Transcript
Introduction to Voltage Swells
I wanna go over the latest white paper called Understanding Voltage Swells. This white paper goes over what voltage swells are and some of their causes, as well as how to spot them and have PMI devices record.
One of the main items that plague swell detection is that they occur less often than sags and tend to be far harder to detect. Most customers don’t even know they happen unless their lights happen to get brighter for a very brief period of time.
IEEE 1159 Definition and Classification
Voltage swells are defined by IEEE 1159 as an increase in RMS voltage from one point one per unit to one point eight per unit, with one point O per unit being your nominal voltage. For instance, a hundred and twenty volts on a regular residential outlet.
Voltage swells are different than transients or steady overvoltage conditions due to how long they last. Transients, for example, last less than a half a cycle, and steady overvoltage conditions last longer than a minute. This puts swells into a small time frame to be classified by, with swells falling into three distinct types.
- Instantaneous swells with a duration of one half to thirty cycles and typically being one point one to one point eight per unit.
- Momentary swells are thirty cycles to thirty seconds, one point one to one point four per unit being typical.
- Temporary swells, which are thirty seconds to one minute, typically with a magnitude of one point one to one point two per unit.
We can see these plotted out via the IEEE 1159 classification chart seen in figure one here. We can see that from half cycle to thirty cycles is considered instantaneous, thirty cycles to three seconds momentary, and three seconds to one minute is temporary.
Causes of Voltage Swells
Single-Line Ground Faults
The largest cause of most voltage swells seen is a single-line ground fault occurring on a phase different than the phases a swell is seen on. Most multi-grounded systems can see a swell from a single-line ground fault from one point one per unit to one point four per unit.
We’ll look at figure two here. We can see how the neutral ground point shifts towards the ground fault point away from the overlay neutral ground on the substation seen in gray. Multi-grounded systems are where the system’s grounded at multiple points throughout the distribution.
Now, on ungrounded or resonant ground systems, we can see an upwards of one point seven three per unit increase in voltage at the top of the IEEE’s classification criteria. Figure three here shows the phasors’ reactions in an ungrounded system with a much larger phasor change as phase A is now considered the system’s ground versus the station’s neutral.
Looking here, we can see the difference between the multi-grounded system and the ungrounded system, with the ungrounded having a very, very large phasor change compared to the multi-grounded system.
One way to visualize this is the example listed in the paper where a branch contacts phase A of a twelve point four seven kilovolt multi-grounded feeder. During the zero point four seconds it takes the recloser to open, phase A sees a deep sag, and phases B and C see a one point three per unit increase, causing a hundred and twenty-volt receptacle to climb all the way to a hundred and fifty-six volts for twenty-four cycles.
Other Causes of Voltage Swells
Faults tend to be the most common causes of voltage swells, but some other causes are:
- Load rejection. This is where a large load switching off causes a voltage drop that was present but disappeared, resulting in a localized voltage step up until a regulator or tap changer can set the voltage back down.
- Capacitor bank switching. Capacitor bank energizing creates a voltage rise, typically a small bit above one point one per unit under light load before controls are spot.
- Recloser operations. Reclosers opening will cause a voltage swell, as mentioned in the above example.
One way to tell a swell seen at a residential suffers interest was caused by a recloser is both monitored phases swell the same amount versus one swelling and another sagging. If one phase swells and one sags, typically the cause is a loose neutral.
Sags Versus Swells
Both sags and swells can be seen in most recordings. A general rule of thumb is that sags are usually caused by something turning on or falling down, and swells are caused by something turning off or letting go.
One important item to remember is that swells can be more damaging to equipment than sags. This is due to the fact sags starve equipment for voltage, whereas a swell force-feeds it.
The tree branch example of one point three per unit increase for zero point four seconds can be referenced against the ITIC curve in figure four. So again, figure four, we can see the one point three per unit, zero point four second duration plotted on the chart, and it shows we are well into the prohibited region for equipment to operate. This results in a very high likelihood of the NA equipment being fed by this swell failing.
Surge Suppressors and Protection
For sensitive equipment, surge suppressors are typically used to help protect them, be it a portable unit or a system designed to be installed into the building’s panel. These devices have MOVs, which are metal oxide varistors, installed that clamps higher voltages to nominal to help ensure safe operations of the devices that they’re actually used on.
One of the largest downsides of these devices is that the components degrade silently over time. Exposure to longer overvoltage conditions degrades the components internally and can cause them to thermally fail.
Setting Up a Recorder to Catch Swells
Catching a swell is more difficult than other issues due to their brief and infrequent nature. Three important items to remember when setting up a recorder to catch these:
- The voltage capture threshold should be set to ten percent off of nominal. This matches the IEEE 1159 definition for a swell and keeps the recorder from triggering on normal regulation events.
- Be sure to record every phase in the circuit. This ensures if one phase sees an issue, we can see the effect on all other phases in the system.
- On split-phase services, ensure both phases are set to monitor and trigger at the same threshold. This is also a good rule of thumb for three-phase circuits as well.
Analyzing Swells with PQ Canvas and Merlin
After a recorder’s been in the field and has captured suspected swells, downloading the recording and uploading it into PQ Canvas allows the user to investigate all the waveform captures being recorded. Scrolling down here to figure five, we can see a waveform character of a swell in recording, with the swell being shown by the overlay of black lines in the figure. This specific swell was about one point one five per unit.
Finding the exact waveform that has what you’re looking for can be tedious, though. With PMI’s Merlin, we’re able to see the meat and potatoes of recording quicker and presented directly. Figure six here shows the power quality report pane of the recording we looked at the waveform in figure five. Note that it is plainly showing we have a rare deep sag and switching-driven swells present in this recording.
If we scroll down here to figure seven, we can see the waveform report that Merlin generates, and with figure eight showing the voltage swell waveform that was present. This allowed us to scroll through the waveforms to see what each one was in plain English and evaluate the waveform we wanted at the click of a button.
Conclusion
Though more elusive than sags, swells are just as detrimental, if not more, to electronics and other equipment being monitored by the affected lines. Utilizing a PQ report such as those offered by us here at PMI allows the user to find these issues quickly and easily.
Appreciate everyone taking the time and coming out, and we’ll see you all in the next one.