I have a permanent magnet generator. AKA alternator. Seven magnets on a rotor that are turning in close proximity to three coils.

Output of the coils is AC. Then converted to DC via a rectifier. Then the DC voltage is controlled by a shunt regulator.

For a constant rotor speed, I observe that the AC output current is a function of the load that is applied. That is, the electrical power being created is dependent upon the load. No load, no power. Bit of a load, more power. More load, more power.

Can anyone tell me in layman's terms why this is happening? I would have expected the maximum electrical power to have been created all of the time.

 

Power is a product of I x A (W) , no current, no power.

 

You will find that many things about electricity conflicts with logical assumptions, or are counterintuitive

I once read a post on this site where the poster didn't understand why turning the crank of a generator didn't get easier when you let more current flow.

 

To expand a bit on @MaxHeadRoom comment...

Current is the flow, so to see current you will have to have electricity flowing. Voltage, on the other hand is the "pressure" so it will be highest when there is no load and reduced as more current is flowing.

In the ultimately terrible water analogy for this, imagine if you had a small pipe and measured the flow from it, then while keeping the pressure more or less constant you switched to a larger pipe and measured. What would you expect?

Of course, if the load is small (it's resistance is high) current will be limited. If the load is large (resistance is low) current is free to increase to the limit of the source.

 

The voltage without any load is the highest it will be without changing the speed of rotation of the alternator. Once you connect a load, current will flow and will depend on the load. A low impedance load will draw more current, a high impedance load will draw less. Either way the power generated will equal the power used in the load.

In a perfect system there will be no losses, but in the real world, the wire in the alternator coils has resistance and as current flows, those coils will heat up. This is power lost and it means that my first paragraph has some caveats. The alternator is not perfect and there will be some amount of power lost, in bearing friction, magnetic saturation and wire resistance.

Lastly, as you add more load, the actual voltage at the output of the alternator will drop due to the losses mentioned above.

 

There is a point at which maximum power will be transferred to the load. We start with an open circuit. No power and decrease the resistance, current will increase and voltage will drop for a given turn rate of the alternator when the product of current and voltage reaches a maximum you have your best energy transfer. Decreasing the resistance beyond this point causes the voltage to drop faster than the current increases. So it follows a bell curve. It is similar to a voltage with a given output impedance say R. When we apply a load say RL the output voltage is given by V*RL/(R+RL) and the current is given by V/(R+RL). The power is the product of the two or P=V^2 * RL/(R+RL)^2. V is a constant and thus V^2 is a constant. Now we can see if RL=open that is extremely large compared to R this becomes about RL/RL^2 or just 1/RL and close to zero. If RL is extremely small compared to R then this becomes RL/R which again is close to zero. The power curve will be sloping upward at the start and sloping downward at the end. When the power is maximum the slope of the power curve will be zero. So if you take the derivative of the function supplied with respect to RL and set that derivative to zero and solve for RL you will find it happens to be when RL=R you get maximum power transfer. I am not going to go through all the math but if you do this you will find max power is when RL=R. The only difference is that the alternator at a given speed has a series resistance that happens to be a complex impedance that involve a resistive part and a reactive part.

 

Voltage is being generated independent of the load. As the load draws current, power is delivered to the load. No power is generated when no current flows. Power is defined as voltage times current, and so when the current is zero no power is delivered.

 

There is a point at which maximum power will be transferred to the load. We start with an open circuit. No power and decrease the resistance, current will increase and voltage will drop for a given turn rate of the alternator when the product of current and voltage reaches a maximum you have your best energy transfer. Decreasing the resistance beyond this point causes the voltage to drop faster than the current increases. So it follows a bell curve. It is similar to a voltage with a given output impedance say R. When we apply a load say RL the output voltage is given by V*RL/(R+RL) and the current is given by V/(R+RL). The power is the product of the two or P=V^2 * RL/(R+RL)^2. V is a constant and thus V^2 is a constant. Now we can see if RL=open that is extremely large compared to R this becomes about RL/RL^2 or just 1/RL and close to zero. If RL is extremely small compared to R then this becomes RL/R which again is close to zero. The power curve will be sloping upward at the start and sloping downward at the end. When the power is maximum the slope of the power curve will be zero. So if you take the derivative of the function supplied with respect to RL and set that derivative to zero and solve for RL you will find it happens to be when RL=R you get maximum power transfer. I am not going to go through all the math but if you do this you will find max power is when RL=R. The only difference is that the alternator at a given speed has a series resistance that happens to be a complex impedance that involve a resistive part and a reactive part.

True but most electrical generation systems are designed for maximum efficiency not maximum power transfer. You design the source impedance to be as close to zero as possible.

https://info.triadmagnetics.com/blog/maximum-power-theorem