ANSYS Maxwell: Double Rotor Axial Flux Motor Using Symmetry

Hello everyone, David Giglio here with Ozen Engineering. In this video, I show you how to model an axial flux motor using the magnetic transient solver in ANSYS Maxwell.

Let's first take a look at the excitation circuit where DC voltage is input to a three-phase inverter, which produces three-phase square wave voltages.

Switches 1 and 4 are for phase A, switch pair 3 and 6 are for phase B, switch pair 5 and 2 are for phase C, and no switches in a pair of a phase are open and closed at the same time.

The sequence of the closing is such as switches 1 and 6 are closed, then 1 and 2, then 3 and 2, then 3 and 4, then 5 and 4, 5 and 6, then 6 and 7, 8, 9, 10, 11, 12, and 12 and 13, 14, 15, 16, 17, 18, then it repeats 1 and 6, and so on. The cycle continues.

Here is a plot of the control voltages for the pulses. We can use time-based control pulses or position control for the pulses. For position control for the pulses, the switches open and close based on the relative position of the magnets relative to the coils.

Let's take a look at the simulation results. At this instant, switches 1 and 6 are closed. Phase A has a maximum positive current, and phase B has the maximum negative current. We see that phase B has a maximum negative current.

The phase A coil current is clockwise and produces a downward field, which attracts the magnet shown in blue, which has a downward field.

These fields are attractive and will induce a torque on the blue magnet to rotate in the positive Z direction, where positive torque is defined as torque pointing up, and the rotation is aligned with the curl of the other fingers.

The phase A coil will repel the magnet shown in red, which has its field pointing upwards. So, the force direction is the same as the force direction for both torques.

Similarly, for this phase B coil, the base current shown here produces a downward field using the right-hand rules for electromagnetic fields. This downward field from the coil will repel the magnet shown in red, which has its fields pointing upwards.

So, the phase B coil will repel the red magnet to rotate in the positive Z direction and will attract the magnet shown in blue, which has this field pointing downwards. The downward field of the blue magnet will try to align with the downward field from the phase B coil.

So, this is true for all the magnetic magnets all around the motor. All the magnets either experience an attractive or repulsive force, and all of these forces work together to produce positive torque in the motor. Let's look at the one deep plot for the torque.

We see that there is a transient for the torque. Initial torque is zero, and then it ramps up and has a ripple at steady state.

The reason why it has ripple is because the windings have impedance, so even though voltage is applied to the windings, maximum current will not be obtained at that instant where the switches close and voltage is applied to the wire. It takes some time for the winding current to rise and then fall.

So, that's why we have this ripple in the torque. Also, torque varies with the position of the stator coils and the magnetic magnets. When a coil has a maximum current and it's in between magnets, that is where the maximum torque is obtained.

Because the torque has ripple, the power output also has ripple. For this design, the output power is 2.2 kilowatts, and the rotor and stator diameters are 120 millimeters. This is a half symmetry model, meaning that there is symmetry with respect to a plane, in this case, the XY plane.

This is actually a double rotor, XO flux motor, where the full motor will have an equal rotor mirrored with respect to the XY plane. Because symmetry is applied, we're only modeling half of the model, and we still obtain the full results in less time.

We also have the rotational symmetry using independent boundary conditions. We have the half symmetry from the XY plane mirror symmetry and another half symmetry for the 180-degree rotational symmetry. Half and half is a quarter symmetry.

We saw in the half symmetry model, it took 28 minutes and 32 seconds. Here, for the quarter symmetry model, it took 17 minutes and 37 seconds. So, we can imagine for a full model, it would take longer. However, for a larger motor, it'll take approximately double the time.

This is a 3D model FEA results you could obtain results in a reasonable time.

You could run multiple simulations and set up a parametric sweep for varying geometry, excitation frequency, the width of the magnets and the thickness of the magnets, the rotor stator diameters, and the number of turns in the coils.

You could vary whatever parameter you want, run a parametric sweep, and have multiple solutions for giving them for your design. Essentially, we can finally seriously make various operations, imagine a wind farm, and produce for example around the diameter of the x-axis a Shell for sections.

This is the guy's this is a stick, these special rem letter tubes have happened. You could make a different FEA model for in this case, a double-sided rotor, and then we could choose the circuit settings.

In Maxwell, we can use CAD operations on the top to modify the geometry, and we can use the Maxwell circuit to edit this excitation circuit however we desire. That is all for this video.

Contact us to learn about our simulation capability and request a demonstration for us to show you how we can help you with your engineering projects. We provide training to use ANSYS tools and offer consulting services.

If you like this video, give it a like, subscribe to our Ozen YouTube channel, and thank you very much. Have a nice day.