No EMF when electromagnet is slightly above or below center of pickup coil with 100% loop area.

[pixelzoom]

Related to #156 ...

On 5/1/24, @arouinfar and I discovered a problem that seems related to the new flux model. In the scenario shown in the screenshot below, the light bulb does not light, and the voltmeter needle does not move. This only happens when the pickup coil has 100% loop area (number of loops doesn't matter) and the electromagnet is slightly above or below the loop's center.

I thought this might be related to the lightsWhenCurrentChangesDirectionProperty, which for the Transformer prevents the light bulb from lighting when the AC power supply changes direction. In Transformer.ts:

        // AC Power Supply is cyclic, and current changes direction. We might 'step over' the zero points in the model,
        // so ensure that the light bulb will go through an 'off' state as the current direction changes. Note that it
        // is still appropriate for the light bulb to light when the current changes direction with the DC Power Supply.
        // See https://github.com/phetsims/faradays-electromagnetic-lab/issues/152.
        lightsWhenCurrentChangesDirectionProperty: new DerivedProperty( [ this.electromagnet.currentSourceProperty ],
          currentSource => currentSource instanceof DCPowerSupply )

Removing this seems to fix the problem for the light bulb. But it does not explain why the voltmeter is affected.

[pixelzoom]

I checked whether recent changes to the flux model in #166 (comment) might have cleared this up, but no -- the problem persists.

[pixelzoom]

With 2 loops and 100% loop area, 10V battery....

When electromagnet.positionProperty and pickupCoil.positionProperty are both at (500,375), flux is ~570816.

Moving electromagnet.positionProperty down 1 unit to (500,376) results in a huge drop in flux to -796 (negative!)

Moving electromagnet.positionProperty up 1 unit to (500,374) results in the same flux of -796.

Running a similar experiment with the AC power supply shows similar behavior.

I don't understand this drop in flux, or why it only occurs with 100% loop area.

[pixelzoom]

Discussing with @arouinfar, I hypothesized that this problem is manifesting because of a relationship between the diameters of the 2 coils.

electromagnet.coil.loopAreaProperty = 7854, loopRadiusProperty = 50.00005846084074, diameter = ~100.

With loop area is 100%, pickupCoil.coil.loopAreaProperty = 70685, loopRadiusProperty = 149.9991143470562, diameter = ~300.

So the pickup coil has ~3x the diameter of the electromagnet coil. When the electromagnet is exactly in the center of the pickup coil, the areas about and below the magnet cancel each other out.

[pixelzoom]

a7a918c also fixes this problem, see description in #166 (comment). I can no longer reproduce using the steps in #164 (comment).

@arouinfar please confirm, close if OK.

[arouinfar]

I'm no longer able to reproduce the bug on main. Closing!