External Hard Drive: A DFMA Redesign

Transcription

1 University of New Mexico External Hard Drive: A DFMA Redesign ME586: Design for Manufacturability Solomon Ezeiruaku

2 1 EXECUTIVE SUMMARY The following document serves to illustrate the effects that Design for Manufacturability and Assembly (DFMA) can have on a product and the savings it can produce for a company. The Boothroyd and Dewhurst (B&D) and Modified Westinghouse (MW) methods are examined using an external hard drive. The individual parts of the external hard drive are compared to each other and parts with long assembly times are discovered. An external hard drive was chosen because of its high retail price of approximately $100. Because of the relatively high price of this product, we believe the manufacturers could be turning a large profit. We also believe that with using DFMA tools, they can reduce their assembly times and in turn, increase their profit. Following the analysis of the original design of the hard drive, Hundal s rules of design are used to redesign and reduce the number of parts. A discussion of each redesigned part and pictures of the parts assist with understanding why the changes and reductions were made. The parts to be redesigned and/or reduced depended solely on the B&D and MW analyses. Of course, the redesign and reduction of some parts had a greater impact than others did. The redesigned parts with the highest impact are: The removal of HD to metal frame washers, which cuts assembly by 12% (B&D) and 8% (MW). The redesign of HD to metal frame screws cut assembly time by 28 seconds or 15% in both analyses. Also, the removal of the rubber feet reduced assembly time by 21 seconds (B&D) and 37 seconds (MW) This resulted in a time savings of nearly 10%. The results of this study show that both the B&D and MW methods give very similar results. It also shows that with a correct redesign of the hard drive the assembly time can be reduced from around 340 seconds (B&D) and 360 seconds (MW) to 167 seconds (B&D) and 198 seconds (MW). This is a 51% reduction for the B&D method and a 45% reduction for the MW method. The cost savings that this represents is $0.57 (B&D) and $0.53 (MW) on the assembly of each part. This also translates into an increase in production of 86 units (B&D) and 64 units (MW) per employee per 8 hour shift. This improvement is achieved through the redesign and removal of eight carefully selected components. Our analysis shows a major improvement in the external hard drive assembly time. If DFMA were applied to the hard drive during the early design stage, the product could have been more successful and more profitable. If the redesign were implemented and the retail price remained constant at $100 dollars, the company could have seen an increase in profit of nearly 50%.

3 Contents 1 Executive Summary Introduction Approach Boothroyd and Dewhurst Analysis (B&D) Modified Westinghouse Analysis (MW) Comparison of B&D and MW Hundal s Rules for Design Redesign Redesign of Threaded Spacers Reduction and Relocation of HD to metal frame insulators Removal of HD to metal frame washers Reduction and Relocation of HD to metal frame screws Redesign and Reduction of metal frame to plastic frame screws Redesign of power circuit plastic screws Redesign of power button Redesign of rubber feet B&D Analysis of Redesigned Hard Drive MW Analysis of Redesigned Hard Drive Economics Conclusions Appendix A Appendix B Appendix C Appendix D References... 25

4 2 INTRODUCTION The team decided to evaluate a Western Digital 500 gigabyte external hard drive. We analyzed and redesigned the hard drive using various Design for Manufacturing and Assembly tools. The hard drive, seen in Figure 1, is made to store information digitally and uses a simple USB connection and is powered from a standard 120 volt wall outlet. The hard drive is required to look visually appealing and withstand years of start-up and shut down. It must also take up the smallest about of space possible. Figure 1: Western Digital 500 gigabyte external hard drive The team chose to analyze this hard drive to evaluate the economic benefit for Western Digital as well as determine if the product can be redesigned to save the company money and possibly lower the price of the unit. Although this model hard drive has been discontinued, similar hard drives today can be purchased for $70 [2]. Four years ago, this hard drive was purchased brand new for $ APPROACH The team decided to analyze the hard drive using two methods. First, the Boothroyd and Dewhurst (B&D) analysis for estimating manual assembly times will be discussed. The B&D analysis focuses primarily on individual part handling and insertion times (see appendix A). The second analysis used was the Modified Westinghouse analysis (MW) which breaks down component assembly into ten categories and tacks on a time penalty for each category (see appendix B). In both cases, the predicted times for manual assembly will be calculated and, using common wages, the total cost to assemble the hard drive can be evaluated. By examining each component, the team expects to find specific parts that can be redesigned to decrease

5 assembly time and in turn, reduce cost. Lastly, an economic analysis will be performed to compare the current design with the teams redesign and determine what kind of savings can be had from the redesign. Part Name # Operations Handling code Handling Time (sec) Insertion Code Insertion Time (sec) 4 BOOTHROYD AND DEWHURST ANALYSIS (B&D) A Boothroyd and Dewhurst (B&D) analysis was performed on the hard drive by first completely disassembling the external hard drive. The hard drive disc was not removed as this would likely cause permanent damage to the entire product. The team justified this decision because the incredible preciseness of the disc drive components require a machine to manufacture and thus, any redesign of these components would not provide any significant reduction in assembly cost. Also, the circuit boards were considered pre-assembled and did not require any soldering techniques as these were likely produced by a separate company. The external hard drive was then assembled in steps to discover any issues with assembly (difficult alignment of holes, hard to place screws, etc.). The external hard drive and all its components can be seen in Figure 2.

6 Threaded Spacers 4 '10' 1.5 '93' 3.5 Circuit Board 1 '30' 1.95 '06' 5.5 Circuit Board Housing 1 '30' 1.95 '29' 11.5 Circuit Board Housing Screws 2 '11' 1.8 '38' 6 Flip Subassembly 1 '00' 1.13 Hard Drive (HD) 1 '30' 1.95 '07' 6.5 HD to metal frame insulators 4 '01' 1.43 '31' 5 HD to metal frame washers 4 '31' 2.25 '16' 8 HD to metal frame screws 4 '11' 1.8 '59' 12 Grounding Tape 1 '00' 1.13 '00' 1.5 IDE connector 1 '01' 1.43 '30' 2 Molex power connector 1 '00' 1.13 '30' 2 Sheet Metal frame 1 '30' 1.95 '08' 6.5 metal frame to plastic frame screws 4 '11' 1.8 '38' 6 power button plastic screw 2 '11' 1.8 '39' 8 power button sheet metal screw 1 '12' 2.25 '38' 6 power button 3 '30' 1.95 '55' 11 rubber feet 2 '11' 1.8 '31' 5 Rotate Subassembly 1 '00' 1.13 wrap around body 1 '30' 1.95 '34' 6 plastic frame 1 '30' 1.95 all parts secured to plastic fram Table 3 in Appendix C shows the B&D analysis of the external hard drive. Figure 2: Disassembled external hard drive Using the B&D analysis, we estimated the total assembly time to be 336 seconds (~5.6 minutes). By plotting the assembly times, we can clearly see that there are parts and procedures that have much longer assembly times than the others. Figure 3 shows a graph of the B&D

7 Estimated Assembly Time (sec) analysis. Using the total estimated assembly time and the average wage of a United States and Chinese worker ($12 and $0.65, respectively) we estimated the hard drive to cost $1.12 if assembled in the US and $0.06 if assembled in China. Very low costs considering the product sold for $ Boothroyd & Dewhurst Analysis of an External Hard Drive Figure 3: Plot of B&D analysis of a hard drive Based on Figure 3, it can be seen that screws, washers and the power button take up large portions of the assembly time. In order to reduce the assembly time, the team chose to redesign and reduce the number of these components. 5 MODIFIED WESTINGHOUSE ANALYSIS (MW) Similarly to the B&D method, the Modified Westinghouse method was applied to the hard drive to estimate the manual assembly time. Figure 4 shows a graph of the MW analysis by part. The analysis estimated a total assembly time of 382 seconds (~6.4 minutes). Again, using common worker wages, we can estimate the cost of assembly for one hard drive: In the US it will cost $1.27 and in China it will cost $0.07). Table 5 in Appendix D shows the completed MW analysis of the hard drive.

8 Estimated Assembly Time (sec) Modified Westinghouse Analysis of an External Hard Drive Figure 4: Plot of MW Analysis of a Hard Drive 6 COMPARISON OF B&D AND MW The MW method estimated an assembly time that was 5.4% higher than the B&D method and both were consistent in showing areas that need improvement. Although the MW method had a longer assembly time the B&D method still estimated some processes to require longer assembly times than the MW method. The area that had the largest discrepancy was the installation of the rubber feet. This is due to the fact that the MW method accounts for adding glue to the feet. Even with the discrepancy, we decided to redesign the rubber feet to reduce assembly time. Figure 5 shows a graph of the B&D versus the MW methods.

9 Estimated Assembly Time (sec) Manual Assembly Time Estimates B&D vs. MW Boothroyd and Dewhurst Modified Westinghouse Figure 5: B&D vs. MW for a hard drive 7 HUNDAL S RULES FOR DESIGN The redesign of the parts is done following Hundal s rules for design, a set of questions that help the designer make sure the design is at its best. These rules focus on clarity, simplicity and safety, and give some general guidelines for the designer. There are nine design components that need to be considered for successful product design: 1. Function- Have a clear understanding of what the product is supposed to do. The product s main function and all of its secondary function (if any) need to be fulfilled. 2. Shape Design- The structural properties of the product need to be carefully considered. Make the product as simple as possible to ease manufacturing. 3. Ergonomics- The product is for human use, it must be comfortable to use/operate. 4. Safety- The product must be safe for use 5. Manufacture- Design the product with DFM and DFA in mind. 6. Transport- Size, weight and shipping regulations need to be considered at all times. 7. Operation- Consider how the product will operate. It should be quiet and nonintrusive. 8. Maintenance- Maintenance should be easy and quick. Spare parts should be readily available.

10 9. Environment and Cost- Designing for the environment must be made important. Reduce manufacturing pollution and design the product so it can be easily recycled. These guidelines are the most used tool regarding the redesign proposed in this study, specifically minimizing the number of parts, using the least amount of assembly tools as possible and using parts that are both easy to handle and easy to assemble, meaning symmetric or noticeably asymmetric parts, large enough to be assembled by hand. The aforementioned guidelines and rules are used in the redesign analysis that follows. 8 REDESIGN We believe that re-designing and/or eliminating the spacers, insulators, washers, screws, power button and rubber feed will greatly reduce the overall assembly cost of the hard drive. Using the new components we will re-analyze the assembly times using both the B&D method and the MW method. 8.1 REDESIGN OF THREADED SPACERS The external hard drive uses internally threaded spacers to accept screws to hold the circuit board and circuit board housing, see Figure 6. These threaded spacers have been pressed into place (most likely by a hand-held tool) and have a long assembly time. There are also 2 spacers not being utilized. The 2 unused spacers could be required for a different model, but we decided to eliminate them to show the cost benefits. We decided to replace the threaded spacers with binding posts shown Figure 7. Figure 6: Threaded Spacers Figure 7: Binding post to replace threaded insert

11 8.2 REDUCTION AND RELOCATION OF HD TO METAL FRAME INSULATORS There are 4 rubber insulators attached the sheet metal frame that accept the screws that hold the hard drive disc in place, see Figure 8. They are required to be pressed in place and we believe 4 insulators to be unnecessary as the back of the hard drive is already up against the sheet metal frame. Reducing the number of insulators from 4 to 2 still allows the hard drive disc to be located by three points and no excessive movement will occur. Rubber insulators Relocate tabs and insulators centrally Rubber insulators Figure 8: Rubber insulators for hard disc drive 8.3 REMOVAL OF HD TO METAL FRAME WASHERS The screws that hold the hard drive disc to the sheet metal frame are awkward to pick up, tangle easily, and have to be held in place until secured, see Figure 9. We believe the rubber insulators work perfectly as washers and these washers can be completely removed. Completely removing the washers would reduce the B&D assembly time by 41 seconds or 12%. Screw Washer Figure 9: Washers for hard drive to metal frame 8.4 REDUCTION AND RELOCATION OF HD TO METAL FRAME SCREWS As with the reduction and relocation of the rubber insulators, we will redesign the hard drive to use only 2 screws to attach the hard drive disc to the sheet metal frame, see Figure 10. Doing this will reduce the B&D time for these parts from 55.2 seconds or 16.5% of the assembly time to 28 seconds or 12.5% of the assembly time.

12 Relocate screws to utilize this hole here Figure 10: Relocation points for hard disc drive to metal frame 8.5 REDESIGN AND REDUCTION OF METAL FRAME TO PLASTIC FRAME SCREWS The sheet metal frame is secured to the plastic frame using 4 very small screws. Locating these screws into their holes is also challenging because they are recessed, see Figure 11. By relocating the screw holes so they are easier to assemble using just your fingers and by reducing the number of screws from 4 to 3 we can cut assembly time by 25%. Figure 11: Metal frame to plastic frame screws 8.6 REDESIGN OF POWER CIRCUIT PLASTIC SCREWS This hard drive is turned on by pressing the power button which presses a smaller button on a circuit board. That circuit board is secured to the plastic frame using 2 self-tapping screws,

13 see Figure 12. By using only one screw and an alignment pin feature on the plastic frame we can ease assembly by no longer needing to hold the board in place and will only need one screw to secure it in place. This screw grounds the circuit board to the metal frame: it must stay Remove this screw and use a centering pin Keep this screw Figure 12: Power circuit screws 8.7 REDESIGN OF POWER BUTTON The power button shown in Figure 13 is secured to the plastic wrap around body using 3 plastic pins and a riveting type procedure. This currently takes 42 seconds or about 12.5% of the total assembly time. This can be greatly reduced by using just one simple snap device. It also has the benefit of allowing easy replacement should the power button need to be replaced. Remove this plastic rivet Figure 13: Plastic rivets for power button 8.8 REDESIGN OF RUBBER FEET This hard drive has the option to be stacked or used vertically. If it is to be used vertically, rubber feet are used to steady the hard drive on a desk. They also elevate the hard drive to allow air to flow underneath to aid in cooling, see

14 Figure 14. Remove rubber feet Figure 14: Rubber feet glued into place Since the user can choose which orientation to use the hard drive, we thought it makes sense to eliminate the step of gluing rubber feet to the hard drive and just package self-adhesive rubber feet with the hard drive. This way, the user can place them where they want; on the long side to allow for horizontal placement and stacking or on the short side with vents to allow for vertical use. See Figure 15. Figure 15: Replacement rubber bumpers 9 B&D ANALYSIS OF REDESIGNED HARD DRIVE The B&D analysis was once again performed using the redesigned components. The results compared to the original design can be seen in Figure 16. The redesigned components show significant assembly time savings. Overall, the estimated assembly time (and assembly cost per unit) was reduced by 51%. The results of the B&D analysis can be found in Table 4 in Appendix C.

15 Estimated Assembly Time (sec) Boothroyd & Dewhurst Analysis of an Exthernal Hard Drive Original vs. Redesign Original Redesigned Figure 16: B&D analysis of an original vs. redesigned external hard drive. 10 MW ANALYSIS OF REDESIGNED HARD DRIVE Again, we re-analyzed the hard drive using the MW method and the redesigned components. Results can be seen in Figure 17. Similar assembly time savings to the B&D method were seen using the redesigned components. Using the MW method, the estimated assembly time (and assembly cost per unit) was reduced by 45%. The results of the MW analysis can be found in Table 6 in Appendix D.

16 Estimated Assembly Time (sec) Modified Westinghouse Analysis of an External Hard Drive Original vs. Redesign Original Redesigned Figure 17: MW analysis of an original vs redesigned external hard drive 11 ECONOMICS Table 1 shows a comparison using both the B&D and MW methods of the original assembly time and cost to the redesigned assembly time and cost. Assembly Time (sec) US Cost ($) china cost ($) B&D-Original MW-Original B&D-redesigned MW-redesigned Table 1: Assembly times and costs for B&D and MW analyses It is interesting to note that the discrepancy between the redesigned B&D and MW is now larger. Table 2 shows the percent differences between the two methods. Overall our analysis shows it will reduce assembly times by nearly 50% (51% for B&D and 45% for MW). % difference Original % difference B&D B&D vs MW Original vs. redesigned 5.37% 50.58% % difference redesigned % difference MW B&D vs MW Original vs redesigned 15.60% 44.59% Table 2: Percent differences between B&D and MW methods

17 If we create a very simple profit calculation, we can say that: Profit = Units Sold * $/unit assembly cost parts cost If we say that the $/unit remains the constant as well as the parts cost, we can say that: Profit gain from redesign = Profit redesign Profit orignial = assembly cost original assembly cost redesign If we use the results from the Boothroyd and Dewhurst analysis and we speculate that approximately 3 million of these products were sold worldwide and assembly was performed in china, the estimated assembly cost for the original design would be $183,000. Similarly the estimated assembly cost for the redesigned product would be $91,000. This tells us that the profit gain from redesigning the hard drive would be roughly 50%. We can assume this to be a low estimate because the parts cost would not remain constant: in fact it would decrease because we were able to reduce the number of screws, washers, and insulators. 12 CONCLUSIONS The Results of this report show that by following Design for Manufacturability and Assembly guidelines, companies can drastically increase their profit margins. This project also showed us that by using either a Boothroyd and Dewhurst or Modified Westinghouse method, you can determine which aspects of your product need to be re-worked in order to decrease the assembly time. By applying Hundal s rules for design, we were able to redesign and reanalyzed the assembly times and determine approximate profit margins. If in industry, a company decides not to utilize DFMA techniques, it is easy to see how a competitor can develop cheaper and more efficient assembly techniques to drive the other company out of business. It s important to notice that the results obtained could be even greater, but not so much more, if the parts that don t have such a large impact on the original assembly time were also redesigned and not only the eight parts that were redesigned based on their assembly times. Though even the time could still improve the difference would be marginal and therefore not considered in this study, which had as an objective to find the largest flaws in the design and re-work them in order to greatly reduce the total assembly time, which as the results clearly show, was achieved.

18 13 APPENDIX A

19

20 14 APPENDIX B

21

22 15 APPENDIX C Table 3: Boothroyd and Dewhurst Analysis- original Part Name # Operations Handling code Handling Time (sec) Insertion Code Insertion Time (sec) Total Operation Time (sec) Threaded Spacers 4 '10' 1.5 '93' Circuit Board 1 '30' 1.95 '06' Circuit Board Housing 1 '30' 1.95 '29' Circuit Board Housing Screws 2 '11' 1.8 '38' Flip Subassembly 1 '00' Hard Drive (HD) 1 '30' 1.95 '07' HD to metal frame insulators 4 '01' 1.43 '31' HD to metal frame washers 4 '31' 2.25 '16' 8 41 HD to metal frame screws 4 '11' 1.8 '59' Grounding Tape 1 '00' 1.13 '00' IDE connector 1 '01' 1.43 '30' Molex power connector 1 '00' 1.13 '30' Sheet Metal frame 1 '30' 1.95 '08' metal frame to plastic frame screws 4 '11' 1.8 '38' power button plastic screw 2 '11' 1.8 '39' power button sheet metal screw 1 '12' 2.25 '38' power button 3 '30' 1.95 '55' rubber feet 2 '11' 1.8 '31' Rotate Subassembly 1 '00' wrap around body 1 '30' 1.95 '34' plastic frame 1 '30' 1.95 all parts secured to plastic frame 1.95

23 Table 4: Boothroyd and Dewhurst Analysis-redesigned Part Name # Operations Handling code Handling Time (sec) Insertion Code Insertion Time (sec) Total Operation Time (sec) Threaded Spacers 2 '10' 1.5 '30' 2 7 Circuit Board 1 '30' 1.95 '06' Circuit Board Housing 1 '30' 1.95 '29' Circuit Board Housing Screws 2 '11' 1.8 '38' Flip Subassembly 1 '00' Hard Drive Disc (HD) 1 '30' 1.95 '07' HD to metal frame insulators 2 '01' 1.43 '31' removed HD to metal frame screws 2 '11' 1.8 '59' Grounding Tape 1 '00' 1.13 '00' IDE connector 1 '01' 1.43 '30' Molex power connector 1 '00' 1.13 '30' Sheet Metal frame 1 '30' 1.95 '08' metal frame to plastic frame screws 3 '11' 1.8 '38' power circuit plastic screws 1 '11' 1.8 '39' power circuit sheet metal screw 1 '12' 2.25 '38' power button 1 '30' 1.95 '30' removed Rotate Subassembly 1 '00' wrap around body 1 '30' 1.95 '34' plastic frame 1 '30' 1.95 all parts secured to plastic frame 1.95

24 16 APPENDIX D Table 5: Modified Westinghouse Analysis- original Part Name End to End Orientation Rotational Alignment Part Thickness Part Size Handling Condition Insertion Clearance Insertion Direction Insertion Condition Fastening Fastening Process Time/Each Operation Number of Repetitions Threaded spacers Circuit Board Circuit Board Housing Circuit Board Housing Screws Flip Subassembly Hard Drive (HD) HD to metal frame insulators HD to metal frame washers HD to metal frame screws Grounding Tape IDE connector Molex power connector Sheet Metal frame metal frame to plastic frame screws power button plastic screws power button sheet metal screw power button rubber feet rotate assembly wrap around body plastic frame Repitition Time

25 Table 6: Modified Westinghouse Analysis-redesigned Part Name End to End Orientation Rotational Alignment Part Thickness Part Size Handling Condition Insertion Clearance Insertion Direction Insertion Condition Fastening Fastening Process Time/Each Operation Number of Repetitions Threaded spacers Circuit Board Circuit Board Housing Circuit Board Housing Screws Flip Subassembly Hard Drive (HD) HD to metal frame insulators removed HD to metal frame screws Grounding Tape IDE connector Molex power connector Sheet Metal frame metal frame to plastic frame screws power circuit plastic screws power circuit sheet metal screw power button removed rotate assembly wrap around body plastic frame Repitition Time

26 17 REFERENCES [1] Boothroyd, G., Design for Assembly - A designer s Handbook, Department of Mechanical Engineering, University of Massachusetts, Amherst, Nov [2] Newegg Inc. (2013) Retrieved April 1, 2011, from % &IsNodeId=1&name=500GB