Will your production line survive load shedding?

Who this is for

Plant owners, engineering managers, and maintenance leads running production lines in South Africa (or any grid-constrained African market) who lose output — and sometimes equipment — every time the power goes. This is a working guide to what actually breaks during load shedding and how to spend the minimum to fix the most.

Load shedding does two different kinds of harm

Owners think of load shedding as lost hours. That is only half of it. The two costs are:

• Downtime — the line is off while the power is off, and then loses more time on restart, cleaning, and re-stabilising. Predictable and visible.
• Damage — the uncontrolled stop and, crucially, the spike and brown-out when power returns, which quietly degrade and sometimes destroy controls, drives, and motors. Invisible until something fails.

The second cost is the one most plants underestimate. The moment the grid comes back is often more dangerous to your electronics than the outage itself.

What actually breaks when the power drops

• PLC and HMI state. A hard power cut can corrupt PLC memory, recipe data, and HMI logs, and leave the line in an unknown state that needs a manual reset. Modern OT is now so integrated that a dirty power-down can take down the controls, not just stop the motors.
• Work in progress. Product mid-process — a bottle being filled, a mould mid-cycle, a batch mid-cook — is usually scrapped. On food and beverage lines, a stalled hot or aseptic process can mean a full clean-down.
• Drives and motors. VSDs and servo drives dislike sudden loss and the inrush on restart. Repeated dirty cycles shorten their life.
• Mechanical stress. A line that coasts to a stop with no controlled sequence can jam, misalign, or shock-load gearboxes and conveyors.
• The return surge. When the grid is restored, voltage spikes and dips can exceed what the equipment tolerates — the single most common cause of load-shedding equipment failure.

The protection layers, cheapest and most important first

Layer 1 — Controls UPS (do this first, always)

Put the PLC, HMI, SCADA PC, network switches, and key instrumentation on an uninterruptible power supply. The control system draws very little power, so a modest UPS lets it ride through short cuts, hold its state, and either keep running (if the rest of the line is on a generator) or shut down cleanly and restart in a known state. This is the highest-value, lowest-cost intervention and it directly solves the "PLC died during load shedding" problem.

Indicative cost: ZAR 30 000–120 000 per line depending on controls load and runtime required.

Layer 2 — Safe shutdown sequence

The controls UPS buys you seconds to minutes. Use them: a programmed controlled-shutdown sequence brings the line down in the correct order on power loss — stopping infeed first, clearing or holding product safely, parking actuators, and logging the state — so restart is clean and nothing is damaged or jammed. This is engineering, not hardware, and it is often missing even on otherwise good lines.

Layer 3 — Surge and undervoltage protection

Protect against the return event. Surge protection devices, undervoltage release on critical contactors, and phase/voltage monitoring relays prevent the line from trying to restart into bad power. This is what stops the spike-on-return from killing your drives. Relatively cheap, and it belongs in the MCC.

Indicative cost: ZAR 15 000–60 000 per line, integrated into the MCC.

Layer 4 — Standby generation for critical loads

Now — and only now — decide how much production to keep running. The mistake is sizing a generator for the whole line by default. Size it for the loads that genuinely cannot stop: cold chain and freezing, CIP and hygiene-critical process steps, anything where a stop means spoilage or a full clean-down. A generator with an automatic transfer switch (ATS) keeps those loads alive and lets the rest stop cleanly.

Indicative cost: from ZAR 250 000 for a critical-load genset + ATS to several million for full-line backup, plus diesel and maintenance.

Layer 5 — Full-line or hybrid backup (only if the economics justify it)

Backing up the entire line — large generator, or a solar-plus-battery hybrid — only makes sense when the downtime cost clearly exceeds the capital and running cost. Diesel for a full line is expensive; solar-plus-battery can shift the maths for daytime operations (covered in our companion guide on running a production line on solar). The decision is a number, not a reflex.

What each layer protects — at a glance

Operational tactics that cost nothing

• Schedule around the timetable. Load shedding in South Africa is broadly scheduled. Plan changeovers, cleaning, and maintenance into the known off-slots, and run production hard when power is on.
• Double-shift recovery. Where demand allows, run double shifts during powered windows to recover the output lost in the dark.
• Stage long processes. Don't start a process you can't finish before a known slot — sequence hot-fill, cooks, and batches around it.
• Brief operators on the restart. A known, trained restart sequence turns a 30-minute scramble into a 5-minute procedure.

Tie the spend to the real loss

The right backup spend is the one matched to your actual downtime cost. A visibility layer that logs every power event and the production lost against it turns "load shedding is killing us" into a number you can budget against — and often shows that the controls layer plus a critical-load generator captures most of the loss for a fraction of full-line backup. See measuring OEE on an old line and total cost of ownership.

What CISH does

We assess a line's load-shedding exposure, then engineer the layers in priority order — controls UPS, safe-shutdown logic, MCC surge/undervoltage protection, and critical-load standby with ATS — as part of a line upgrade or maintenance engagement. New lines are designed load-shedding-aware from the start; see also when to upgrade your PLC and our SADC plant case study, where grid instability drove the whole design.

Frequently asked questions