05110

7 Proceedings of the Winter KGCOE Multi-Disciplinary Engineering Design Conference Page 7 Thanks to Dave Hathaway and Steve Kosciol in the Mechanical Engineering Machine Shop for their technical expertise. Copyright 2005 by Rochester Institute of Technology