Testing high speed DC circuit breakers

Electrical Tester - 8 October 2021

 

Author: Wim D'Hooghe

Although many of our readers will be familiar with circuit breakers used in AC power systems, it’s possible that they may know less about the high-speed DC circuit breakers (DC HSCBs) that are frequently found in transportation systems. To remedy this, Wim D’Hooge, a specialist for Megger’s BALTO product range, provides a useful introduction to DC HSCBs, their applications, and how they are tested.

Introduction

What is a high-speed DC circuit breaker? It’s a protective device with a very simple characteristic – it should always trip at the set maximum current (Ids) within a specified time, typically 10 ms, irrespective of the magnitude of the overcurrent. When testing these devices, the parameters that need to be measured are Ids, the opening time of the breaker, and the contact resistance for the main contacts. These breakers are clearly very different from the types found in the power industry, so where are they used?

Their main application is in the transportation sector. In many countries, national railway systems that employ electric traction are split approximately 50/50 between AC and DC traction current. Typically, long distance highspeed links use AC, whereas local rail services, as well as trams, metros and other light rail services use DC. And wherever DC traction is used, DC HSCBs are needed. However, this is not their only application; around 10 % of DC HSCBs are used in other sectors, such as marine and mining. This percentage may well increase in the future with the growth in popularity of renewable energy sources, many of which are DC.

In the transportation sector in particular, DC HSCBs have been around for a long time, and it’s not at all unusual to find them in equipment that has been in service for years or even decades. DC HSCBs are found in the traction power substations and also in the rolling stock. As an interesting aside, it’s worth mentioning that it’s quite usual for the people working in the substations to have very little to do with those who work on the rolling stock; they tend to operate as two separate groups.

There are numerous manufacturers of DC HSCBs producing a diverse range of products, but all of them can be tested with BALTO systems. The first of these goes back to 2008 when, rather surprisingly perhaps, it was virtually impossible for users to test their DC HSCBs! Some built cumbersome test systems the size of a large room, while others didn’t test their breakers at all, except by giving them a visual inspection during cleaning. This situation was clearly unsatisfactory and led to the development of the BALTO system.

Testing DC HSBCs

The principal occasions when DC HSCBs need to be tested are:

  • During commissioning
  • In preparation for a line upgrade (when more trains are to run on a line, more current must be available, and the substation breakers must be set to higher trip current, which requires recalibration and retesting)
  • As part of routine maintenance (typically every 2 to 3 years in a substation, more frequently on rolling stock)
  • After a breaker has been repaired or refurbished

Breakers on rolling stock can be tested conveniently with a BALTO Compact test set, but for the larger breakers used in traction substations, the BALTO Modular system, which can accommodate up to five power units, is used. The power units are rated at 4000 A, which means that a fully loaded system, mounted on a trolley for mobility, can supply up to 20 000 A. In a master-slave configuration, this can go up to 40 000 A.

This high current rating may suggest that the test set is a potentially dangerous piece of equipment, but that’s definitely not the case. The power source is batteries and ultra-capacitors that provide a 15 V supply to a DC/DC converter. The output of the converter – the test current – is at just 4 V, so it’s as safe as an AA battery! Of course, when a test is being carried out that involves injecting, say 10 000 A into a breaker, the cables will still sway a little due to the electromagnetic field produced by the test current.

There are separate IEC standards for testing traction power substations (IEC 61992-2) and testing rolling stock (IEC 60077-2/3), but both specify that the tripping current, Ids, should be measured while the current through the breaker is increasing at a rate of 200 A/s. In practice, before making this measurement, a quick test is made to determine (approximately) the breaker’s trip setting.

When carrying out the quick test using, for example, a test set fitted with two 4000 A power modules, the test set injects a current that increases from zero to 8000 A in 600 ms. That’s far too fast to make an accurate measurement of Ids, but it does give a rapid indication of the breaker setting. With the quick test, a breaker set to trip at, say, 6000 A will probably show a trip current of around 6500 A – too high, but still useful information.

The next step is to carry out the test according to the IEC standards, and this is illustrated in Figure 1. Since the approximate value of Ids is known from the quick test, the test current can be arranged to increase very rapidly until it approaches this value. This makes the best use of the charge in the test set batteries and ultra-capacitors. When the anticipated Ids value is approached, the rate of change of the test current is reduced to 200 A/s, as stipulated in the standard. The tripping point, shown by the solid vertical line in Figure 1, must be in line with the IEC standards.

Figure 1: Measuring Ids

It is worth mentioning that the energy stored in a BALTO test set is enough to supply the test current for up to 5 seconds, during which time it will increase by 1000 A (5 seconds x 200 A/s). This is a large sweep of current and will enable the trip current to be measured accurately even if the result of the preliminary test was a little wide of the mark. In contrast, test sets from some other manufacturers can only supply the test current for around 0.5 s. That gives a current sweep of just 100 A, so it’s very easy to miss the tripping point, which can lead to a lot of inconveniences and wasted time.

Figure 2: Measuring the breaker opening time

After measuring Ids, a test is carried out to determine the breaker’s opening time, as shown in Figure 2. For this test, the current is increased slowly until it is 10 % less than the measured value of Ids, then it is increased in a step change to 10 % above the measured value of Ids. The breaker will trip, and the time between the step change in current and the cessation of current flow is measured. This is the trip time. In the case shown, it is 11 ms, which is a good value for a DC HSCB.

The final test is to measure the contact resistance, which determines the voltage drop across the breaker’s main contacts. This should be in the region of 30 μΩ – values significantly higher than this indicate a problem. The test current used to measure the contact resistance is specified by the breaker manufacturer. Typical values are 1000 A for small breakers and twice that for larger types.

Between them, these tests provide a complete evaluation of the performance of a DC HSCB and, provided that suitable test equipment is used, they can be carried out very quickly. In most cases, the tests themselves take only around five minutes to perform, but depending on the circumstances, making the necessary test connections can add significantly to the overall testing time.

Breakers with protective relays

Up to now, we have considered DC HSCBs with direct release. That is to say that they are tripped solely by the current flowing through them. Breakers fitted in traction substations may, however, be tripped indirectly via a protective relay (see Figure 3) although this arrangement is not used by all transport undertakings. An option is available for the BALTO system that allows breakers with protective relays to be tested by simulating the output of a shunt, which can be 60 mV, 150 mV or other values, and injecting an adjustable current into the protective relay. Note that users who only want to test breakers with direct release will not need this option.

Figure 3: A breaker with a protective relay

Conclusion

With the right equipment, testing high-speed DC circuit breakers is fast and easy. And it is an effective safeguard against the high costs and disruption that can result from a circuit breaker failure. In a transport network, a breaker that trips unnecessarily can inconvenience thousands of passengers, and one that doesn’t trip when it should can destroy expensive equipment leading to even greater cost and disruption. The consequences of breaker failures in other sectors are equally dire so, for all users of DC HSCBs, a modest investment in test equipment will surely be money well spent!