Transcript
Introduction
Hello, everyone. Welcome to today’s Ask a Pro session. I’m David Keaton, a firmware engineer over here at PMI, and I’m going to be talking about reverse power flow and distribution feeders.
How Reverse Power Flow Happens
Everything on a radial feeder was designed around one-way flow. Power out from the substation, voltage highest there and dropping along the line. That assumption sits in regulator settings, capacitor placement, protection coordination, and planning models.
Now, add generation at the far end where the design never expected it. On a clear day around noon, a feeder section can produce more than it consumes. The surplus flows back the way it came, and real power at that point goes negative. One house with an oversized array exports onto its secondary. A few on the same transformer and the segment nets out negative. Enough penetration and you see export at the feeder head or back to the substation transformer. It also doesn’t arrive evenly, so one phase can be exporting while the other two are still importing.
The shape is the duck curve. Solar ramps up through the morning, net demand sags into a midday trough, then climbs steeply in the evening. At the system level, that curve flattens. At the feeder section, it can cross zero, so a substation look can look unremarkable while a section downstream is exporting.
When Reverse Power Flow Becomes a Problem
Reverse power flow isn’t inherently harmful. Modern planning increasingly assumes some bidirectional flow, and a feeder that exports for a couple of hours a day isn’t in trouble. Problems come when export becomes frequent, widespread, or large on infrastructure set up for one direction. The question is how much, how often, and what it’s doing.
Voltage Rise and Impedance
The impedance that produces voltage drop on the way out, conductor, service drop, transformer, produces voltage rise when you reverse the current through it. It matters more here than on transmission because feeder resistance is significant, so real power export alone moves voltage noticeably.
The field pattern is consistent. Solar climbs, net real power falls through zero, surface voltage rises over the same window, and the day’s highest voltage tends to line up with its deepest export. On a one-twenty volt service, sustained voltage above one-twenty-six volts may exceed the ANSI C84.1 Range A limit.
Effects on Regulation and Protection Equipment
Load tap changers, line regulators, and switched capacitor banks all assume voltage falls with distance from the substation. Reverse flow inverts that profile. A regulator may no longer see the condition it was set to correct, and the capacitor bank installed to hold voltage under heavy load may push an already high voltage higher.
That shows up as more operations, extra tap changes, capacitor switching, sometimes hunting. It’s mechanical wear on equipment with a finite number of operations, so you pay in shortened life and maintenance rather than outages, which is why it often goes unnoticed.
Radial coordination assumes the substation is the dominant source of bulk current. Large synchronous distributed generation can feed a fault locally, reducing what the upstream relay sees, contributing to protection blinding, which can delay detection. Inverter-based DER contributes less individually, but aggregate penetration still affects protection studies. Sympathetic tripping is the other case. DER on a healthy feeder contributes to the fault on an adjacent feeder, and the healthy feeder trips.
Normally, conductors and transformers don’t care about direction, but protection, regulation, metering, and control schemes may not be configured for frequent reverse flow. Export raises conductor loading, changes transformer loading profiles, and shifts regulation requirements. Hosting capacity studies estimate all that, but they’re models. Measurements tell you whether those conditions actually occur, how often, and how severely.
Why Reverse Flow Is Hard to See
Reverse flow is hard to see for a structural reason. Monitoring is concentrated where the assets are expensive, substations and major equipment. Export starts where the DER is, at the edge of the system. Instrumentation and the thing being observed are at opposite ends of the feeder.
Export begins on services, secondaries, and laterals before it reaches the substation. SCADA covers a limited number of points, and some record magnitude without direction. Five- and fifteen-minute averages smooth short export events out, and a three-phase aggregate can look ordinary while one phase exports underneath it. Add solar variability on a partly cloudy day, and you get export that starts and finishes between two reporting intervals.
So what’s needed is data that shows direction at a location far enough downstream to matter at a resolution fast enough to catch it.
Three Signatures of Reverse Power Flow
- Real power crossing below zero is the primary indicator. Under the load convention, positive kilowatts is import, negative is export. A verified transition from positive to negative is direct evidence a location is sending power upstream.
- Voltage rising during export is corroboration. Record voltage on the same timeline as real power, and you can see whether they move together. When the day’s voltage peak sits on its deepest export, that’s a reasonable case the DER is contributing. How much depends on impedance, load, regulation, and inverter controls, so don’t expect a fixed ratio.
- Power factor going unstable near the crossover is a tell rather than a problem. As real power approaches zero, it stops carrying much meaning and small changes swing it hard. Reactive power is much more informative here, especially with smart inverters absorbing VARs during high export as part of VAR control.
Voltage rises for all sorts of reasons, and power factor near zero is hard to interpret. Real power is the primary indicator. Min, average, and max within each interval matter here more than usual. A passing cloud can flip a section between export and import in minutes, and batteries move faster. Averages hide those extremes.
Recording Example
Here we have a recording that actually illustrates those three measures in particular. This is at a service entrance. As we notice overnight, real power is positive, and as people start getting ready to leave for work, midday sag starts to happen and real power goes negative. Through midday, it stays negative, then climbs back to zero into the evening peak.
Voltage here rose from one-twenty-four to one-twenty-five to one-twenty-six and stayed within that region while real power was exporting. And as you notice, the peaks coincide with the export peaks as well. Power factor, as it crosses over zero, also tends to dip quite heavily. But otherwise, power factor is near one.
Monitoring Location
A substation monitor tells you whether the feeder exports, not where it starts. A service or transformer monitor catches localized export earlier. One at a regulator or capacitor bank tells you whether export is affecting voltage control equipment.
Reverse power flow is becoming an operating condition rather than an exception. On most DER-rich feeders, whether it happens isn’t the question. Where, how often, and what impact are. It’s not inherently harmful, but it does require awareness. Monitoring location is key. And again, crossing zero with real power is the primary indicator that export is happening.
Contact Information
Thank you for attending today’s webinar. If you have any questions, you can reach us anytime by calling 800-296-4120 or emailing support@powermonitors.com. Thank you, and have a great day.
