Inside a control cabinet, the power supply that converts incoming AC to the low-voltage DC that PLCs and RTUs run on is a quiet single point of failure, because if that one converter dies, everything downstream of it goes dark. Redundant power supply design removes that risk by running two supplies in parallel through a module that lets either one carry the load, so a failed supply never interrupts the equipment. This guide covers how dual power supplies and a redundancy or diode-OR module work together, what dual DC buses and dual AC feeds add, and how cabinet-level power redundancy pairs with a UPS. This is hardware inside the panel, distinct from facility power sources like a standby generator.
Redundant Power Supply in one line: A redundant power supply is an arrangement in which two or more power supply units feed a load in parallel through a redundancy module, so that if one supply fails the others continue powering the equipment with no interruption. In a SCADA cabinet this usually means two DC power supplies whose outputs are combined by a diode-OR or redundancy module that isolates a failed unit and lets the healthy one carry the full load. It removes the power converter as a single point of failure at the panel level and is a common, low-cost step toward a fault-tolerant control system.
The core of a redundant power arrangement is a pair of power supplies each sized to carry the full DC load on its own, combined through a device that lets either one feed the equipment without back-feeding the other. The simplest such device is a diode-OR: each supply feeds through a diode into a common output bus, and the diodes prevent current from flowing backward into a failed or lower-voltage supply. If one supply drops out, its diode simply stops conducting and the healthy supply carries the whole load through its own diode, with the transition happening in the electronics rather than through any switch that has to move. Purpose-built redundancy modules refine this idea, often using low-loss active components in place of plain diodes and adding monitoring so a failed supply raises an alarm instead of failing silently.
The key design rule is that each supply must be rated to carry the entire load by itself, not half of it, because the whole point is that one can fail and the survivor must cope alone. Running two supplies that each carry half the load and calling it redundant is a false economy, since losing one then overloads the other. Correctly sized, the arrangement means a power supply can fail, be flagged, and be replaced, ideally while the system runs, without the controllers or field devices ever seeing a dip. That single change removes one of the most overlooked single points of failure in a control panel, because the DC converter is a component that does fail with age and heat, yet is easy to forget when it has been humming reliably for years.
Redundancy at the supply itself is only the first layer, because the power still has to reach the equipment along wiring, and shared wiring reintroduces a single point of failure. A more thorough design uses dual DC buses: two separate distribution rails, each fed by its own supply, so that critical devices with two power inputs can draw from both buses independently. A redundant PLC or a managed switch with dual power inputs then survives the loss of an entire bus, not merely the loss of one supply, because the failure of a rail, a terminal, or a wire on one side leaves the other side intact. This pushes the redundancy past the converter and into the distribution.
The same logic extends upstream to the AC side. Two power supplies fed from the same single AC circuit are only protected against a supply failing, not against that circuit tripping, so a fully redundant cabinet feeds its two supplies from two independent AC sources, ideally on separate breakers or even separate distribution panels. Now a tripped breaker, a failed feed, or maintenance on one circuit leaves the cabinet fully powered through the other. Tracing the power path from the AC feed through the supplies, the redundancy module, the DC buses, and into each device is exactly the single-point-of-failure hunt applied to power: every point where the two paths merge into one is a place the redundancy quietly stops working, and a well-built cabinet keeps the two paths separate as far as the equipment inputs allow.
Redundant supplies protect against a component failing, but not against the incoming power itself going away, which is where an uninterruptible power supply comes in and why the two are complementary rather than alternatives. A UPS rides through a utility outage on battery for a period, keeping the whole cabinet alive long enough for a generator to start or for the system to shut down cleanly, while redundant supplies keep the cabinet alive through the death of an internal converter. A well-designed control panel uses both: the UPS defends against loss of source power, the redundant supplies and dual buses defend against internal failures, and together they cover a much wider range of faults than either alone. Facility-level backup such as a standby generator sits further upstream again, handling extended outages that outlast the battery.
This hardware layer matters to a cloud SCADA deployment because the field devices at each site depend on it, and no amount of central redundancy helps if a site's gateway loses power. A remote well pad, pump station, or substation whose controller and communications gateway share a single power supply has a SPOF that will eventually take the site offline, so redundant supplies and a UPS at the cabinet keep the site sending data through both internal faults and short outages. On a platform such as Merobix, the central supervision is already made resilient across availability zones, so the practical reliability limit for a given site becomes the power and communications hardware inside its cabinet. Getting the cabinet power right, redundant supplies feeding critical devices through dual buses, backed by a UPS, is what lets a remote site keep reporting through the small failures that would otherwise create silent blind spots.
Two or more power supplies, each rated to carry the full load alone, feed the equipment in parallel through a redundancy or diode-OR module. The module lets any healthy supply power the load while isolating a failed one, so if a supply dies its share is picked up instantly by the others with no interruption. Good modules also raise an alarm on the failed unit so it can be replaced, ideally while the system keeps running.
No, they protect against different failures and are best used together. A redundant power supply protects against one of the power converters inside the cabinet failing, keeping equipment powered through an internal fault. A UPS protects against the incoming power itself going away, riding through a utility outage on battery. A UPS does nothing if a converter dies, and redundant supplies do nothing if the source power fails, so a robust cabinet uses both.
Because the purpose of redundancy is to survive the loss of one supply, and after that loss the surviving supply must carry everything on its own. If two supplies each carry only half the load, losing one overloads the other and the whole arrangement fails, defeating the redundancy. Sizing each unit for the entire load is what guarantees that a single failure leaves the equipment fully powered.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.