|
|
|
|
Home
Projects Services |
|
|
|
|
|
High-Throughput Production Processing Five- (5) Axis
Titanium Components III (HITHRU III)
The stresses created by the need to maintain DoD’s fleet of aging aircraft at combat readiness are well known. Sources for needed repair parts diminish, sometimes to zero; materials commonly used 50 years ago are no longer available; technical data is sometimes missing or incomplete; and so forth. These issues and others combined push depot maintenance cycles to unacceptable levels. It is only through concerted efforts that the average cycle for the KC-135, for example, was reduced to its current level of less than a year. When structural repair parts are needed but technical data is missing or incomplete, a reverse or re-engineering process is triggered. Designers must create a (3D) solid model from which manufacturing engineers and planners create a manufacturing process, including any necessary numerical control (NC) programs and work control documents. Five-axis machine tools are the most efficient for making complex aerospace structural parts but are expensive and extremely difficult to program. These complex parts can also be made on less efficient three- and four-axis machine tools, but at the cost of additional machine setups and perhaps more expensive work-holding fixtures. Depot manufacturing engineers must therefore trade off recurring and nonrecurring costs plus programming time versus additional setups as well as machine utilization costs. The CTMA HITHRU project (March 1999 – May 2002) sought to make the choices easy by increasing process planning productivity for high-speed machining of aluminum parts made on five-axis machine tools; encapsulating manufacturing knowledge to reduce the requirement for manufacturing knowledge of programmers, and producing NC programs that produced higher quality parts. It did so by addressing in detail a subset of the entire part space, concentrating on structural parts in aerospace applications where the part can be made from slab stock in two setups (front-side/back-side) on a five-axis machine tool with no intervening operations—a subset that covers a large portion of aerospace parts. HITHRU measured the dynamic responses of a Cincinnati V5 vertical machining center and created an algorithm that calculated cutting parameters (axial and radial depth of cut, feed rate, and spindle revolutions per minute) that used maximum spindle horsepower while avoiding chatter zones. Automated feature recognition and NC program generation was successful but suffered from knowledge base limitations due to limited testing on real parts. HITHRU III (July 2001 – December 2003) targeted titanium parts, multiple setups with complex work-holding strategies, and embedding advanced controller functions such as torque-controlled machining and probing cycles. Cimskil™, the software product that emerged from this project, was also ported to the Windows environment. All project objectives save one were met, but with varying degrees of success. Optimized titanium machining processes for the Cincinnati H5 were developed and validated as were the advanced controller functions. Automation of feature recognition and NC program generation was greatly enhanced as the system was exposed to new parts but still remains somewhat problematic. For parts having similar types of features that were processed through Cimskil™, automation of feature recognition and NC program generation frequently exceeded 80%. For parts with features outside the current Cimskil™ feature set, however, automation was often less than 50%. Multiple setups with complex work-holding strategies proved too difficult to accomplish within the scope of this project. Project benefits are wide-ranging and should be of significant benefit to both the depot community and to defense original equipment manufacturers (OEMs): 1. Automated feature recognition and NC program generation has been a topic of research for over a decade. This project took both to levels never before achieved, proving that five-axis manufacturing features can be addressed in this way. 2. A commercially available product, Cimskil™, emerged as a direct result of this project. 3. Cimskil™ can provide levels of automation up to and sometimes exceeding 80% for parts having recognized manufacturing features that can be made from slab stock in two setups. 4. The time required to develop NC programs for five-axis machines can be reduced by more than 50%. 5. The optimized machining processes developed in the project maximize machine productivity and produce parts of excellent quality. Accurately summarizing benefits that can accrue through the use of Cimskil™ is difficult because there is so much variation in component complexity. Consequently benefits are presented here in ranges: 1. Using traditional means to create a NC program for a five-axis milling machine can consume one week for simple parts to six weeks for large, complex ones. With programming productivity increased by as much as 80%, the savings can range from two to 25 days in the creation of the NC part program. 2. Once the part program is created, it must be validated by cutting a trial part before release to production. If problems such as dimensional errors, surface finish issues, collisions, or warping are found, the programmer must make corrections and try again. Using traditional methods, the program validation process can consume time that can easily amount to half the time taken for programming, dependent mostly on the experience and skills of the programmer. Cimskil™ programs are driven by actual solid model geometry and process parameters are based on proven best practices so the potential for human error is greatly reduced. The first part cut for validation has a much higher probability of being correct than when programs are created using traditional methods. Savings there are on the order of one to 10 days. 3. Finally, the machining processes created through Cimskil™ are optimized for high productivity. In one real case for an F-15 part at WR-ALC, machining time was cut from 13 to 3.5 hours. The part was ordered for stock by DLA with an order quantity of 200 pieces so actual time saved in manufacturing was about 1,900 hours. 4. Total time saved can therefore range from zero (when all manufacturing features and manufacturing methods are new to Cimskil™) to 35 days—just to get the part into production. Once in production, savings can also be substantial but are entirely dependent upon order quantity. For the example given, savings were on the order of 80 days. The HITHRU III project team partners were: · Cincinnati Lamb · Boeing · Sikorsky Aircraft Corporation · Technology Answers Corporation · Warner Robins Air Logistics Center · Cherry Point NADEP · Elmendorf AFB · NCMS Program Manager: Tony Haynes, tonyh@ncms.org, (734) 995-4930.
|