JEC Group have brought together the international community of composites leaders and executives in our Composites Circle as an unique networking opportunity to meet with both peers and future partners.
Aircraft and rotorcraft parts made from high-performance composite materials are superior to earlier generation metals due to their increased strength, reduced weight, and increased service life. For these reasons, most aircraft part designers choose to create new parts out of carbon-fibre-reinforced plastic despite the challenges involved, namely that the manufacture of CFRP parts is more complex than traditional materials and can therefore make parts more costly to produce. This paper describes the method that Accudyne Systems has created for developing automation equipment to manufacture composite parts economically.
JOHN MELILLI VICE PRESIDENT OF SALES AND MARKETING ACCUDYNE SYSTEMS (published on April 2010 - JEC Magazine #56)
This “part centric” approach produces a “part purpose” machine that takes into consideration the specific design of a part with regard to ply orientation, number of laminates, and shape of the part. This paper describes this process and details several part and machine examples derived from the process. As shown in Figure 1, traditional part design is a linear process. Downstream steps are constrained by choices made in previous steps, however new decisions are not informed by considerations further down the line.
With each step, the barriers to effective automation intensify because of the increased constraints. Plies that are difficult to place, fibre angle variation, short courses may not cause problems during manual part fabrication; however challenges such as these are the source of difficulties and expense when trying to automate the fabrication of a CFRP part.
Tier One and Tier Two part producers have typically considered automation solutions that are “machine centric”, meaning that they perform in a broad mode independent of specific part characteristics such as geometry or ply orientation. Many commercially available fibre placement and tape placement machines excel at making large parts such as wings and fuselages. However, these types of machines operate inefficiently if the part has short courses, pad-ups, ply drops, significant contours or other such features. In these cases, and when production rates, reduction of scrap and quality consistency are most important, the industry requires a bold new paradigm. In a “part centric” design approach, a robust examination of the part design and the planned automation drives the overall design effort. It also leads to a series of steps which lead the part designer to developing a machine specification that is idealized for the manufacture of a specific part or a series of similar parts, hence a “part purpose” machine. Aerospace parts can have complex geometries that complicate their automated manufacture. By taking a part centric approach during initial part design, engineers can consider design changes to minimize automation challenges. By working “hand in hand” with a custom automation supplier, this process methodology allows harmonization of part and machine design to yield a completely connected process.
Although flat parts may seem simple at first examination, the inclusion of internal padups and unusual ply boundaries pose challenges to even the nimblest “machine centric” machine. If optimizing fabric placement would improve automated production and yield, it may be important to consider processes other than automated fibre placement (AFP) or automated tape laying (ATL). In some cases, a complete work cell may be the most appropriate approach as it combines multiple aspects of the manufacturing process. This use of work cells is common in many other manufacturing industries.
Clearly, the future of automating the manufacture of composite based is going to take a new approach.
The following is one option to address the new challenges:
Following are examples of various parts and the machines that were produced to manufacture them.
Traditional ATL/AFP machines were deemed to be too expensive and too slow to produce the “hat”-shaped stringers used to reinforce the fuselage shown above (Figure 4).
After careful analysis of the part the flat charge laminator was conceived and built. Shown in Figure 5, this machine manufactures the part using a combination of fabric and unidirectional prepreg material at a rate of 25 kg/hr or a stringer every 15 minutes. This rate will allow the manufacturer to supply the customer with parts to accommodate the rate of 10 planes per month (published target manufacture rate).
Figure 6 shows another common aircraft part, a composite radome. Radomes typically utilize materials that are transparent to radar signals and are best manufactured using fabrics due to the shape and required performance characteristics. Manufacturing these parts entails the use of precut pieces of prepreg fabric which are then hand laid on a mandrel. Making each unit by hand is expensive and time consuming.
After careful examination of the part, material type and orientation, engineers developed a work cell to manufacture the inner laminates for a radome using glass/epoxy fabric. The concept utilizes a means for manipulating fabrics and adhesive films onto a tool (Figure 7).
Within this cell, both the tool and the material move, allowing straightforward and accurate application of fabric and film. After the material is placed, a compaction step is required and a butt joint between plies is executed to adhere to the part designer’s specification that all material joints are lapped by other material. The cell also incorporates a novel bladder system for compacting and debulking each fabric ply. AFP and ATL machines cannot effectively or efficiently manufacture this part. The part centric approach recognizes the need to harmonize machine, material, method and operator to increase the rate at which this part can be manufactured.
“T” and “Z”-shaped stringers
Some aircrafts utilize a T, Z or C-shaped stringer which is placed on the surface of the fuselage and co-cured in an autoclave. The current technology consists incutting plies on an ultrasonic table, assembling them into a flat laminate and forming them over a tool. In the case of the T-shaped stringers, a “radius filler” or “noodle” must be manufactured on a separate machine and inserted into the void created at the base of the T. All of this is done by hand (except the ultrasonic cutting) and is both time consuming and expensive. It also impacts the ability to increase rate.
Figure 9 shows a work cell that combines pick and place, hot drape forming, debulking and the manufacture and insertion of a noodle. Most of the previous activities done by hand are replaced by automation. Rate is significantly increased as compared with the manual process and each part has consistent, predictable quality.