Will an AC electromagnet respond to harmonics?

[neazoi]

Will an AC electromagnet respond to harmonics?
In what sense?
Each harmonic of the AC will add to the magnetic field strength?
The electromagnet coil itself can act as a filter so that some higher order harmonics of the AC will be attenuated?

 

[Borber]

Electromagnet respond to current that flows through coil. AC current depends on voltage applied and impedance of coil. Coil impedance has fixed ohmic component and series reactive inductive component which is proportional to frequency of applied AC voltage.
Magnetic force produced by coil current is proportional to current absolute value (like full wave rectified AC) . Fundamental harmonic and higher harmonics act together.

 

[neazoi]

Electromagnet respond to current that flows through coil. AC current depends on voltage applied and impedance of coil. Coil impedance has fixed ohmic component and series reactive inductive component which is proportional to frequency of applied AC voltage.
Magnetic force produced by coil current is proportional to current absolute value (like full wave rectified AC) . Fundamental harmonic and higher harmonics act together.

Since these act and affect the same magnetic field, then is the next thought true?

Two electromagnets placed near each other (and electromagnetically shielded) so the magnetic field of the first induces a voltage at the output of the second electromagnet coil (magnetic coupling) could reassemble a bandpass filter that rejects any harmonics of the original AC.

Ant this I think applies to transformers, however the output of a mains transformer does contain harmonics of 50Hz as well I think...

 

[FvM]

I'm missing any reasoning for the claimed "bandpass effect".

Two closely coupled inductors are acting electrically like one (if you disregard the voltage isolation). Means the secondary (AC) voltage of a 1:1 transformer is identical to the primary. For less than perfect coupling, the voltages become different. But partially coupled indutcors do in no way reject harmonics. You'll add capacitors, forming a resonant LC circuit to filter harmonics.

 

[crutschow]

It's true an inductor will filter the harmonics of a waveform but, as RvM noted, a transformer does not act like an inductor to the signal passing through.
It only acts like an inductor to the primary magnetizing current, not the signal current from primary to secondary.

It's a common error to think that the transformer magnetizing inductance affects the signal passing through, but it doesn't. The signal sees no inductance (except for any stray and undesired leakage inductance).
Thus an ideal transformer is transparent to the signal.

 

[neazoi]

It's true an inductor will filter the harmonics of a waveform but, as RvM noted, a transformer does not act like an inductor to the signal passing through.
It only acts like an inductor to the primary magnetizing current, not the signal current from primary to secondary.

It's a common error to think that the transformer magnetizing inductance affects the signal passing through, but it doesn't. The signal sees no inductance (except for any stray and undesired leakage inductance).
Thus an ideal transformer is transparent to the signal.

That is why I mentioned "electromagnetically shielded". The additional transformer question was just to see if this applies to unshielded transformers as well.
I am asking for two identical electromagnets electrostatically shielded from each other, so no other component can pass from the first to the second apart from the magnetic field.
Will this configuration act like a short of BPF?

I am thinking it that way:
The two electromagnets are electrostatically shielded from each other with a non-ferrous material (eg aluminium) so that only the magnetic field can pass through.
The first electromagnet coil accepts mains 50Hz plus the mains harmonics.
All these components combine together to create the alternating magnetic field in the first electromagnet.
The second electromagnet (ideally closely spaced to the first) sees this alternating magnetic field and generates an AC to it's windings.
This AC is purified from harmonics, although there are losses in the process.

My point is, if the frequency of the harmonics in the first electromagnet will affect the frequency of it's alternating magnetic field, or if the higher frequency (but lower power) magnetic fields generated by the harmonics, will be dominated by the magnetic field produced by the fundamental frequency (net magnetic field).

 

[volker@muehlhaus]

I am asking for two identical electromagnets electrostatically shielded from each other, so no other component can pass from the first to the second apart from the magnetic field.

What you call "electromagnet" is the same thing as an inductor (with core). And the two magnetically coupled electromagnets are a transformer. No more, no less.

Will this configuration act like a short of BPF?

As much as any other inductor/transformer: by the frequency dependent impedance Z = jwL. Depending on the core material, there are some frequency dependent losses also.

I am thinking it that way:
The two electromagnets are electrostatically shielded from each other with a non-ferrous material (eg aluminium) so that only the magnetic field can pass through.
The first electromagnet coil accepts mains 50Hz plus the mains harmonics.
All these components combine together to create the alternating magnetic field in the first electromagnet.
The second electromagnet (ideally closely spaced to the first) sees this alternating magnetic field and generates an AC to it's windings.
This AC is purified from harmonics, although there are losses in the process.

For a core that is not saturated: The current is the second coil is proportional to the current in the first coil. Same harmonic content, as with any transformer. This does not change when you call the transformer "electromagnet".

My point is, if the frequency of the harmonics in the first electromagnet will affect the frequency of it's alternating magnetic field, or if the higher frequency (but lower power) magnetic fields generated by the harmonics, will be dominated by the magnetic field produced by the fundamental frequency (net magnetic field).

If you drive the core into saturation, things change indeed. But that does not mean you will get a "sine" output from input with harmonics. The saturated core might also distort the waveform and add harmonics that did not exist in the input waveform.

 

[neazoi]

For a core that is not saturated: The current is the second coil is proportional to the current in the first coil. Same harmonic content, as with any transformer. This does not change when you call the transformer "electromagnet".

Is that also the case even if the primary and the secondary are electrostatically shielded? (only net magnetic component pass through)
This is not the case with an ordinary transformer.

 

[FvM]

Is that also the case even if the primary and the secondary are electrostatically shielded? (only net magnetic component pass through)
This is not the case with an ordinary transformer.

All previous contributions assumed a kind of ideal transformer with only magnetic coupling, ignoring the inter-windings capacitance of the "ordinary" transformer. It can be essentially modelled by two inductors representing main and leakage inductance. As said, all harmonics will be transmitted by the magnetic coupling.

If you connect a resistive load to the transformer output, it forms a first order low-pass with the leakage inductance which slightly reduces the magnitude of higher order harmonics. For a usual transformer with some percent short circuit impedance X%, the effect will be rather weak.

You don't need a transformer to see the leakage inductance effect, a simple series choke will do.

 

[volker@muehlhaus]

Is that also the case even if the primary and the secondary are electrostatically shielded? (only net magnetic component pass through)
This is not the case with an ordinary transformer.

The working principle of a transformer is based on magnetic coupling.