A layman’s guide to power blackouts

A primer on a hot-topic


by Kumar David

There is a buzz around this topic for obvious reasons and this brief primer is an attempt to provide a few theoretical insights in laymen’s language. I will deal mostly with general principles or with events of some time ago and not touch on the blackouts of September 2015, February 2016 and March 2016 because of my involvement in some aspects of an ongoing investigation. There are different ways in which to approach the topic and I have chosen to do so under the subheadings ‘The System as a Whole’, ‘Planning and Management’ and ‘Equipment’. Power system blackouts are topics esoteric to non-specialists, so occasionally a didactic tone may creep in – apologies for that.

Not only in recent months but previously too, our system and systems around the world, have shut down. It is the behaviour of the network, that is the functioning of the system as a whole, that holds the key to the problem; it is not the repetition of the same defect time and again. I will make a start by throwing light, as simply as possible, on the cohesion of the network.

The system as a whole

The electricity supply system consists of dozens (in a large country hundreds) of alternating current (AC) generators connected to hundreds of load points by a network. I am referring to high voltage of transmission interconnections. Lower distribution networks that tap power from the transmission network (grid) and convey it to a multitude of individual loads are usually not significant in the security and stability of the system of generators and transmission lines as a whole. Each generator does not service a particular load; rather, the output of all generators is "pooled" into the transmission grid. Individual distribution systems (municipalities, industries, rural regions) tap power from the grid in much the same way as a housing estate taps the water-main and then distributes water to individual housing units. There is one critical difference though; electricity cannot be stored in the grid, therefore there has to be a near instantaneous balance of production and consumption of electric energy.

The crucial point about the grid is that it is the critical structure that binds all the generators and loads into what in the jargon is called a synchronised system. A large-scale blackout is a failure of all or part of the interconnection. Power supply networks are interconnected for three reasons; viz as load varies the cheapest generators can be run and the more expensive ones held back for use at peak time only; second, power from different areas and types of plant (think Laxapana and Norochcholai) can be sent to far flung load centres depending on needs and availability, and third a large strong interconnection is stable against all but the most serious disruptions. Nicola Tesla proved conclusively that AC is superior to direct current (DC) for electricity supply. The reasons will take me too far from my topic today.

Interconnection however hides a secret; it is like the mighty warrior Achilles who could be fatally wounded in the heel. His mother Thetis dipped the infant in the underworld river, the Styx, to make him invulnerable, but overlooked the detail that she was holding him by his heels. AC systems, excellent for handling large amounts of power, too have their Achilles heel. In normal operation all generators must work at strictly the same frequency, the sinusoidal shape (if you remember it from school) of their voltage waves must all oscillate at the same rate. The waves can, and must, be slightly out of phase (shifted) from each other – see figure - but must maintain a lockstep of frequency during steady operation. All system wide blackouts eventually are the result of synchronism (lockstep operation) being disrupted by big external or internal shocks, causing generators to attempt to run at different frequencies leading to failure of harmonized operation.

This cannot happen easily. When faults, unexpected load changes and other disruptions of small or medium magnitude disturb synchronous operation, powerful inherent corrective effects (synchronising power flows) take place between different portions of the system and pull generators and regions into stable synchronism again. However, very large disturbances, improper design, or incorrect operation can undermine this, causing too many important transmission lines to trip-out or too many generators to disconnect. If there is large loss of transmission capacity the paths available for stabilising power flow between errant generators are inadequate. Or if too many generators become disconnected, load demands exceed the simultaneous production of power and the whole system slows down, frequency declines and a cascade of events force the system to grind to a halt. This is how blackouts occur.

I don’t know if the following analogy is helpful. The body is ceaselessly exposed to germs, radiation and genetic mutations but the immune system continuously overcomes and corrects the ‘faults’. Similarly in power systems, the natural laws of physics, not automatic control mechanisms (though they can help), cause synchronising power to flow in the correct direction between disturbed generators in an attempt to restore the normal lockstep. If the assault on the immune system exceeds the ability of the immune system to cope, the body goes down; likewise no electricity supply system can ensure 100% reliability (the best like Hong Kong offer 99.9%). No immune system can promise eternal life. The converse is not true; every time poor reliability is manifested in electricity supply it is not proof of some eternal law of physics. It could well be the outcome of poor planning, design or operation. It all depends; and this is why big blackouts need to be investigated at a high professional level. The cost to society of an eight-hour all-Island blackout is in the range of tens of millions of rupees.

Transmission lines are exposed to regular lightning strikes, fully loaded generators may trip out for unforeseen reasons, and transformers may fail. The system is designed and operated to be secure against any one such "single outage", also called "N-1 secure" (N is jargon for the whole lot). Computers in System Control Centres continuously repeat simulations of what could happen after any single-outage and human controllers adjust the operational status (usually change load sharing between generators) if the system is seen to be insecure for a "credible outage". If in a real world case, a single transmission line trip, a single generator disconnection, or one transformer failure, causes a system-wide botch, it is usual to blame system operators for failing to live up to expected standards of security.

Planning and management

The expansion of the power system has to be planned at least a decade ahead and the plans regularly updated (rolling plan) and Lanka’s experience from the Laxapana days through the great Mahaweli Complex period (both irrigation and power) has been exemplary up to the 1990s. The Irrigation Department, Mahaweli Authority and CEB planners coordinated with top-class foreign consultants and international funders to do Lanka proud. However, from the time private power projects were introduced (and Puttalam coal power inexplicably delayed for two decades) the planning experience has been murky. Allegations of a financial nature have often surfaced, mostly against high political personalities at the taking-end and business on the giving-side. I am not implying that private sector involvement in power projects is necessarily corrupt, but I am saying that Lanka’s experience has been problematic.

The involvement of politicians in project decision making (these are not in the multi-million but the multi-billion rupee range) has been a curse as it contributes to wrong decisions, wrong sequencing of project options, cost overruns and interference with investigations of system failures (yours faithfully has first-hand experience of one such bigtime mystery in 2009). CEB engineers are technically sound, but the overall structure of the organisation has not undergone radical rethinking for a long time – maybe not since LECO was spun off (I may be wrong here). Privatisation is no solution but restructuring will be a boost. An open structure with greater transparency and public participation and the creation of advisory committees that include outside experts (even a foreigner or two) which meet say twice a year to review ongoing work and make recommendations will, I believe, strengthen the organisation both in fact and in image.

Prof KKYW Perera has mentioned to me the usefulness of a twin-ladder system for cadre advancement for engineers, after an initial say 10 year period needed to round off experience. One ladder to be a specialist stream (like consultants, to take an analogy from medicine); the other will be generalists whose aspirations are more managerial and seek a location along the corporate ladder. It would be useful to revive consideration of this option.

The CEB has, by its intrinsic nature and traditions, evolved into an organisation which will not function well unless ‘pole-positions’ are occupied by able engineers. But there are thousands of others, technicians and administrators, who are overlooked when discussion focuses on engineers. This is not good and I would like to devote a separate piece to this facet at a later date. In any case this article is not, and cannot pretend to be, a comprehensive review. As indicated at the start, it attempts to bring some important topics to the attention of interested laymen.


What goes without saying is that the best available, at an affordable price, must be procured and coordination ensured, for example among protection systems and between protection systems and the plant to be protected. Most laymen are unaware that protection accounts for no more than 1% or 2% of the total costs of a power plant or transmission line, but at critical moments its role is supreme, like the umpire in a cricket match – a clothes peg, momentarily transformed into kingmaker.

Everybody in Lanka has heard ad nauseam of the need to strengthen procurement and tendering procedures and exasperation has brought many to the point of exclaiming: "Let the bastards take their 10%, but for heaven’s sake make the right choices!" I need say no more.

Aside from drawing attention to protection equipment and the importance of proper procurement, there is one other matter about power stations that will be of interest. Hydro plant is mechanically robust and uncomplicated; gas turbines are a little more complex. Thermal power stations involve a boiler, steam circuit, turbine, generator, condenser, feed-water pump and thousands upon thousands of other auxiliaries; this is a different ballgame. Not only are there thousands of pumps, relays, switches and whatnot that can malfunction causing shutdown or derating of the main plant, but the start-up and shutdown routines are far more complex and time consuming. Norochcholai with 900 MW is the largest power station now, Sampur-1 (500 MW Indian) will be the next giant to be followed by Sampur-2 (1,200 MW with Japanese help) which will be the biggest monster among them all. A lot of operational training needs to be done.

What next after these last two coal projects? We need to plan two decades ahead. Encouraging green energy (wind, solar, bio and mini-hydro) is vital but will not add to more than about 15% of total energy needs, at best, for another generation. What are our power planners thinking longer term? I will bite my tongue and sign off at this point.

animated gif
Processing Request
Please Wait...