Coil Construction. Details

Coil construction is the most complex and frustrating part of turbine design. I suspect there are engineers out there who could "work out" what the best design would be but for those like me who can not even understand the complex equations required, trial and error is a better allbeit more time consuming and frustrating way to go.

The points to consider when constucting your coils

Always obey Flemings left hand rule

Avoid over engineering the design so that the blades will barely rotate

Try for a design that gives the voltage you need at a low RPM ie

Build a test jig that you can run with a power drill

Be aware that the more watts you generate the more heat you generate so more coils allow you to spread that heat

Be aware that bigger wire size allows for bigger currents but less voltage due to less turns

• This is probably the best picture I found on magnet / coil spacing. It is from http://www.fieldlines.com/ and it is a source of a lot of good information.


• The key is to ensure that as one magnet exits the coil the other enters. This ensures Felmings law is followed and the currents do not "fight" each other. In fact in this way they compliment each other.


• We can also see that with this arrangement the inside and outside diameter of the coil is determined by the gap between the magnet.


• This is the determining factor on how many turns can used. Thicker wire = less turns and more current capacity, thinner wire = more turns and more voltage but less current carring capacity.

Another good site is http://www.reuk.co.uk/Wind-Turbine-Alternator-Basics.htm

It shows how the magnets interact with the coils to provide the best output.

Now the problem comes with calculating the placement of the magnets so that they obey all the rules.

The picture on the left is a result of much thought and some time spent with my maths teacher brother.

It is propbably the most crucial information you will need.

After you have determined how many magnet you will use you will need to know where to place them from an angular perspective.

In my case 16 magnets divided by 360 degrees gives 22.5 degree spacing.

But where to place the magnets relative to the centre of disk is crucial. The further out, the further apart, closer in, closer together.

They have to be placed exactly or your turbine will not run. I can not stress this enough. I have three prototypes that all failed because I did not follow this setup EXACTLY.

This applies if you have round, square or oblong magnets.

Back to the picture. In my case my magnets are 30mm in diameter, the coils are mounted on a bobbin with a hole in the centre of 30mm and an outside diameter of 50mm. This means there is 10mm of coil on each side of the magnet. See picture.

To ensure therefore that one magnet leaves the centre just as the next is entering the coil we need the magnet centres 40mm apart. ( Half magnet diameter of 15mm x 2 for each magnet plus 10mm gap). See picture.

How do we then calculate what the radius (half the diameter) needs to be.

We know the following. Angle between magnets 22.5 degrees, we know the magnets have to be 40mm apart. So using the above picture, we can draw a line down the centre of the triangle, giving us a right angle. The degrees are halved to 11.25 and the gap goes from 40mm to 20mm.

Now all we need to do is calculate the hypotenuse (http://en.wikipedia.org/wiki/Hypotenuse) and that will be our radius. If we call the hypotenuse x, then x=20/Sin 11.25 (trust me)

Go to the Windows calculator and "View" Scientific (yes it has a scientific calculator!) and you should get the answer 102.559

To this we then add the radius of the magent ie 15mm and we get an outside diameter of 117.5mm. Subtract 15mm and get inside diameter

If we then draw this circumference in your metal disk and place your magnets, they will be correctly placed.

Wire size.

Now a bit of work is needed on test jig to find the thickest wire you can use that not only fits between the magnet gap (10mm) but also generates the voltage you need ie 14 Volts, 28 Volts or even 50 Volts at a reasonable rotation speed. Note the extra volts are needed as 12 Volt batteries are in fact 13.8 Volts and you need to charge them to say 14VDC.

Another consideration is wiring of the coils together. There are two ways, star or delta. You can look these up on the net but I used star as it provides higher output at lower RPM. Delta perfoms better at higher RPM. This chart is from http://www.windstuffnow.com/main/ another great sourc of information.

My Setup

This is my test jig. Pretty basic. Bottom rotor, 4mm MDF sheet, top rotor and a bit of wood with a bolt through it to attach a drill.

The wound coils were simply stuck to the MDF with Super Glue. I used 3 multimeters, 1 for current, 1 for volts and a 3rd to measure RPM. This meter had a Hertz setting so I simply tapped off one of the coils. Note: You have to adjust the hertz reading for the number of magnets. With 16 magnets you get 8 pulses pulses per revolution. So in my case I needed to divide the hertz reading by 8 to get the correct RPM.

A jig was also made to roll the coils. This was two circular pieces of 4mm MDF slightly larger than the coils and a bolt through them that sandwiched the coil and connected to the drill.

Counting turns was far too tedious so I counted one, then unwound it and then used that length as my benchmark. I then measured out 5 equal lengths and wound them on. A couple of turns here or there is not going to matter. It was aprroximately 300 turns.

I decided to use just one guage of wire as there were too many variables. From the chart at http://www.powerstream.com/Wire_Size.htm I chose 24 AWG and bought a big industrial roll.

I won't bore you with the many combinations I tried but in the end I settled for 6 coils wound right to the outer edge of the holder/bobbin ( I used the bobbins from the tape you use to seal water pipes as they were the exact dimensions required ie 30mm ID and 10mm of winding). I figure there is about 300 turns.

Not very scientific but on the jig it tested OK. A 3 phase solution was decided upon with two coils per phase, total 6 coils.

What's Next.

Positioning the coils.

The next step is to imbed the coils in resin to protect them from the elements. The crucial part of this operation is that the coils are positioned at the right angles AND with the right diameter so that the coils are directly centered with the centre of the magnets.

I found the best way was to draw accurately on a piece of paper the position of each coil and super glue them in position. This way they are fixed and can not move. It also allows you to wire the centres of each coil while in position. The angle in this case is 360 degrees devided by 5 = 72 degrees apart. The paper is cut in a circle of 250mm diameter. Magnet edge is on a 230mm diameter disk so we add the width of the coil ie 10mm and edge.

Embedding the coils in resin

You will see in the picture a pie dish with the position of the coils clearly marked. Simply stick tape the alignment paper to the bottom of the tray. A pie dish is perfect to hold the resin mould. It is metal so the resin does not stick to it and the flat bottom is removable making it easy to remove the mould. Holes for the coil wires can be easily drilled in the side. (Note: make sure these holes are a tight fit with the wire or the resin will dribble out.) They come in various sizes. 280mm was chosen as the outside diameter needs to wider than the rotors to allow the stainless steel rods to attach to it and hold it in position.

Next another jig needs to be placed in the centre to allow space for the bolts connecting the inner and outer rotors. Back to the pie shop and a circular pie cutter was found to be perfect. The 140mm one was selected but again come in various sizes. To the left is the first layer of clear resin poured 2-3mm thick. The brick keeps a tight enough seal and stops the centre moving.

The resin mix

The resin is made up up of two parts, the resin and the hardener. BE WARNED the resin sets at different rates depending on the temperature and the amount of hardener used. I have many tubs of hardened resin that have set well before it was time to pour them in.

BIG TIP: Have everything ready prior to mixing the resin.

Two resin mix's were used, one clear or straight resin the other with talcum powder mixed in. The talc is used for two reasons. 1. It allows the more expensive resin to go further and 2. it helps dissipate the heat build up created by the coils. The clear was used in the first 2-3mm layer to see through it to see where to place the coils.

TIP: The resin / talc mix is 50/50 in weight. Mix these two together THEN add the hardener.

Here we can see the coils in place and the second layer of resin/talc mix added to cover the top of the coils by 3-4mm. The coil wires can also be seen coming out the sides. Do not add more than is necessary to cover the coils as the thicker the coil assembly the further the coils are from the magnets and believe me, every mm counts. Do not worry too much about the bubbles. These can be sanded out with a sander when the resin is set. Note: if the resin sets too fast the bubbles may not have time to rise to the surface.

BIG TIP: Err on adding less hardener than more. In fact I would suggest half the recomended amount especially on hot days.

The removed cast. You can see it is messy and still has the alignment paper attached to the bottom. Simply sand away both top and bottom, being sure not to expose the coils. Note: they look closer to the surface than they are due to the clear resin at the bottom. Once it is out of the pie dish, place it between two flat surfaces larger than the cast and apply a lot of pressure ie several bricks to ensure the cast sets flat. Leave overnight.

VERY IMPORTANT: See big tip below.

BIG TIP: Remove the cast before it has hardened completely. If allowed to completely set in the pie dish it will buckle. It expands as it hardens and climbs up the edges of the pie dish. This is a little difficult to judge. Once rubbery, continue to test until it can be removed without cracking. Twist the outer edges of the pie dish to break the seal and assist by pushing the wires through.

Here we can see the sanded down version of the coil assembly. Some cracks did appear but they were filled with Araldite and sanded back. This picture also shows the coil assembly connected the base (no hub or rotors) with the stainless steel rods. Stainless was used as it is non magnetic. Magnetic bolts were initially used but the rotating magnets were attracted to them causing a ratcheting effect ie preventing it from rotating. Alignment is crucial so that when the hub and rotors are installed, the magnets must past directly over the centre of the coils.