Consider a "small" autonomous grid, which is based mainly on RES and has a few diesel generators, like in the following figure

enter image description here

Everything runs smoothly, feeding the small load. Suddenly, the big angry load tries to connect and its transformer draws a large magnetization current (let's say 3 or 4 times the nominal current).

My question is: Is there a reliable strategy to follow so that the DC/AC converter manage that current without shutting down? Obviously, there will be a current limiting function, but is it fast enough?

Some conditions

  • Load's transformer in inaccessible
  • We cannot depend only on the diesel generators to supply that current
  • We want to avoid a partial shut down of the system
  • Voltage drop, for 10-15 cycles is allowed

I know the question is broad and the answer depends on various parameters, but is there an "standard" method to deal with this problem?


The nominal current is not strictly specified, but if we consider that the big load is ~ 1MW @ 400 phase-phase, the current is \$ I = \frac{1MW/3}{\sqrt3 \cdot 400} = 480 A \$


The big load is mostly resistive - it heats up water. There may be some motors, but this is not the issue.

  • \$\begingroup\$ Do you know the nominal current? Please share in original post. \$\endgroup\$
    – Huisman
    Apr 1 '19 at 18:27

Magnetizing current flows through the primary winding of a power transformer to energize the core. This is independent of the loads connected to the secondary winding and is voltage dependent.

When approaching primary saturation voltage, from high input voltage;

  • the primary current harmonics (5th and 3rd) begin to rise
  • the inductance drops 10% at threshold of saturation
  • thus reactive current rises
  • and lowers the input impedance.

When the AC voltage is switched off , there is likely to be a Remanence or magnetization charge in the core that remains for a long time. Thus if voltage is applied in the different phase crossing from the previous shutoff the magnetization can get out of sync and if the core does not have enough margin to saturation may cause a sharp drop in inductance until it self-balances. ABB makes smart reclosures ($) to resolve this issue for large XFMR's.

If I understand correctly, instead of 10% excitation current you are getting 300% to 400% rated current (or perhaps just normal operating current) on startup. This is likely due to core saturation from Remanence.

This tells me something about your nasty load but leaves me guessing about my assumptions and your measurements for margin to saturation, X, L/R and, harmonic content, load characteristics and other ratings.

A little more info on your loads and transformer tests will help suggest a better cure.

It this surge is purely motor-load surge then it is not "transformer surge" from Remanence, Core, Saturation, overvoltage etc.

Your options then are ;

  • Choose an HMCP circuit breaker with a higher instantaneous trip range.
  • Substitute a thermal magnetic circuit breaker with a higher instantaneous trip range.
  • reduce tap voltage
  • 1
    \$\begingroup\$ The magnetization current is "independent" of the load, but the transformer's "dimensions" are not. Bigger load needs bigger transformers, hence bigger inrush current. This technique from ABB seems interesting, so I will look further. Thanks! \$\endgroup\$
    – thece
    Apr 1 '19 at 18:51
  • \$\begingroup\$ Can the startup load be regulated externally? Motors are often 5x but can be cured by VFD's or sequencing multiple loads. But this load induced then it won't cure Remanence saturation. \$\endgroup\$ Apr 1 '19 at 18:53
  • 1
    \$\begingroup\$ Nope, unfortunately it starts at full power. This was the first thing we tried to do :P \$\endgroup\$
    – thece
    Apr 1 '19 at 18:54
  • \$\begingroup\$ So you must determine if it is caused by loads or Remanence ( no load) \$\endgroup\$ Apr 1 '19 at 18:56
  • \$\begingroup\$ Is this a problem with RES? lacking a soft start? \$\endgroup\$ Apr 1 '19 at 19:21

Ok, so after some time I have found some suggestions, based on several papers from IEEE and other sites. I will cite them in this post.

  • Employ a larger (relatively to what the nominal power will be) T/F so that it does not go into deep saturation and draw a significantly lower inrush current. Obviously this increases the cost of the T/F. [1]
  • Employ a VSI (Voltage Source Inverter) to moderate the voltage applied to the T/F. This kind of solution is known as Voltage Sag Compensators (or Voltage Dip Compensators). [1] [2]
  • Sequential (not instant) closing of each phase of the switch that connects the T/F with the grid.In this way, dramatic reduction (or zeroing) of the inrush current is achieved, since each phase is connected at the appropriate time. Depending on the application, a residual magnetization measurement system may be required. The implementation also depends on the topology of the T/F (3 phase, 3x1 phase, Y / Δ). [3] [6]
    • This method can also be combined with the placement of a resistance, from the common point of the windings to the earth, which during the transient phenomena is in series connected to the windings. [4] [5]
  • Pre-magnetizing the iron core so that it does not need a large inrush current. This technique requires a system to measure the residual magnetization of the T/F since its previous deactivation. [7]


[1] Po-Tai Cheng, Wei-Ting Chen, Yu-Hsing Chen, Chia-Long Ni, and Jarsun Lin, “A Transformer Inrush Mitigation Method for Series Voltage Sag Compensators”, IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 22, NO. 5, SEPTEMBER 2007

[2] Yu-Hsing Chen, Chang-Yi Lin, Jhao-Ming Chen, and Po-Tai Cheng, “An Inrush Mitigation Technique of Load Transformers for the Series Voltage Sag Compensator”, IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 8, AUGUST 2010

[3] Mukesh Nagpal, Terrence G. Martinich, Ali Moshref, Kip Morison, and P. Kundur, “Assessing and Limiting Im-pact of Transformer Inrush Current on Power Quality”, IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 21, NO. 2, APRIL 2006

[4] Yu Cui, Sami G. Abdulsalam, Shiuming Chen, and Wilsun Xu, “A Sequential Phase Energization Technique for Transformer Inrush Current Reduction—Part I: Simulation and Experimental Results”, IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 20, NO. 2, APRIL 2005

[5] Wilsun Xu, Sami G. Abdulsalam, Yu Cui, and Xian Liu, “A Sequential Phase Energization Technique for Trans-former Inrush Current Reduction—Part II: Theoretical Analysis and Design Guide”, IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 20, NO. 2, APRIL 2005

[6] Yu-Hsing Chen, Ming-Yang Yeh, Po-Tai Cheng, Jen-Chuan Liao, and Wen-Yin Tsai, “An Inrush Current Reduc-tion Technique for Multiple Inverter-Fed Transformers”, IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 50, NO. 1, JANUARY/FEBRUARY 2014

[7] Rickard Ekström, Senad Apelfröjd, and Mats Leijon, “Transformer Magnetizing Inrush Currents Using a Directly Coupled Voltage-Source Inverter”, Hindawi Publishing Corporation, ISRN Electronics, Volume 2013, Article ID 361643, 8 pages, http://dx.doi.org/10.1155/2013/361643

[8] KATSUJI SHINOHARA, KICHIRO YAMAMOTO, KENICHI IIMORI, YOSHITAKA MINARI,OSAMU SAKATA, and MICHIO MIYAKE, “Compensation for Magnetizing Inrush Currents in Transformers Using a PWM Inverter”, Electrical Engineering in Japan, Vol. 140, No. 2, 2002, Translated from Denki Gakkai Ronbunshi, Vol. 121-D, No. 7, July 2001, pp. 730–738

[9] Liana Cipcigan, Wilsun Xu, and Venkata Dinavahi, “A New Technique to Mitigate Inrush Current Caused by Transformer Energization”, 0-7803-7519-X/02/$17.00 © 2002 IEEE


Hope it helps anyone with the same or familiar problem.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.