Semiconductor Plant Energy Management: Meter First

Start semiconductor plant energy management by metering the incoming electricity supply and major distribution feeders, then add utility meters according to the plant's process. This lets operators trace a demand peak or unexplained consumption to the equipment responsible.

At 2:05 p.m., for example, a chiller restarts while production tools and air compressors are already running. The overlap lasts 30 minutes and may set the month's maximum demand. Operators need to see both the event at the incoming supply and the systems that caused it.

Many plants still rely on the Tenaga Nasional Berhad (TNB) incoming meter and monthly bill. These show the site total, but avoidable idle load, poor scheduling and utility faults can remain hidden within it.

Electrical and electronics (E&E) products are Malaysia's largest export product group. The Malaysia External Trade Development Corporation reported record E&E exports of RM601.18 billion in 2024, accounting for 39.9% of Malaysia's exports. In the northern manufacturing corridor, Infineon opened the first phase of its Kulim 3 silicon carbide fab in August 2024, with a second phase of up to five billion euros planned.

In short

• Start metering at the incoming electricity supply and major distribution feeders, then add utility meters that match the plant's specific process.
• Overlapping equipment starts, such as a chiller restart while production tools and compressors are already running, are the usual cause of an unexplained demand peak.
• Comparing energy use within the same production state, not month to month, is what actually isolates waste in a semiconductor plant.
• A metering plan that traces a peak back to the responsible equipment turns troubleshooting from a guess into a lookup.

Semiconductor plant energy management must match the plant type

Choose utility meters according to the production process. A wafer fabrication plant, an outsourced semiconductor assembly and test (OSAT) plant, and a general electronics factory use different combinations of utilities.

Use this table to choose where to look first. It does not show each utility's share of energy use: equipment, product mix, cleanroom class and operating hours can change the order. Another site's energy breakdown may not fit your plant.

A fab may need deionised (DI) or ultrapure water (UPW) metering early in its plan. A general electronics plant may get more value from line, oven and compressor meters. An OSAT site often sits between these profiles.

Build the metering plan from the incoming supply to equipment

Start with a reliable site total, then add meters that explain where that energy goes. Each level should account for a useful part of the level above it.

Electrical meters record power in kilowatts (kW) and energy use over time in kilowatt-hours (kWh). Use these readings alongside utility measurements and production records:

Check the site total against the submeters

Use the TNB incoming supply and main switchboard as the reference, recording active power, energy and power factor. Power factor describes how effectively the electrical supply is used to deliver active power. Add meters at the major distribution boards or transformers below this level.

Compare the sum of these submeters with the incoming total, allowing for known losses and unmetered loads. A large unexplained difference means the meter coverage is incomplete or a meter needs checking.

Measure cooling delivered as well as electricity used

Process cooling is often a major site load, so meter the whole chilled-water system early. Include pumps and cooling towers as well as chillers. Pair electrical meters with flow and temperature sensors to compare electricity use with the cooling delivered.

For water without glycol, estimate cooling load from the flow in litres per second (L/s) and the difference between return and supply temperatures. The symbol ΔT represents that temperature difference in degrees Celsius:

$$ \text{Cooling load (kW)} \approx 4.186 \times \text{flow (L/s)} \times \Delta T\ (^\circ\text{C}) $$

A 100 L/s flow with a 5°C temperature difference delivers about \(4.186 \times 100 \times 5 = 2{,}093\) kW of cooling. Confirm sensor accuracy and fluid properties before using this estimate for formal measurement and verification.

Keep cleanroom conditions separate from comfort cooling

Track cleanroom supply and return conditions, fan power, cooling, reheat, humidity and operating schedule separately from general comfort cooling. Air-change and pressure settings affect contamination control, so changes need process, quality and safety approval. Cleanroom optimisation needs its own engineering study.

Compare utility power with flow and production

Meter each major compressed-air, vacuum, exhaust and abatement system. Pressure in the main header alone cannot show electrical cost; compare input power with flow and production state. Our guide to compressed-air energy efficiency in Malaysia covers leak checks, pressure and controls in more detail.

Map water from the plant inlet through treatment to high-use areas. For DI or UPW systems, measure and balance feed, product, reject and make-up flows. This helps the team investigate changes in water recovery or unexpected night flow.

Link every utility trend to what production was doing at the time. A fall in energy use could mean better control, lower output or a stopped line.

Manage maximum demand by controlling which loads run together

Operators can reduce the billing maximum only by changing the loads that set it. Under Regulatory Period 4 (RP4), the TNB tariff structure effective 1 July 2025, medium-voltage demand charges are RM89.27 per kW for General supply and RM97.06 per kW for Time of Use (ToU) supply.

The charge uses the month's single highest 30-minute demand interval. TNB defines the ToU peak period as 2 p.m. to 10 p.m. on weekdays. Check the current schedule on TNB's tariff page before making an investment decision.

Consider a ToU plant whose billing maximum includes 180 kW of avoidable loads running together. A standby chiller may have started while compressors were loaded and non-critical charging was active. Multiplying that load by the ToU demand rate gives the potential demand-charge difference:

$$ \text{Demand-charge difference}=180\ \text{kW}\times\text{RM }97.06/\text{kW}=\text{RM }17{,}470.80 $$

The amount at stake for that month is RM17,470.80. It covers demand charges only, and the saving is not guaranteed. The plant saves this amount only if the action reduces the actual billing maximum; moving the load to another already-high interval may create a new peak.

Prepare the operating response before an alert arrives:

1. Set an alert below the site's agreed demand limit.
2. Agree which loads operators may delay and how they may adjust chiller staging within process limits.
3. When the alert triggers, follow that response list and avoid starting large systems together.

Every response must protect output, product quality and safety. See the fuller guide on how to cut TNB maximum-demand charges.

Compare energy use within the same production state

Link energy readings to production states so the team can understand why consumption changed. Use labels operators already know: idle, running, ramp, changeover, maintenance and shutdown.

For periods with output, calculate specific energy: the energy used for each good unit, panel or lot produced. Divide energy use by good output over the same period:

$$ \text{Specific energy}=\frac{\text{kWh in the period}}{\text{good units, panels or lots in the period}} $$

A line that uses 3,600 kWh to make 1,200 good units has a specific energy of \(3.0\) kWh per good unit. Keep idle energy separate. Dividing 500 kWh by zero output during an idle shift gives no useful result, so report the idle baseline in kW and total idle kWh instead.

Compare periods with similar product mix, yield, ambient conditions and planned maintenance. Each can change the result, so higher kWh per lot needs investigation before it can be judged as poor performance.

These comparisons support a wider factory energy-monitoring plan and provide a defensible baseline for checking changes to chiller sequencing, pressure or operating schedules.

CobiNeural connects meters, utilities and operations

CobiNeural can collect equipment-level energy and water data so teams can view utility readings alongside production states. It can monitor maximum demand, issue alerts and display dashboards. It can run on its own or work with an existing building management system (BMS), programmable logic controller (PLC) or supervisory control and data acquisition (SCADA) system.

The design depends on the site. Engineers can reuse reliable existing signals and add submeters or sensors where data is missing. The platform also covers indoor air quality, chilled water, billing, measurement and verification, reporting workflows under the Energy Efficiency and Conservation Act 2024 (EECA), and energy-management reporting for International Organization for Standardization (ISO) 50001.

Engineers still need to check meter locations, current transformer ratios, naming, time alignment and data quality. Current transformer ratios let meters convert the measured signal into the correct current reading. Start with the meters that explain the largest loads, and add detail when it supports an operating decision.

For the full mechanics of a TNB bill, see our free course Understanding Your Electricity Bill.

To map the first metering points for your Malaysian plant, request a practical CobiNeural walkthrough.