NO LOAD CURRENT 1

[rukmal]

We have designed a 100kVA 11kV/415V distribution transformer & in final testing it was identified that the no load current exceeds to specified limits that are laid under IEC 76.I would like to know the possible causes for such a obersvation

* Can it be the air gap deviations?
* Can it be the quality of the core steel used ? (Actually the design flux density is within the specified limits of the core steel supplier)
* Can there be any other causes for that ?

It would be a great help if someone can find us a proper theoritical explanation for that or a suitable source document.

 

[electricpete]

From the equivalent circuit, the no-load current is limited by the series combination of primary leakage reactance X1 and magnetizing reactance Xm. Since Xm>>X1, Xm will be the determining factor.

If it is an established design which has previously performed properly, then you need to look at manufacturing deviations which could decrease Xm. As you mention quality of core steel may be one. Also I remember that a single lamination is composed of multiple plates with a butt joint or something similar. If there is an air gap at that butt joint it will increase the reluctance of the path and therefore decrease the inductance.

If it is a new design you may want to ask if you have enough primary turns to create a high enough magnetizing reactance. Also of course the core mean length being too long or area not big enough would cause high reluctance and low magnetizing reactance.

It seems remotely possible that dc testing may have left the core in a magnetized state which could cause saturation every half cycle and increased (nonsinusoidal) magnetizing current.

Likewise a test voltage that is too high or test frequency too low may force you into saturation and therefore higher (nonsinusoidal) magnetizing current. And of course improper design may place you in saturation even under normal voltage/frequency.

One good starting point to eliminate all of the saturation-related causes might be to check the harmonic content of the current.

I apologize if I'm telling you what you already know.

 

I understand that the magnetized current is superior to the norm.
The data of the manufacturer of the plate are more theoretical than practical.
The experience is essential in transformers designe.
With oriented grain plate, the usually problems are more of heat and losses, that of magnetized intensity.
The mainly solution is to modify induction.

 

[jbartos]

Suggestion: The higher magnetizing current can also be caused by somewhat inferior workmanships, such as stacking of turns, insulation, magnetic characteristics of transfomer metal sheets, and transformer assembly. Some, transfomers have special provisions for adjustment of the transformer impedance Z that is stated on One Line Diagrams. It significantly affects short circuit levels in the downstream power distribution system.

 

Sr. JBartos.
the insulators do not have appreciable relation with the case.
High and wide disposition of the window has importance, but little.
The type of plate defines the induction to apply.
The turns are function of the induction.
One is a normal distribution transformer.
The short circuit tension is another different war.

 

[prc]

Some more details are required for a proper analysis. It is said -no load current is higher than IEC76.1 specified limits. To my knowledge IEC is not specifying any specific values for no-load current ,but only gives a tolerance of 30 %max over guaranteed value. From the question it is not clear how the guaranteed value was taken.You will never get current as mentioned by steel supplier .A building factor depending on design of lamination joints ,extent of lamination handling in factory, whether in-house annealing was done etc,etc,has to be added to the value given by steel maker.It will definitely be more than 30 %,sometimes even 100 %!

 

[rukmal]

To prc,

Do you think in house annealing would solve the problem? At the moment we don't do in house annealing.Actually the guarenteed value was based on our design programme output which is based on on B-H curve supplied by our steel supplier.

 

[prc]

In house annnealing will reduce the excitation current. But the extent of reduction ,you have to conclude from experiment.
Easier solution is to modify the building factor used in your programme based on actual test results.