Axial flux pmg construction


Below is the story of the other machine which did reach fruition.  We start with the wheel hub, which came from a Chevy Cavalier (not a Vauxhall Cavalier but similar in size).  The alternator is loosely based on the axial flux 'pmg' (permanent magnet generator) which I designed for use in the developing world.  Brakedrums and laminations are not easy to find in some countries (in fact some people have problems finding them at all).

The magnets are glued to two steel disks so that they face each other.  See the disks being drilled (5 holes at 1/2") to fit the hub below right.  More details of the magnets later.


The stator for this alternator has no laminations - it is just a disk of coils which sits between the two magnet rotor disks.  Here (left) is the mould for the stator (simplified from the one described in my pmgbooklet).  On the right you can see the coil winding former, which is split into two parts for easy removal of the coils after winding.  The coils are circular to suit the circular magnets which we chose in this case.

Nora Woofenden helped out with the coil winding.

The coils were cast in polyester, watched closely by Michael McGuiness (a local man, in the brown hat) who is also building one of this type.  Michael was our 'scrounger' whose task was to find things in a big hurry when we realised we had forgot to get them before.  Thanks for keeping us supplied Michael!  Here we have Kelly and Brian doing the dirty work, with respirators, filling up the mould with coils and polyester.

Next day the casting was still soft.  We had to heat the mould up, so as to kick the polyester into action, but after a few hours we were able to lever off the top and pry out the completed stator.  Hugh Glass and Chuck Morrison inspect the final product.  Much easier to produce than a laminated stator.  Hugh Glass is a talented stage hand from Utah.

Now that we had the stator we were able to accurately position, and weld three supporting lugs on the yaw frame (see below).  We placed the stator on a magnet rotor disk and centred it.  then we fitted 3 pieces of steel angle to the stator using 1/2 inch 'all-thread' or 'threaded-rod' or 'studding'.  The all-thread allows us to adjust the position of the stator so as to place it accurately centred between magnet rotors with the minimum of clearance.  The right hand picture shows the yaw frame set up on a stub tower, with the tail also attached.  The tail boom is welded to an inclined hinge arrangement designed by Tod Hanley and Jeff Gilbert.

Meanwhile we fitted the magnets to the rotors.  For test purposes they were 'super-glued' to the steel disks, and then later we set them in polyester for extra strength.  In previous designs I have used ferrite magnets, but in this machine we used neodymium ones (NeFB), which are stronger.  They are also more expensive but the price is coming down all the time, so they are becoming a very attractive option.  We used 12 magnets, each 1.5inches in diameter and 3/8inch thick.  This works well with 9 coils for 3 phase output.

We are grateful to Wondermagnet.com for supplying the magnets.  Wondermagnet's sister website Otherpower.com has a very interesting low tech alternator design using wood.  Our design uses steel which makes it more efficient and also more practical for permanent duty.


Here is Kelly casting his magnet rotors in a level mould 
We did a quick test of the alternator - cranking it by hand and measuring the voltage at a low rpm.  It came out exactly as predicted.  Using simple calculations I matched the speed of the alternator to the speed of the 8 foot diameter wind-rotor (propeller) so that it would run at or near its design tip speed ratio of 7.  We were able to adjust the cut-in speed of the alternator to suit the blades by careful grouping of the coil connections.

Below I am showing Bob Peterson a design spreadsheet for matching up the rotational speeds.

return to main page for SEI workshop