Design and develop a rapid tooling solution for composite lay-up
I secured a graduate research assistantship at the Boeing Advanced Research Centre (BARC) at UW during my MS course. During my assistantship, I worked on helping Boeing implement new tooling options into a research phase at their facilities. Currently Invar, a nickel-iron metal alloy is used for tooling purposes owing to its high stiffness and strength and low coefficient of thermal expansion. However, Invar has drawbacks that include high costs and extremely long lead times.
Due to the increase in demand for parts and components, it is no longer feasible for Boeing to wait for 6 to 12 months for a completed tool to be available. This project focuses on solving this requirement by finding alternative tooling options.
Project duration
September 2018 - June 2019
Role
Graduate Research Assistant, Design lead and ceramics investigator
Relevant skills
Catia v5, Instron 5944, vacuum bagging process, 5-axis CNC router (Fibreworks), Fusion 360
Team members.
James, Kenrick Chan, Paritosh Raghunand, Rebecca Renfrow, Sierra Bishop, Tayler Hoftell
Understanding the problem
Through this project we sought to develop a rapid tooling solution using new methodologies and materials. The tool would be required to exhibit these properties:
High service temperature of at least 370˚F
High vacuum integrity
A surface finish of at least 63rms
Withstand pressures ranging between 9-100 psi for extended periods of 4-5 hours.
Endure at least 4 cycles of autoclave
The manufacturing time for the new tool should be within 14 days (compared to 8-10 months using Invar)
Solution
The initial phase of the project focused on investigating the different types of materials and tool manufacturing processes. These options were: Castable Ceramics & 3D Printing, Machinable Foams, and Sealants/Primers to be used. The different classifications of materials investigated required different experimentation and methods to come to a conclusion about the suitability of the material in a tooling process.
I began researching potential ceramic materials, obtaining their samples, and carrying out preliminary tests to gauge feasibility of these materials meeting temperature, time, vacuum integrity, and surface finish requirements. I also took upon the task to CAD design the geometry of the tool and guidelines to manufacture the same.
Investigating Ceramics
We began by investigating the manufacturing method to be used to create a tool from Ceramics. Ceramics naturally has a low Coefficient of Thermal Expansion(CTE), low porosity and very high service temperatures of above 2500˚C. Although 3D Printing has made giant leaps in the stock materials, the ceramics that can be 3D printed unfortunately do not have a high enough glass transition temperature to withstand the autoclave cycle.
After careful consideration and multiple brainstorming sessions, I presented a technique to merge the two technologies of using a castable ceramic and an additive manufacturing process to create a negative of the tool. The ceramic would then be cast into the shape of the mold to result in the final tool.
We then began by researching the different ceramics that could be used for this purposes and exhibit the required properties. We were able to find ceramics from Aremco, Cotronics which could be used for this purpose. The properties of these materials have been listed below.
Using Scanning Electron microscope imaging of cross sections of these various samples, we were able to narrow down our choices to Cotronics cer-cast 780 and Aremco’s Ceramcast 900-N based on the number of porosity sites and overall strength.
We determined the thermal expansion of the material by running a thermomechnical analysis (TMA) on the two selected ceramic samples. The tests confirmed the low CTE of the material and could be considered for the tool.
After confirming the low CTE of the ceramics, we moved on to establish the strength capabilities. Using ASTM C1161, we used the 4-point bend test to determine the flexural strength of Cotronics 780. The tested samples were of size 0.24” × 0.3” × 3.5” (6 mm × 8 mm × 90 mm) and were placed in a 4-point bend instron setup. The exterior rollers were 80 mm apart, the interior ones were 1.57 inches (40 mm) apart, and a ramp rate of 9.375 × 10-5 /s was used for the instron.
We went ahead and cast a test sample using the Cotronics 780 using the guidelines provided by the Boeing Team and found significant warping and shrinkage after demolding the same. This can be clearly seen in the images below.
CAD Design
I designed the geometry of the tool to be tested using the guidelines provided by the Boeing Advanced Manufacturing (BAM) team in Auburn structure using CATIA V5. The tools were designed for a skin panel and a hat stringer. The tools also include features which are used to accommodate the vacuum pressure valve and the bagging material.
The models would be 3D printed at the BAM facility housing one of the world’s largest 3D printer. The tool for the hat stringer was 3ft x 1.4ft x 0.6ft while the tool for the skin panel was 2ft x 2ft x 0.62ft.
Reflections
Some of my major Takeaways from my experience at BARC were:
Experience is key, but method still matters! This was a major trait I noticed while working with the engineers from Boeing. Though they have worked on similar designs for decades, the process in which they approach the design is still methodical without skipping over any steps.
The castable ceramic and 3D printed mold process is promising as a result of combining the properties of both technologies. The strengths of ceramic are that it has a low CTE, high glass transition temperature and high density for stable vacuum integrity. It can also have a very fine grain size which allows for a smooth surface finish.
Although ceramic is a brittle material with risk of fracture in flexure, there are many high strength ceramic options which will combat this, including those from Cotronics, Aremco and Geopolymer. The observed shrinkage and warping can also be further understood and controlled.