If Henry Ford walked into most any factory today, his jaw would likely hit the ground. The modern factory floor looks little like the assembly line of Ford's day. Massive humming machines, robotic arms, and a proliferation of sensitive electronics (such as programmable logic controllers (PLCs), variable-frequency drives (VFDs), and network communications devices) have led to tremendous gains in productivity. Ironically, these high levels of productivity translate to high profit loss when an electrical problem causes a shutdown.
One hour of lost production today costs a whole lot more than it did even five years ago. Why? Many of the electronic devices that contribute to productivity gains also contribute to productivity losses because of their sensitivity to such problems as sags, swells, and voltage transients. To further complicate matters, some of the electronic equipment susceptible to damage from power quality problems contribute to power quality problems—as in the case of VFDs.
What does this mean to the average industrial facility? It means the need to monitor, troubleshoot, and preempt power quality problems is greater than ever before. This article focuses on one type of power quality problem: the voltage transient.
In the past, troubleshooting electrical problems often led to disrupted or failed industrial equipment. This was especially true with transient voltage spikes, due to their short duration and often unpredictable pattern of occurrence. Identifying these transients involved the use of expensive, specialized, portable monitoring equipment. This meant renting or purchasing equipment (at typical costs of $15,000 to $20,000) or even bringing in a consultant specializing in power quality troubleshooting and analysis.
Then, with the portable monitor in hand, personnel were faced with the difficult task of deciding where to look for the transient. Installing the portable device was equally difficult. Engineers were sometimes required to shut down a system, and they frequently had to relearn how to set up the device. Finally, they needed to find a safe location to place the unit while it waited for another occurrence. Unfortunately, all of this cost and complexity didn't always lead to a satisfactory resolution of the problem.
Today, specialized portable power quality monitors are no longer the only option for troubleshooting voltage transients. Some high-end, permanently installed circuit monitors can also do the job. These circuit monitors serve two roles: One is a highly accurate digital meter (measuring and logging electrical quantities, such as current, voltage, power factor, and energy); the other is a monitor (watching for voltage transients and other power quality problems—alarming on occurrence and capturing waveforms and other data useful for getting to the root cause of a problem).
Although circuit monitors have been able to detect and record steady-state power quality disturbances for some time, the ability to capture sub-cycle transients today requires a great deal more horsepower. Only recently has the technology become available to design compact, affordable circuit monitors that perform high-speed transient detection without compromising accuracy, metering performance, data logging, and other functions.
If you need a circuit monitor capable of detecting and capturing voltage transients, what are the technical requirements you should look for? At minimum, the device should provide the following features:
A fast sample rate. The required sample rate depends on the frequency of the transients you monitor. Since most low- and medium-frequency oscillatory transients occur in the kilohertz frequency, circuit monitors that sample voltage input channels at 256 or 512 samples-per-cycle can sometimes detect and capture oscillatory transients. On the other hand, impulsive transients are often shorter in duration and require much higher sample rates. ANSI and IEC standards require that electronic equipment designed for industrial environments (e.g., programmable controllers, drives, and circuit monitors) undergo a series of lab tests. These tests ensure the equipment can withstand high voltage transients, such as a 1.2 2 50-ms 2,000V impulsive transient.
Technicians select test criteria that represent potential real-world scenarios. A 1.2 2 50-ms 2,000V impulsive transient rises from zero to its peak value of 2,000V in 1.2 ms, then decays to half its peak value in 50 ms. To reliably catch the peak of a 1.2-ms rise-time transient, a monitoring device must sample data at a rate of at least twice per microsecond, which equates to a sample rate of 2 MHz per voltage channel or 33,333 samples-per-cycle.
A monitoring device that samples data at the rate of 512 samples-per-cycle, samples data only once per 32.5 ms. This is adequate for general disturbance monitoring and some oscillatory transients, but it is far too slow to characterize a 1.2 2 50-ms impulsive transient. The fastest transient-detection-capable circuit monitors sample voltage data at 5 MHz per voltage channel or 83,333 samples-per-cycle, which is fast enough to detect even very short impulsive transients.
Ample onboard memory. The circuit monitor should have ample memory to store multiple, high-resolution captured waveforms and the associated rms data while still performing data logging functions (e.g., load profiling, energy usage, etc.) for standard monitoring purposes. It also should store transient waveform data in nonvolatile memory to prevent data loss if the circuit monitor should lose control power.
Sag/swell alarms. The circuit monitor should support high-speed alarms with a detection rate of less than half a cycle for capturing sag/swell disturbances. It's also useful when this alarm operates output relays. Alarms should trigger a waveform capture up to 60 cycles in duration; record the event in the circuit monitor's onboard event log, including a date/time stamp and the rms magnitude of the most extreme value of the sag or swell; and force the circuit monitor to log other pertinent data in its onboard memory.
Compatible software. Although compatible software is not a requirement of the monitor itself, it is necessary to annunciate alarms, view captured waveforms, store waveforms and related data to a disk, and so on. Other useful features include the ability to view harmonic spectral plots and rms plots, zoom on and overlay waveforms, and export waveforms in a standard file format such as Comtrade.
Reaping the Benefits
By installing circuit monitors capable of transient disturbance monitoring at key locations throughout your facility, you'll be prepared to detect and troubleshoot problems before they lead to damaged equipment and production losses. These benefits include:
Lower cost. Cost savings come in several areas. Considering that it costs about $1,500 to $2,000 to add transient monitoring to a standard circuit monitor, you could install 10 transient monitors for the cost of one specialized portable power quality monitor. Additional cost savings come from reducing the need to hire consultants to troubleshoot power quality problems.
Ease of use. Personnel only need to learn one software package to perform basic power monitoring tasks and to alarm upon, view, and analyze disturbances. Circuit monitors eliminate the need for equipment shutdowns to connect portable power quality monitors, and they reduce the safety risk created if someone tries to connect a portable monitor to live equipment.
24-7 monitoring. Permanently installed circuit monitors watch for transient voltage disturbances all day, every day. Even if you shut down the associated PC software, the monitors will retain captured power quality data in nonvolatile memory for later retrieval.
Immediate notification. Since voltage-transient damage is often cumulative, it could be the 100th transient that causes insulation failure and a short circuit. Personnel may not be aware of problems—especially with low-amplitude transients—until a failure occurs. Networked monitors with application software to annunciate alarms will provide immediate notification of voltage transients.
The Bottom Line
Portable power quality monitors still serve a useful purpose. In existing facilities, there may be a need to monitor for troublesome phenomena at a point in the power system where a transient-detection-capable circuit monitor does not exist. But in new facilities, a well-designed power system has transient-detection-capable circuit monitors at the service entrance, on mains, and on other key circuits serving sensitive loads. When they're combined with a high-speed communications network and software to perform analysis and annunciate alarms, they can go a long way toward providing the data necessary to detect and troubleshoot voltage transients before they shut you down. Even Henry Ford would see the sense in that.
Who Spiked My Transient?
A committee of the Institute of Electrical and Electronics Engineers (IEEE) recently reaffirmed a document titled, Recommended Practice for Monitoring Electric Power Quality (Std. 1159-1995). The document has two purposes: to define electromagnetic phenomena that can cause power quality problems; and recommend methods to measure and record those phenomena.
To accomplish the first goal, the document had to present standard definitions for the multiple confusing terms used to describe power quality aberrations.
So what exactly is a transient? Transient is the term for a disturbance lasting less than 1 cycle. Transients are either impulsive or oscillatory in nature. An impulsive transient is normally a single, very high impulse. Lightning is the most common cause of impulsive transients.
An oscillatory transient oscillates at the natural system frequency and normally dies down within a cycle. These transients (sometimes called switching transients) occur when you turn off an inductive or capacitive load, such as a motor or capacitor bank. An oscillatory transient results because the load resists the change. This is similar to water hammer in pipes. Water hammer occurs when you suddenly turn off a rapidly flowing faucet and hear a hammering noise in the pipes. The flowing water resists the change, and the fluid equivalent of an oscillatory transient occurs.
Both types of transients are subdivided into three categories related to the frequencies. Rise time and duration characterize impulsive transients, while frequency and duration characterize oscillatory transients. Low-frequency transients are the types most likely to occur in a power system. Medium-frequency transients are not as common as low-frequency transients, but have much higher amplitudes. You generally observe high-frequency transients only near the source of the disturbance, and the response characteristics of the instrument transformers affect your ability to measure them. A transient-detection-capable circuit monitor should be able to detect low- and medium-frequency impulsive and oscillatory transients.
The effect a transient has on a power system depends on the amplitude of the transient and its frequency. In the case of oscillatory transients, the frequency of the transient often causes problems—although the amplitude can cause problems as well. For example, oscillatory transients can wreak havoc with variable-frequency drives.
In the case of impulsive transients, the amplitude of the transient often causes problems. The damage caused by a transient can be immediate, as is sometimes the case with a lightning strike, or gradual, as in a response to a series of low-amplitude transients. These low-amplitude transients slowly degrade insulation until a short circuit eventually occurs. Instead of immediate damage by one high-amplitude transient, the insulation suffers a slow “death by a thousand cuts.”