Exciting Current harmonics in Transformer

[Carl Pugh]

sophiecentaur, a transformer designer should know how a transformer operates. If such a basic question as whether some of the core exciting current flows in the secondary is not known, then the designer does not have a complete understanding of how a transformer operates. It is true however that rarely does it make a difference.

cabraham, your model for calculating the exciting current is incorrect.
One correct model is an inductor in parallel with a resistor in parallel with a perfect transformer primary.

Then a capacitor should be placed in parallel with the perfect transformer secondary.
If the perfect transformer has a 1 to 1 ratio, the perfect transformer can be removed and the capacitor will be in parallel with the inductor and resistor.

If the reactance of the capacitor and inductor are equal, then the exciting current only includes that of the resistor. The inductive part of the exciting current flows in the perfect transformer and then in the capacitor.

In the preceding discussion, all components are perfect and there are no harmonics.

YES I KNOW ABOVE IS NOT CLEAR, BUT IT IS THE BEST I CAN DO.


It would be easy to design a transformer with a core gap where the primary exciting current would decrease over 25% when a capacitor is connected to the secondary.
Transformer would be a small standard transformer using E-I laminations, except the core would have a non-magnetic gap. Harmonics and core saturation would be avoided as much as possible.
If such a transformer was built and the exciting current decreased over 25% when a capacitor was connected across the secondary, would you agree that some of the transformer magnetizing current flows in the secondary?

 

[sophiecentaur]

I agree that a designer should know as much as possible but what is actually possible? I have only a limited knowledge of transformers beyond fairly basic equivalent circuits. I do, however, have some reasonable knowledge of antenna theory. I know enough of that to be aware that you can only go so far in describing precisely what goes on - even for a simple dipole.
They are both 'classical' systems but they don't have complete solutions afaiaa. Do "what's really happening" is a rather naive question. 'Near enough' has to be good enough and, amongst other quantities, Magnetising Current is only something introduced in order to make predictions. Why demand anything more of it? It's useful enough as it is.

 

[cabraham]

sophiecentaur, a transformer designer should know how a transformer operates. If such a basic question as whether some of the core exciting current flows in the secondary is not known, then the designer does not have a complete understanding of how a transformer operates. It is true however that rarely does it make a difference.

cabraham, your model for calculating the exciting current is incorrect.
One correct model is an inductor in parallel with a resistor in parallel with a perfect transformer primary.Then a capacitor should be placed in parallel with the perfect transformer secondary.
If the perfect transformer has a 1 to 1 ratio, the perfect transformer can be removed and the capacitor will be in parallel with the inductor and resistor.

If the reactance of the capacitor and inductor are equal, then the exciting current only includes that of the resistor. The inductive part of the exciting current flows in the perfect transformer and then in the capacitor.

In the preceding discussion, all components are perfect and there are no harmonics.

YES I KNOW ABOVE IS NOT CLEAR, BUT IT IS THE BEST I CAN DO.


It would be easy to design a transformer with a core gap where the primary exciting current would decrease over 25% when a capacitor is connected to the secondary.
Transformer would be a small standard transformer using E-I laminations, except the core would have a non-magnetic gap. Harmonics and core saturation would be avoided as much as possible.
If such a transformer was built and the exciting current decreased over 25% when a capacitor was connected across the secondary, would you agree that some of the transformer magnetizing current flows in the secondary?

Carl, you tell me that my exciting current is incorrect, then present a simplified equivalent circuit using all linear elements, R, L, & C. These linear elements cannot describe why harmonics are present. Harmonics occur because of the non-linear nature of the core material B-H curve. Your equivalent RLC circuit is a good starting point for a beginner learning xfmrs for the 1st time. But to understand why the harmonics exist, & why they can be found in primary or secondary windings requires an advanced understanding of the subject.

I present a rather advanced & detailed view of harmonics due to non-linearity for all to benefit from, & you rebuke me with a simple amateur level RLC equivalent circuit. Do you fully understand the info I posted in great length? If not, that's ok, but are you at the level where you can tell me I'm wrong?

Your RLC is really sophomoric claptrap. There are no harmonics w/ RLC model. This is the problem w/ forums like this. A few people w/ extensive knowledge on a subject are being rebuked by people who have a very crude beginner understanding of the material. If I erred, I'll accept correction. If my model omits details, please add them. But you take my non-linear model, & replace it w/ the most crude primitive model useful only to a beginner.

Not to be rude, but you just aren't what you think you are. Are you a designer of magnetics? How much experience have you developing power systems & magnetics? How much formal education, including machines & power, do you have? I'm not being confrontational, but when somebody tells me that I'm wrong, I expect them to back up what they say, & you haven't. You don't know enough to challenge me. Nothing personal, but unless you can address where I went wrong, invoking oversimplified circuit models is not going to get us anywhere. Again, it's not personal. I would advise you & everyone to be very careful about telling someone they are wrong, unless you are positive, & can cite credible references.

Claude

 

[Carl Pugh]

Well I have a BSEE with 25 years of transformer design experience. Most of the design experience was in shops that would accept specs from any customer with money.

Harmonics don't have anything to do with the imaginary part of the core exciting current flowing in the secondary.

Do you agree with following analysis?

With capacitor not connected, see attachment
Source current=1.0 amp resistive + 1.0 amp inductive=1.414 amp RMS
Secondary current=0
With capacitor connected to transformer secondary
Source current=1.0 amp resistive + 1.0 amp inductive + 1.0 amp capacitive
The inductive currents cancel out, so the
Source current=1.0 amp resistive.
The capacitive current in the secondary will be 1.0 amp

This proves the theory that the reactive part of the core exciting current can flow in the transformer secondary.

Carl

 

[cabraham]

Well I have a BSEE with 25 years of transformer design experience. Most of the design experience was in shops that would accept specs from any customer with money.

Harmonics don't have anything to do with the imaginary part of the core exciting current flowing in the secondary.

Do you agree with following analysis?

With capacitor not connected, see attachment
Source current=1.0 amp resistive + 1.0 amp inductive=1.414 amp RMS
Secondary current=0
With capacitor connected to transformer secondary
Source current=1.0 amp resistive + 1.0 amp inductive + 1.0 amp capacitive
The inductive currents cancel out, so the
Source current=1.0 amp resistive.
The capacitive current in the secondary will be 1.0 amp

This proves the theory that the reactive part of the core exciting current can flow in the transformer secondary.

Carl

But the capacitive current in secondary is load current because it is amp-turn balanced by the primary. Exciting/magnetizing current is not amp-turn balanced at all. No, I don't agree.

Without the cap sec load, the flux is sinusoidal because all the mag current harmonics have a low Z path to flow. A substantial portion of the mag current is fundamental, 1st, lagging the primary voltage by approx. 90 deg. Whan the cap loads the sec, another current comes into play in the sec, which leads the secondary voltage by approx. 90 deg. This leading current however, gets counter-balanced by leading primary current. By definition this is not exciting/magnetizing current, but rather, it is load current.

The fact that the leading load current amp-turns cancels the lagging magnetixing current is incidental, as it does not shape the core flux. Under no circumstances can the capacitive leading secondary current be considered secondary exciting/mag current. If we parallel several caps, one at a time, any leading sec current gets counter-balanced by additional leading primary current.

That is prima facie evidence that this leading current is not exciting current. Not even close. I alopogize if I came across as rude.

Claude

 

[m.s.j]

According to some definitions, the primary side of a transformer is the side that connected to power supply. Please specify active power supply or reactive power supply. Please note to attached page, as you know, illustrated situation is possible in power system, which side of transformer (1 or 2) consist magnetizing current.