Simple generator

[broli]

Well here's a video of connecting a dc source to the coil and inspecting the rotor interaction.

https://www.youtube.com/watch?v=nPTEpe6wRNA

I also did a simulation by applying a DC current and calculating the torque all the way around. Strangely enough it seems the torque is almost constant untill the magnets cross the coil edges and then the torque flips and remains constant again. Again this is not what conventional generator does, usually you see a nice sinusoidal torque graph.

[gyulasun]

Thanks for this test. It suggests thinking of some core saturation from the input 6.3 A current
which surely biases the core and probably makes the ring core magnetically asymmetric.
This is the only explanation I can think of why the rotor magnet poles align with the coil ends:
those are the places where the electromagnet poles are created.
It comes from this that at lower input current levels, say under 1 Amper or even less, the interaction
gradually disappears I think because the core cannot become magnetically asymmetric and the
magnetic poles are able to close into each other within the ring core as in any 'normal' closed
magnetic circuits without an air gap.

My observation from the video is that the 'rotor' would always want to align with the coil edges in attraction.
Should you have a ball bearing with even less friction, I think the torque would 'behave' in a more 'sinusoidal' way.
When the rotor is turned to the 12 o'clock and the 6 o'clock direction, it is magnetically the farthest position
from the coil edges and the attraction forces at the coil edges are simply not enough to influence rotation.
Have you thought of replacing the rotor with a compass and see the effect when input current is say 0.1 Amper only?
At such lower level core excitations the saturation should get to a minimum hence the asymmetry
cannot yet develop as much. With the compass 'needle' the sensitivity of your "rotor" may get increased.
Then you could increase input current too and see the behaviour.
Sorry that I dare to assume the bearing is what mainly causes the bitty rotor movement,
especially in the first half of the video,  though it works more readily in the second half of the video.

[lumen]

I wonder how it would respond with the DC current and only a single magnet on the bar.
Suspecting that the magnets could be acting on the field outside the winding and not the field in the core.

[broli]

gyulasun, very good point. I haven't considered saturation. I did a quick check in FEMM with sillicon steel and indeed it seems 5A is saturating the core. It's only when I go below 1A the magnetic field strength in the core goes below saturation. Attached graph shows the field strength in the core at different currents in FEMM.

Here's a repeat of the experiment with 0.5A of current instead of 5A:
https://www.youtube.com/watch?v=lnU54W_ikkY

And repeating the experiment at this low currents indeed reduces the torque considerably as you can see, but the question is whether this just a linear relation ie. 10x less current = 10x less torque?


However it might be my imagination but now the torque also seems to be only concentrated near the edges of the coil whereas previously it seemed to be uniformly around the core.

I'm currently doing a torque simulation with much higher elements count to get rid of the jitter I have been getting in the data. The current torque data indicates a non zero AVERAGE torque all the way around...
So I upped the accuracy by decreasing the element size in FEMM and I'm collecting data per degree instead of 5 degrees. The initial few data points show a much cleaner data distribution, there's hardly any jitter now, but it will take many hours before the complete run finishes.

[broli]

Well the more accurate simulation is complete and as I thought the anomaly all but disappeared, the average is near zero now and the graph is almost jitter free.

However this brings us to the next subject, the torque is almost CONSTANT until it reaches the edges and then it does an instant flip....it's the exact opposite of what I thought would happen. A single pole motor/generator usually has a sinusoidal curve for it's torque vs angle graph. I added the counter torque graph based on the current torque graph and previous induced current graph.

Now let's go a step further and do some power calculation, the average counter torque seems to be 0,00165Nm at 1A output. If we want to keep the rotor spinning at 10000 RPM and extract 1A from the coil it will cost us 1.727W of mechanical power. This seems rather low .