Designing a Carbon Fibre VAWT Model for All Energy
Part 3: The Manufacturing Process
It was my daughters 5th birthday this week and one of her presents was a plaster of paris moulding kit. You mix it up and pour it into the mould, wait a while and crack it open and you have, in this case, a perfect model of Dora the Explorer! This basic process of using a mould to define the required shape is the basis of manufacturing with carbon fibre .
Welcome to Part 3 of our series on designing a quarter scale model of the Slipstream S20 VAWT for the All Energy Show. In this post we will take you through the carbon fibre manufacturing process and take a look at some parts that are hot from the autoclave this week.
With the deadline fast approaching, progress on the turbine has accelerated this week with all the individual carbon and metal parts having been completed. The main thrust in the coming week will be on assembly of the model ready for shipping on the 11th May. The Slipstream team have been working on site at ACG to fit the first blade to the struts and central shaft.
Let’s talk about how we made the hollow blades of the quarter scale model as a way of describing the manufacturing process. The diagram below gives a simplified overview of the processes involved.
The basic process is to create a positive pattern of the shape to be made out of a resin based block, make a negative carbon mould off this and then use the mould to lay up the pre-preg. The pre-preg is then forced into the mould under pressure and heat to complete the process.
CAD & CAM
The required shape of the blades were drawn up as a 3D model in CAD (Solidworks in this case) as part of the design process. The blades were then split along the length of the blades to make two positive “patterns”. CAM (Computer Aided Machining) software is then used to define the route and position that the cutting tool will take when making the patterns.
The purpose of the pattern is to make a negative carbon mould into which the carbon fibre pre-preg will be laid up. The pattern is a positive of one side of the final blade, that is, it’s surface geometry is exactly the same as the desired component. The patterns for the blades were machined from epoxy resin tooling block. We used ACG TB750 because of it’s fine surface structure. The pattern was machined using a 5 axis CNC machine, using epoxy tooling block makes for fast and easy machining.
Once the patterns have been machined, the surface is sealed and rubbed back to a high gloss surface finish. As well as providing a smooth finish, this final coat is also a release agent that prevents the mould from sticking to the pattern. Typical achievable tolerances on the pattern to CAD are +/-0.25mm.
Here’s a picture of the pattern for the quarter scale model, note the run off areas each side of the blade surface. These will be used to make the flanges of the mould later in the process. Also, note that the edges of the pattern are vertical, this also helps when making the mould.
We decided to use a carbon tool for the model for a number of reasons, it produces an excellent surface finish on the final part, new moulds (or tools as they are often called) can easily be made from the patterns if required (good when manufacturing in volume) and the mould tool is temperature expansion matched to the component which prevents problems when curing.
The carbon tool for the model was made by laying up an approximately 5 mm thick laminate over the pattern and curing in the autoclave. We’ll go into more detail of this process later in the post.
Here’s a photo of one side of the final mould tool. Note the great finish obtained. This is mainly due to the quality of the finish on the pattern. Also note the run off areas each side of the blade and also the vertical side walls. These are added for practical reasons to make the mould tool stand up on a bench without support, and were made by running the carbon fibre over the edge of the pattern.
Component Layup for a Hollow Blade
When laminating or laying up the pre-preg for the hollow blade we used the technique shown in the diagram below. That is a leading edge return is manufactured and bolted to the inner tool. The layers of pre-preg are layed up into the mould tool and around the return to create an overlap of pre-preg material. The outer tool is layed up flush with the edges of the mould. The two halves are then brought together and the overlap is pressed into the outer pre-preg to create a strong lap joint. This results in a single part coming out of the mould when cured. We will discuss bagging and curing in the next section below.
The laminating process for the blades was carried out by hand. Pre-preg was cut on a computer controlled cutting machine to define the exact shapes required. These were then laid into the mould as described above. Multiple layers are added until the design thickness is reached. A laminating book is produced as part of the design process. The layup book defines the material type, fibre orientation and placement in the mould tool. There is a single page for each ply or layer.
When the pre-preg has been laminated into the mould tool, any air trapped between layers is removed by a process known as de-bulking. This involves placing the tool in a bag and sucking the air out to compress the pre-preg into the mould tool.
All materials used in the process such as vacuum film, breather materials, sealing strips etc.. were supplied by Richmond Aerovac Ltd
For making the hollow blades of the model, the two haves of the laminated mould tools were bolted together with a film tube placed up the middle of the blade (this is left open at each end). Then a bag is placed around the whole of the mould tool but with the tube up the middle still open to the atmosphere.
A breather material is placed between the vac bag and the pre-preg. The breather material is very much like the fleece used as a filler in jackets. It’s purpose is to hold the vac bag off the pre-preg to allow the vacuum to reach all parts of the mould tool, this prevents pockets of air becoming trapped which would lead to poor compression or consolidation of the pre-preg onto the mould tool surface.
The picture below shows the blades bagged up with the vacuum hose fitted. You can see the open end of the tube that runs up the middle of the blades. When a vacuum is applied the pre-preg is sucked into the mould tool. The additional pressure of the autoclave oven acting on the inside and outside of the mould tool “pushes” the carbon fibre material harder into the mould tool surface to give better consolidation.
Once bagged up, the blades were cured in the autoclave. The autoclave is, simply put, a large pressure cooker that applies heat and pressure. The heat causes the resin within the pre-preg to reduce in viscosity and flow around the fibres whilst the pressure forces the plies of pre-preg into the mould tool surface to consolidate the multiple layers together into a single laminate fully wetted with resin.
Here’s a picture of the blade being prepared for curing, thermocouples are being added to monitor temperature during curing. The temperature must be carefully controlled.
Once cured, the blade was released as a single hollow part. There is often some bleeding of resin from the edges, called flash. This is easily cleaned up.
And here’s a nice picture of Liam Maloney (Advanced Composite Engineering’s - Engineering Manager) holding the first blade fitted to the struts and un-coated central shaft.
Ok, that’s it for this time.
Next time, we will be taking a more in depth look at the process of FEA modelling of carbon fibre parts and of course keeping you up to date with progress as things come together on the model.
Written by Mike Roberts : Engineer at Slipstream Design