Reliability 101: The Basics of Electrical Distribution System Reliability

By Dan Steiner, PE, CEM, CDSM, CEA, CMVP; President

Editor’s note – This article is the first in a four-part series on the important subject of electrical power systems reliability.

Reliability is one of those words with varied meaning, depending on its use. Its root word, reliable, means “dependable,” something you can count on. But reliability also carries with it the ideas of honesty (as in “performs as promised”), fitness for purpose, and resistance to failure.

All of these admirable concepts and many others are rolled up in the term electrical system reliability. And if you’re a plant manager, maintenance supervisor, or other professional with responsibility for the successful operation of your facility, as much as you want reliability from the car you drive or the weather forecast you listen to, you’ll want it even more from your facility’s electrical power system.

In this article, I’ll lay the groundwork for future discussions on electrical reliability – and how to achieve it.

What Is Electrical Reliability?

To some, the “reliability” of a system means simply having the system available when it is needed most of the time. But what constitutes “most of the time”? Sixty percent? Ninety? What if the system doesn’t work right when it is available? Few industrial or commercial facilities can get by on such a loose approach to system reliability. This is especially true for their electrical distribution systems.

In his book Maintainability & Maintenance Management, Joseph Patton defines reliability as “the probability that an item will perform its intended function without failure for a specified time period under specified conditions.” This captures the essentials of what we mean by electrical reliability. A reliable electrical system is one that works as designed without fail, and it does so with ideally 100% reliability.

Now, the questions become is this even possible? If it is, is it affordable? In both cases, the answer is yes. And especially in light of what system unreliability can cost, it’s really the only alternative for a company that wants to guarantee its present and future competitive position.

The High Cost of Unreliability

I recently saw a headline for a seminar advertisement that asked if my company could afford to lose $7,000 an hour. Naturally, this caught my eye, so I read further. The ad summarize the average cost per hour for different industries when they experience an electrical power failure. The numbers were stunning: $30,000 per hour for the food processing industry, $100,000 for metal casting, $200,000 for the automotive industry – and these figures were from a 1996 study!
Maintaining a reliable electrical system does cost money. But how does this compare to the cost of an hour, a day – a week – of lost production?

There are other potential costs for electrical unreliability. Valuable equipment can be destroyed, personnel can be injured (or worse), customers can experience their own financial catastrophe due to the loss of your products and service. Rarely is a power outage a “small” incident.

Main Factors of Reliability

Electrical system reliability is a function of three basic factors:
- Design and installation, including protective device coordination and selective zones of protection
- Proper preventive maintenance (a.k.a., preventative maintenance)
- Proper system operations

Take away any leg of this stool and you have an unreliable system. A poorly designed or installed system is unreliable from the start. Even good systems begin to deteriorate as soon as they are put into service; poor maintenance hastens this process. And studies show that roughly three-fourths of all unplanned production shutdowns are due to human errors.

Quantifying System Reliability

Can reliability be quantified, not just estimated? Yes, but, first, you have to decide what a “loss of power” is. Loss of power, not unreliability, is the opposite side of electrical reliability.

To put this in perspective, a facility experiencing an annual loss of power for only 0.003% of the year is said to have 99.997% reliability. This sounds great until you realize that 0.003% is 26 hours – more than one full day! – of lost power. For nearly every facility today, this would be a disaster! Reliability can be quantified…but is it a level you can live with?

Keep in mind that historical data on past system downtime and the resulting losses, plus other investigative records, are important to setting the benchmarks used for quantifying electrical system reliability.

The Benefits of Reliability

Because reliability can be quantified and loss-of-power costs can be calculated, both numbers help determine ROI on projects designed to improve reliability. Generally, in today’s electrically dependent world, improving electrical reliability yields far more benefits – financial and otherwise – than the costs of implementation.

Important Reliability Terms

To get the most out of future discussions on electrical reliability, you need to speak the language. Here are a few terms you should know:
Availability – The probability that the system will operate as designed under specific conditions, expressed as a percentage (e.g., 99% availability).

Critical loss of power – A loss of electrical power for any time period that poses a threat to the three critical Ps: product, property, and people.

Mean time between failure (MTBF) – The average time between when a piece of electrical equipment breaks down and when it breaks down again.

Mean time between replacement (MTBR) – For electrical equipment, the average time between complete replacements of specific equipment.

Mean time to repair (MTTR) – The average time between when a piece of electrical equipment breaks down, and when it can be repaired and put back in service.

N+ – A method of identifying an electrical system’s redundancy, or ability to cope with a power loss. An N+1 system has one redundant path that will energize during a power loss, an N+2 system has two such paths, etc.

Probabilistic risk assessment (PRA) – A statistical analysis of an electrical system that attempts to answer three questions: What can go wrong? How serious could the impact be? How likely is this to happen?

In my next article, I’ll discuss reliability and the electrical preventive maintenance (EPM) program.