Building the nacelle body for our small wind turbine out of glass fiber and epoxy was a hard piece of work. Especially our lack of experience made it difficult to conclude the process in a timely manner. However, it feels great to have learned something and we are excited to share our experiences with you.
Building the mould
We decided to not build the original mould but to start directly with the negative mould. We decided on Styrodur which is similar to Styropor but it is a hard foam so it does not crumb and has a good structure to work with. We used a simple carpet cutter to work on the material. We only had a CAD draft in 1:1 to get the measurements from. We drew the different diameters on both sides of the 2 cm Styrodurplate, always taking an extra 2 mm for glue and plywood which would come into the mould later. First we cut out the smaller diameter and tried to even the radius to the bigger one througout the wall thikness. We needed a stabilizing wall every 8 cm otherwise the plywood would have too much space to bend. For glueing we used different types of glue.
In the end, the best glue was a simple handicraft glue you get in every supermarket.
Between the steps the glue had to dry for several ours, we used everything we found to put some preassure on the mould to keep it in form. Before we glued the plywood inside we put both moulds together to see if they fit. There was a mistake in the middle of one mould so we had to rebuild one plate. In this stage it was easy to fix, later would have been a lot harder.
Butting in plywood was an idea from the “Tretboot AG”. Very thin plywood can be bent extremely and has a very smooth surface. We had to divide it into four parts because of the changing bending at the end. It was hard to get the parts right, we had to calculate the first cone surface and try to make the other parts fit. As you can see in the picture on the left the edges were not quite right so we had to fill the edges later with epoxy which was a lot of extra work. The better way would have been to go to the CAD Laboratory of the Flensburg University of Applied Sciences and ask for a wind-up of the CAD module to prepare the mould parts more easily. Between the gaps of the plywood and on top of the soft edges of the styrodur we put some 2-Component-Epoxy-Putty. This is resistant to the further layers and we are able to grind it to a smooth surface.
The Putty has to dry for 24 hours before it is completely hard and in the end it was harder than we thought to grind it. Apparently there are two different types of Epoxy-Putty. One is a very strong, water proof one (that is the one we had) and there is another one which is a lot easier to work with. (So next time, get the other one!)
Anyway after a lot of sanding the mould looked pretty good. At the beginning we thought about building the lashes with the form but we decided that it is way too difficult to laminate into the mould with the lashes so that is why we detached them again.
I guess the styrodur powder was not the most healthy thing either but since we started to use the epoxy products we tried to wear gas masks and gloves most of the time.
I cannot believe how scared I was about making a mistake in the barrier layer. It was my biggest nightmare that we would laminate into the mould and at the end it would not come off again and we would have to destroy the mould and start all over again. A friend told me to put epoxy resin once all over the mould. We were short on time and thought we might be able to skip this step which includes another 24 hours of drying so we started to built up the barrier layer on top of the plywood. You start the barrier layer with a few layers of separating wax. For a new form you usually use 4-5 layers instead of one until you get a pore-free surface. The wax was nothing like we expected. It is very liquid, most of it dissolves in the air and only a really thin layer of wax stays on the mould. After putting on 5 layers the surface was still very porous. We put on some more layers of the wax and later several layers of the release agent. The wood seemed to soak it all up. It did not look like a pore-free surface at all when we finaly started with the gelcoat. The gelcoat was a two component one as well. We put on one layer with a lacquerroller and a foam brush. After 2 hours we put on another layer. After we finished the second layer I decided that I did not want to go on laminating into the mould. I was too scared the form would not come off the mould so I decided to let the gelcoat dry and restart with a non-porous surface again. In the end the gelcoat came easier off the mould than the old gelcoat of the new gelcoat. But you never know and we only had one shot.
Note for next time: Put on a layer of epoxy resin into the mould first and you will not have any problems. Also it is very good to cover.
All parts of the styrodur mould you want to work with epoxy. As we saw, the wax is very aggressive. Parts of the styrodur just disappeared (luckily they were all not part of the main mould).
Laminating the Nacelle
Who would have thought this? So much work before we could finally laminate. After leaving the second layer of gelcoat another 2 hours to dry and before we started with a thick layer of epoxy resin we put some epoxy resin mixed with cotton flox (like a self-made putty) into the corners to not get over the maximum bending radius of the glassfibre matts. Next step is to put in a layer of fibre glass. We decided on a 150 g/m² glassfibre webbing as a first layer. After every layer you put on more epoxy resin and soak the whole fibre with it before putting on the next layer. Also you have to make sure not to have too many air bubbles between the layers. Our second layer was a 300g/m² fibre glass matt. The fibre glass matt was horrible to work with, it was very stiff so it was impossible to get it in without bubbles. Also it needed a lot of resin to soak probably. We asked several people whether we should leave the matt in one or cut it into stripes. All said it would be fine to leave it as one part. In the end I would definitely cut the second layer in small stripes. It is a lot easier to work with and for a form like this it does not really affect the stability of the component. Furthermore I would use more thin layers instead of the big matt. Another critical point was the hardening time of the resin. Doing both moulds at the same time without an extra hand is not the smartest thing to to.
Take your time, get someone to help you and buy a resin with a long hardening time. We had 90 minutes and that was really absolutely the minimum.
After 24 hours of drying the big moment was finally there… getting the nacelle out of its mould. With a few hard strokes from a hammer on the backside of the mould the component came off the mould. Getting of the old gelcoat was a bit of a pain but hey it came of the mould!!! The soft remains of the fibre glass were cut off with a scissor and the hard bits with a “Dremel”. After evening the edges and putting in some holes only the lashes for the hook and loop fastener had to be attached. We built the lashes out of the plywood (little L’s) and attached it to the body with epoxy putty.
About the author:
I am studying Renewable Energy Tecnology at Flensburg University of Applied Sciences. Wind Energy is a big part of my studies so I was more than happy to participate in this project and do some practical work. My job was to manufacture the aerodynamic body of the turbine.