DC component in transformer magnetising current

Status
Not open for further replies.

matts0

Member
Hello
I have done some tests on the magnetising current in 4 types of transformer winding connections. They are Yy,Dy,Yd,and Dd(All with secondary winding open).
Tests are done on the primary phase A.
I found that in Yy,Dy,Yd connections the waveforms of the magnetising currents all contain high amout of DC component but in Dd connection, it is quite low( Also the waveform of the magnetising current in Dd has the best shape of sine-wave).

I understand the theory of the flow of 3rd order harmonic in different windings but could someone give some idea on where the DC component come from?
Thanks
 

matts0

Member
Thanks for the answer.I just realised that I used the voltage and current transducers to take the signals of voltage and current to the oscilloscope and then reconnect it to computer to display the waveforms on the computer screen.
is it possible that the DC component come from this process?
 

SG-1

Senior Member
All the current & voltage transducers I have ever seen had a pure DC output ( like 4- 20 mA ) that was in proportional to the AC input signal. What make & model of transducer did you use ?

Scopes will not generally read current without some type of current sensor that can transform the current signal into voltage.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110402-0828 EDT

The unbalanced current, DC if you want to call it that, is that current which is used to rebalance the flux state of the ferromagnetic core. This is a transient event taking place over a large number of cycles.

See my photographs P6-P8 at http://beta-a2.com/EE-photos.html . P6 and P7 illustrate two different initial residual flux states, one high and one much lower. P8 is the steady state condition after the flux state of the core has become balanced. This steady state condition is the same for any initial condition and therefore has a zero unbalanced component because there can be no sustained unbalanced component.

Note: in both P6 and P7 the AC applied voltage to the transformer primary started near a positive slope voltage zero crossing.

.
 

matts0

Member
The DC component showed in the oscilloscope screen is about 15mA.It is small.However compared to the magnitising current in the coil it is still very significant. Will it be a problem?
I do not have details of the transducers.I just did my experiment on the PS110 transformer trainer and I used the on-board transducers.
Thanks.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110402-1411 EDT

matts0:

Steady state there is no DC component in the input current to the primary of a transformer, loaded or unloaded. Even if you put a half-wave rectifier on the secondary with a load, which produces a DC current in the secondary, there will be no primary DC component on a steady state basis. You have an instrumentation error, your zero balance of the scope amplifier and/or transducer, if you see a DC component and you are truly at a steady state condition, and the source is not producing a DC component.

If your source is an isolation transformer and your transformer under test was the only load, then the primary side of the test transformer would have no DC component. Put a pure resistance in series with the primary of your transformer under test. For 120 V nominal source voltage use 1 ohm if currents are in the range of possibly 1 A. Connect a DC voltmeter across the current sensing resistor. if there is a DC current, then you will read a voltage. 1 A at 1 V is 1 W so use a 5 W resistor. You should see a blip on the meter when you first apply power to the transformer. This will be different from one time to the next. And zero after steady state is reached in a few seconds.

Using my 175 VA transformer, 1 ohm, and a Fluke 27 I saw above and below 0.1 V for the peak and decay to 0 within a couple seconds. With my Simpson 270 on the 2.5 range sometimes about 0.3 V. Note can go either + or -. Note that both the Fluke and Simpson perform integration on short pulses like these. Thus, the reading is not a direct measure of the peak that occurs.

You spoke of a 15 MA offset, but did not mention full scale or the peak magnetizing current.

I have no idea of what a PS110 trainer is, or what is the rating of the transformer with which you are experimenting. This is information you should provide.

You will note in my scope photos I referred you to, that the transformer was rated at 175 VA. So under full load with a pure resistance load the input RMS current would be slightly higher than 1.46 A, or a peak of 1.46/0.707 = 2.06 A. Also note that I indicated the peak steady state magnetizing current was about 250 MA, but the scope trace puts it a little less than 200 MA. If you scaled these values to some other size transformer the results would be quite similar. Most transformers are designed with the same kind of steel for the core and criteria for the normal flux density. New higher efficiency transformers might be somewhat different, and toroid transformers will likely have a more square loop core material.

.
 
Last edited:

matts0

Member
Thank you,gar for introducing the DC inrush current and the measuring method.
But what do you mean by "rebalance the flux state", is the principle same as in the motor start-up?
Also I have another question..The transformer I used is rated at 3 kVA for three-phase and the magnetizing current I measured was about 200MA.The unbalanced current is presented at steady state and I think it is truly an instrumentation error problem.However because I directly connected the transducer in series with the primary coil,will the DC component introduced by the transducer have negative effect on the transformer?As it was in the same scale(150mA) with the magnetizing current.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110403-1233 EDT

matts0:

I suggest you find a book on electric and magnetic fields and study hysteresis curves. One suggestion is "Electric and Magnetic Fields", Steven S. Attwood, 1949, John Wiley, and Sons. Probably hard to find.

When you remove excitation to a ferromagnetic circuit you usually leave it with some residual magnetic flux. The new shape of the hysteresis curve at this residual point is a function of that residual point, and the new excitation applied to the circuit. Transformers are magnetically driven hard, meaning somewhat into saturation in normal operation. As you continue to apply this excitation the hysteresis will gradually move to a balanced equilibrium state. Initially it is not and this is the cause of the unbalanced current.

The primary cause of high startup current in a motor is the mechanical load (inertia). Some very short time current is a result of the magnetic circuit. A motor has a much softer saturation characteristic than a transformer because of the large air gap in the magnetic circuit compared to a transformer.

To learn about transformers and inrush current you would be much better off using a single single phase transformer. But maybe that is what you effective have if it is three single phase transformers in a Y to Y configuration.

You need to define your current sensing transducer. If it is a Hall device, then somewhere there is a zero adjustment or capacitor coupling of the output signal.

If the current sensor is a current transformer, then there is no steady state DC component, and your problem would be the adjustment of zero on the scope.

I do not believe there is any series inserted current transducer that would introduce any DC component in your test circuit.

Study the characteristics of every component in the circuit with which you are experimenting. Get books out and study each component.

.
 

LarryFine

Master Electrician Electric Contractor Richmond VA
Location
Henrico County, VA
Occupation
Electrical Contractor
What happens when you swap the meter leads? If the indicated polarity changes, it's real. If it doesn't, it's instrument error.

I've been reading this thread off and on, and I suspect an outside DC voltage source, possibly even galvanic action induced.
 
Status
Not open for further replies.
Top