E07 The Electricity System Supplying Malaysia Data Centres

Key Takeaways

• The Electricity System Supplying Malaysia Data Centres is a hierarchy of interconnected networks, moving electricity from generation through 500 kV, 275 kV and 132 kV transmission networks to lower-voltage distribution and consumers.
• High voltage reduces transmission losses because transmitting the same power at higher voltage requires lower current, while resistive losses increase with the square of current.
• Large data centres connect at high voltage because of their electricity demand, not simply because they are data centres. TNB’s supply guidelines relate the required supply voltage to maximum demand and system capability.
• A major data-centre connection can involve more than a new cable, potentially requiring substations, transformers, connection circuits and reinforcement of existing network infrastructure.
• Location matters: the same data-centre load can present very different connection requirements depending on available grid capacity, network configuration and system-security considerations.
• The electricity system serving a data centre combines existing, dedicated and reinforced infrastructure. Understanding these physical relationships is essential before examining how the resulting infrastructure costs are allocated.

Why Electricity Infrastructure Matters

A data centre may look like a building, but its ability to operate depends on infrastructure extending far beyond its physical boundaries.

Among all the resources required by a modern data centre, electricity is one of the most fundamental. Servers, storage systems, networking equipment, cooling systems, security systems and other critical infrastructure ultimately depend on a continuous electrical supply.

This means that understanding a data centre also requires understanding the electricity system that supplies it.

The electricity consumed by a data centre does not originate at the boundary of the site. It is produced somewhere within the wider electricity system, transported through transmission networks, transformed between different voltage levels, routed through substations and distribution infrastructure, and eventually delivered to the customer’s connection point.

In simplified form, the external electricity system can be represented as:

Power generation → transmission network → transmission substations → distribution network → customer connection → data centre

This article focuses on everything up to the point where electricity is supplied to the data centre.

It does not examine the electrical distribution system inside the data centre itself. That internal system—covering equipment such as switchgear, transformers, UPS systems, generators and server distribution—was introduced in →E03 — Why Data Centres Are Built This Way and is outside the scope of this article.

The distinction is important because the electricity infrastructure outside the data centre is part of a much larger system shared, in different ways, with other electricity consumers.

A transmission line may serve many customers.

A transmission substation may supply an entire area.

A distribution network may serve industrial facilities, commercial buildings and residential developments.

A power station may generate electricity that ultimately serves millions of consumers.

The data centre is therefore connected to an electricity ecosystem rather than operating as an electrically isolated facility.

This also helps explain why the location and scale of a data centre can have implications beyond the boundaries of its own site.

A facility requiring hundreds of megawatts of electrical capacity is not simply another building connecting to the grid. Its connection may require significant consideration of transmission capacity, substations, network configuration, system reliability and future demand.

TNB itself describes electricity network planning as a long-term process involving load forecasts, capacity planning and the development of new or expanded transmission and distribution infrastructure. Its current planning information refers to 20-year demand forecasting and progressively shorter planning horizons for different parts of the distribution system.

This wider perspective connects directly with earlier MDCO articles.

→E01 — What Is a Data Centre introduced the data centre as a piece of digital infrastructure rather than simply a large building.

→E03 — Why Data Centres Are Built This Way explained why modern facilities require extensive electrical, cooling, backup and resilience systems.

→A06 — Loudoun County Case Study — When Infrastructure Becomes Background Noise illustrated another aspect of the same principle: infrastructure built to support large concentrations of data centres can have consequences beyond the boundaries of individual facilities.

The present article now follows one of the most fundamental inputs identified in →E02 — How Data Centres Interact with the World—electricity—backwards from the data centre to the wider electricity system.

The objective is not to turn the reader into a power-system engineer.

It is to establish a basic understanding of the physical system before later MDCO articles examine questions of supply applications, tariffs, infrastructure costs and cost allocation.

MDCO Insight: A data centre may occupy a particular site, but the electricity that powers it comes from a much larger infrastructure system extending far beyond its boundaries.

From Power Plant to Data Centre

To understand the relationship between a data centre and the electricity network, it is useful to follow electricity through the system.

A simplified representation is:

Generation → high-voltage transmission → grid substation → lower-voltage network → customer connection → data centre

Each stage performs a different function.

Generation

Electricity begins at generating facilities.

Power stations convert different forms of primary energy into electrical energy. Depending on the technology, this may involve natural gas, coal, hydroelectric power, solar energy or other sources.

The generator produces electricity at a voltage suitable for the generating station, but this is not normally the voltage used to transport large quantities of electricity across long distances.

The electricity therefore enters another important part of the system: the transmission network.

Transmission

Transmission networks transport large quantities of electricity across substantial distances.

They operate at high voltages because transporting large amounts of electrical power at lower voltages would require much higher currents and would result in greater electrical losses.

Malaysia’s electricity system includes a high-voltage transmission network operating principally at 500 kV, 275 kV and 132 kV. TNB identifies these as its transmission voltage networks.

Transmission is therefore not simply a collection of wires carrying electricity from power stations to customers.

It is a highly interconnected system designed to move power between generation sources, substations and major load centres while maintaining acceptable levels of reliability and system performance.

Transmission Substations

Electricity cannot remain at one voltage throughout its journey.

Transmission substations provide the points at which different parts of the network are interconnected and where voltage is transformed.

For example, TNB’s published connection guidelines describe a Transmission Main Intake (Pencawang Masuk Utama, or PMU) as an interconnection point involving 132 kV or 275 kV, with standard transformations including:

• 275/132 kV;
• 132/33 kV; and
• 132/11 kV.

A substation therefore performs much more than a simple switching function. It can provide transformation between voltage levels, switching, protection and controlled interconnection between different parts of the electricity system.

Distribution

After electricity has travelled through the high-voltage transmission system, it is progressively transformed to lower voltages for delivery to different categories of consumers.

TNB identifies 33 kV, 11 kV and 400/230 V as distribution voltages. In certain parts of Johor and Perak, 22 kV and 6.6 kV are also used.

The distribution network therefore brings electricity closer to the point where it will actually be consumed.

At lower voltage levels, the network can serve commercial premises, industrial facilities, residential developments and other consumers.

Customer Connection

The final stage is the connection between the electricity network and the customer.

This is where an important distinction appears.

Not every customer connects to the electricity network at the same voltage.

A typical household ultimately receives electricity at low voltage.

A large industrial facility may receive electricity at a much higher voltage.

A very large data centre may require a connection directly to the high-voltage transmission system.

TNB’s published supply information states that supply may be provided at 275 kV, 132 kV, 33 kV, 22 kV, 11 kV, 6.6 kV or 400/230 V, depending on the applicable supply arrangement and the customer’s load requirements.

This means that the electricity journey is not identical for every consumer.

A household does not necessarily pass through the same network configuration as a major industrial customer.

Similarly, a large data centre connected at 132 kV or 275 kV has a very different physical relationship with the electricity system from a small commercial building supplied at low voltage.

This difference will become particularly important when considering how electricity infrastructure is planned and how the associated costs are allocated.

The Grid Is a Hierarchy, Not a Single Wire

It is tempting to imagine the electricity system as a simple line:

Power station → customer

The actual system is much more complex.

There are multiple generating facilities, transmission circuits, substations, transformers, distribution networks and customer connections. Electricity flows through an interconnected system rather than along a dedicated path from one particular power station to one particular customer.

This is particularly important when discussing large electricity consumers.

A data centre may have a specific physical connection, but the electricity system supporting that connection may contain infrastructure that serves a much wider area.

That distinction between a customer-specific connection and the wider electricity network will become central to the later MDCO series examining who pays for electricity infrastructure.

MDCO Insight: Electricity reaches a data centre through a hierarchy of interconnected networks, and a customer’s physical connection is only one part of the wider electricity system that supports it.

Why Electricity Is Transmitted at High Voltage

The use of high voltage in electricity transmission is not arbitrary.

It follows directly from fundamental electrical physics.

The relationship between electrical power, voltage and current can be expressed as:

P = VI

where:

• P is electrical power;
• V is voltage; and
• I is current.

For a given amount of power, the relationship can be rearranged as:

I =P \ V

This tells us something fundamental.

If the amount of power being transmitted remains the same, increasing the voltage reduces the current.

Why does that matter?

Because electricity flowing through a conductor encounters resistance.

The electrical power converted into heat as a result of that resistance is approximately:

P_loss = I²R

where:

• P_loss is the resistive power loss;
• I is the current; and
• R is the resistance of the conductor.

The important feature of this equation is the square of the current.

This means that reducing current has a disproportionately large effect on resistive losses.

Suppose, conceptually, that the voltage used to transmit a particular amount of power is increased by a factor of ten.

From:

I = P \ V

the current becomes one-tenth of its original value.

Now consider the loss:

P_loss = I²R

If the current becomes one-tenth, the resistive loss therefore becomes approximately one-hundredth of the original value, assuming the conductor resistance remains unchanged.

In simple terms:

10 times the voltage → 1/10 of the current → approximately 1/100 of the resistive loss.

This is the fundamental physical reason electricity is transported at high voltage.

Why Not Simply Use Extremely High Voltage Everywhere?

If higher voltage reduces transmission losses, it might appear that the logical solution would be to increase voltage as much as possible.

In practice, it is not that simple.

Higher voltage introduces its own engineering requirements.

Electrical equipment must withstand higher electrical stresses.

Greater clearances may be required.

Insulation systems become more demanding.

Switchgear, circuit breakers, protection systems and substations must be designed for the relevant voltage level.

Transmission lines also require appropriate physical clearances and insulation arrangements.

The cost and complexity of the infrastructure therefore increase with voltage.

The electricity system consequently uses a hierarchy of voltage levels, with each level serving a different purpose.

High voltages are appropriate for moving large quantities of electricity over significant distances.

Lower voltages are more appropriate as electricity approaches the point of consumption.

This is why the electricity system does not operate at one universal voltage.

It is a carefully engineered sequence of voltage transformations.

Why This Matters for Data Centres

The principle becomes particularly important when considering very large data centres.

A facility requiring a relatively small amount of power can be supplied through a lower-voltage network.

A facility requiring tens or hundreds of megawatts represents a fundamentally different electrical load.

Supplying such a load at a low voltage would require extremely high currents and correspondingly greater electrical losses and infrastructure requirements.

Connecting large loads at higher voltages therefore becomes technically attractive and, depending on the size and location of the load and the available network, may be necessary.

This provides the physical foundation for understanding why some of Malaysia’s largest data centres connect at 132 kV or 275 kV.

The issue is not that data centres have been arbitrarily assigned a special voltage.

It is that their electrical demand can be large enough to interact directly with the high-voltage electricity system.

The next section examines how these voltage levels fit together within Malaysia’s electricity network.

MDCO Insight: High-voltage transmission is fundamentally an exercise in managing current: for the same power, higher voltage means lower current and substantially lower resistive losses.

Malaysia’s Voltage Hierarchy

Malaysia’s electricity network is organised around multiple voltage levels.

The precise configuration of a particular supply depends on the location, network conditions, customer demand and the applicable supply scheme. However, TNB’s public information provides a useful overview of the principal voltage levels used in Peninsular Malaysia.

Voltage	Broad function
500 kV	Highest-level transmission
275 kV	High-capacity transmission
132 kV	Transmission and major customer supply
33 kV	Distribution / bulk supply
22 kV	Distribution in selected areas
11 kV	Distribution
6.6 kV	Distribution in selected areas
400/230 V	Low-voltage consumer supply

TNB currently identifies 500 kV, 275 kV and 132 kV as transmission voltages, while 33 kV, 11 kV and 400/230 V are identified as distribution voltages. TNB also notes that 22 kV and 6.6 kV are used in certain parts of Johor and Perak.

These voltage levels should not be understood as eight completely separate systems.

They are interconnected through substations, transformers, switching facilities and network equipment.

500 kV

The 500 kV network represents the highest principal transmission voltage identified in TNB’s public description of the Peninsular Malaysia system.

At this level, very large quantities of electrical power can be transferred through the transmission network with comparatively low current for the amount of power being transported.

The 500 kV network can therefore form part of the backbone of the wider electricity system.

275 kV

The 275 kV system is another major transmission level.

It provides high-capacity transmission and interconnection between important parts of the network.

For very large electricity loads, 275 kV can also become relevant as a customer supply voltage, subject to the applicable technical assessment and supply arrangement.

132 kV

The 132 kV system occupies an important position between the higher-voltage transmission network and lower-voltage distribution systems.

It is also particularly relevant to large electricity consumers.

TNB’s public supply information lists 132 kV as one of its available supply voltages, with the actual voltage determined by the applicant’s load requirements and applicable supply scheme.

This is one reason 132 kV is particularly important when examining Malaysia’s data centre expansion.

33 kV and Below

As electricity moves towards ordinary consumption, progressively lower voltage levels are used.

TNB’s published connection guidelines describe a 33/11 kV Main Distribution Substation (PPU) as a facility connecting the 33 kV and 11 kV networks. They also describe distribution substations as points where 11 kV, 22 kV and, in some circumstances, 33 kV systems are transformed to the low-voltage 400/230 V network.

The result is a cascading structure:

500/275 kV → 132 kV → 33 kV → 11 kV → 400/230 V

This is a simplified representation rather than a statement that every customer passes through every level.

That distinction is important.

Not Every Consumer Uses Every Voltage Level

A common misconception is that electricity supplied to every consumer must travel through the entire hierarchy.

It does not.

The actual route depends on where the customer is connected and the amount of power required.

A household might ultimately receive a 230 V or 400 V supply.

A commercial or industrial customer may receive supply at 11 kV or 33 kV.

A very large customer may connect directly at 132 kV or 275 kV.

TNB’s public information explicitly states that the actual supply voltage depends on the magnitude of the applicant’s load requirements.

This distinction becomes particularly important when examining data centres.

A large data centre may bypass much of the lower-voltage distribution system because its demand is sufficiently large to justify a direct high-voltage connection.

The fact that it connects directly to a high-voltage network, however, does not mean that it is electrically independent of the rest of the system.

It remains part of the interconnected grid.

A Network Designed for More Than One Customer

Another important point is that the electricity network is designed as an integrated system.

TNB’s published network-planning information describes forecasting demand across supply zones and planning additional capacity where future demand is expected to exceed available capacity. This may involve new Transmission Main Intake substations or additional transformer capacity.

Therefore, when a major new electricity load such as a data centre is proposed, the question is not simply:

“Can a cable be connected from the nearest substation?”

The more fundamental question is:

Can the wider electricity system reliably accommodate the additional load, both now and over the planning horizon?

That may involve examining transmission capacity, substation capacity, network configuration, fault levels, protection arrangements and future demand.

It is one reason the connection of a very large electricity customer can require much more extensive technical planning than the connection of an ordinary building.

MDCO Insight: Malaysia’s electricity network is a hierarchy of interconnected voltage levels, and the voltage at which a customer connects reflects the scale and characteristics of its electrical demand.

Transformers — The Bridge Between Voltage Levels

The voltage hierarchy would not be practical without transformers.

A transformer allows electrical energy to be transferred between circuits operating at different voltages, making it possible to use high voltage for transmission and progressively lower voltages closer to the point of consumption.

This is one of the fundamental technologies underlying the modern electricity system.

Stepping Voltage Up

At or near a generating station, electrical power can be passed through a transformer that increases the voltage.

This is known as a step-up transformer.

The purpose is closely connected to the physics discussed earlier.

For a given amount of power, increasing voltage reduces current.

The electricity can therefore enter the transmission network at a much higher voltage than the voltage at which it was initially generated.

This reduces the current required to transport the power and, consequently, reduces resistive losses in the transmission conductors.

Stepping Voltage Down

The process is reversed as electricity moves towards consumers.

At substations, transformers can reduce the voltage from one network level to another.

For example, TNB’s published connection guidelines identify standard transformations at Transmission Main Intakes including:

• 275/132 kV
• 132/33 kV
• 132/11 kV

Further downstream, a Main Distribution Substation can transform 33 kV to 11 kV, while distribution substations can transform medium-voltage supply down to the 400/230 V low-voltage network.

The voltage therefore changes progressively as electricity moves through the system.

The Transformer as a Bridge

It is useful to think of the transformer as the bridge between different parts of the electricity system.

The transmission network needs high voltage because it transports large quantities of electricity efficiently.

The distribution system needs lower voltages because it is progressively bringing electricity closer to customers.

Customers ultimately require voltages appropriate for their electrical equipment.

Transformers allow these different requirements to coexist within one interconnected system.

The result is a system that can be represented conceptually as:

Generation → step-up transformation → high-voltage transmission → transmission substation → lower-voltage network → customer connection

For a very large customer, the sequence may be different from that of a household.

A large data centre may receive electricity directly from a high-voltage transmission system, with the customer connection designed around its substantial electrical demand.

The important point is that the transformation between voltage levels does not occur randomly.

It is part of a planned architecture.

Why This Matters for Data Centres

Transformers also illustrate why the electricity system supplying a data centre cannot be understood simply by looking at the cable entering its property.

The connection may depend on infrastructure upstream of that point.

There may be a transmission substation.

There may be transformers providing the necessary voltage transformation.

There may be transmission circuits connecting that substation to the wider grid.

There may also be additional network infrastructure required to provide sufficient capacity or an appropriate level of supply security.

TNB’s connection guidelines recognise that network planning for large developments can require land for transmission main intakes, substations and associated transmission line rights-of-way or wayleaves. The required infrastructure depends on the estimated demand and the applicable network design.

This is particularly relevant when a large data centre is connected to an existing electricity network.

The customer connection is visible.

Much of the system that makes the connection possible is not.

Understanding that distinction is the first step towards asking a more difficult question:

Which parts of this electricity infrastructure are built specifically for a new customer, and which parts already form part of the wider electricity system?

That question moves beyond engineering and into the economics and governance of electricity infrastructure.

It will become increasingly important in the next articles in this series.

MDCO Insight: Transformers make the entire electricity hierarchy possible, allowing power to move efficiently at high voltage while ultimately being delivered at voltages appropriate to different categories of consumers.

Why Data Centres Connect at 132 kV or 275 kV

One of the most visible features of a large data-centre development is its connection to the electricity grid.

Unlike a typical household or small commercial building, a hyperscale data centre may require an electricity supply measured in tens or even hundreds of megawatts. At this scale, connecting through the ordinary low-voltage distribution network would not be practical.

The reason large data centres connect at 132 kV or 275 kV is therefore primarily the magnitude of their electricity demand.

It is important not to interpret this as a special voltage level granted simply because a customer is a data centre. The electricity system does not fundamentally distinguish between a data centre and another very large electricity consumer on this basis. Rather, the required supply voltage is determined by the magnitude and characteristics of the load, together with the capability and configuration of the surrounding network.

TNB’s Electricity Supply Application Handbook establishes minimum supply schemes according to maximum demand. Its published guidance indicates, broadly, that:

• loads from 5,000 to 25,000 kVA are associated with a 33 kV minimum supply scheme;
• loads from 25,000 to below 100,000 kVA may require 132 kV or 275 kV depending on the applicable arrangement; and
• loads of 100,000 kVA and above have a minimum supply scheme of 275 kV.

TNB also makes clear that the final supply arrangement is subject to system capability studies and that alternative arrangements may be considered after taking into account factors such as location, economics and system security.

This distinction is important.

The voltage level is a consequence of the required electrical capacity and network conditions, not simply the identity of the customer.

From megawatts to megavolt-amperes

The relationship between the commonly quoted capacity of a data centre and the electricity connection requirement also requires some care.

A facility described as having a 100 MW electrical load represents approximately 100,000 kW of active power. However, the electricity system must also account for reactive power and power factor. The corresponding apparent power is expressed in kVA, rather than kW.

The two measures are related through power factor:

Power factor = kW / kVA

Consequently, 100 MW does not automatically mean exactly 100,000 kVA. The actual connection requirement depends on the electrical characteristics of the facility and the conditions specified by the utility.

Nevertheless, the order of magnitude illustrates the issue.

A load approaching 100 MW is not simply another ordinary distribution customer. It represents a substantial demand on the electricity system.

At this scale, TNB must consider not only whether electricity can reach the site, but whether the surrounding network can accommodate the new demand while continuing to operate securely.

Among the considerations are:

• available generation and network capacity;
• transmission constraints;
• substation capacity;
• transformer capacity;
• network configuration;
• voltage performance;
• system security;
• contingency requirements;
• anticipated future demand; and
• the geographical location of the proposed connection.

The objective is not merely to make electricity available under normal conditions. The network must also remain capable of operating within acceptable limits when equipment is unavailable, demand changes or other system conditions occur.

This is why a large electricity connection is fundamentally a system-planning exercise, rather than simply the installation of a cable from the nearest substation.

TNB’s published connection guidance expressly provides for system capability assessment in determining the appropriate supply scheme.

The implication is significant for understanding data-centre development in Malaysia.

A proposed data centre may have secured land, obtained planning approval and completed its building design, but its electricity connection still depends on whether the wider electricity system can accommodate the proposed load at the proposed location.

MDCO Insight: A data centre connects at high voltage not because it is a data centre, but because its electricity demand can be large enough to require direct connection to the high-capacity transmission system.

What Does It Mean to Connect a Data Centre to the Grid?

It is tempting to think of an electricity connection as a relatively simple physical arrangement: a power line reaches the site, electricity enters the facility and the customer begins consuming it.

For a large data centre, the reality can be considerably more complex.

From the utility’s perspective, connecting a major new customer can involve a combination of infrastructure and system-planning activities extending well beyond the boundary of the customer’s property.

Depending on the supply arrangement, this may involve infrastructure such as:

• transmission lines or cables;
• transmission substations;
• Transmission Main Intake (PMU) facilities;
• transformers;
• switchgear;
• protection systems;
• metering facilities;
• connection circuits;
• dedicated connection assets; and
• reinforcement or modification of existing network infrastructure.

TNB’s public connection guidelines specifically address the need to plan adequate land for transmission main intakes, substations and routes for cables or overhead transmission lines.

This illustrates an important characteristic of large electricity connections: the physical infrastructure required to serve the customer may extend beyond the customer’s site.

For example, a proposed development may require a new connection circuit from an existing transmission facility. In another case, the surrounding network may require modification or reinforcement before the additional load can be accommodated.

The exact arrangement depends on the location, load, existing network configuration and system studies.

Existing infrastructure, new infrastructure and reinforcement

It is therefore useful to distinguish three broad categories.

Existing infrastructure

This may include:

• existing transmission lines;
• existing substations;
• existing transformers;
• existing distribution networks; and
• other network facilities already serving the electricity system.

Infrastructure built specifically for the connection

This may include:

• dedicated connection circuits;
• new substations;
• customer-specific connection facilities;
• dedicated transmission or distribution assets; and
• associated protection, metering and control equipment.

Infrastructure that is reinforced or expanded

This may include:

• upgrading an existing substation;
• increasing transformer capacity;
• strengthening transmission circuits;
• modifying network configuration; or
• adding system capacity to accommodate additional demand.

These categories are useful for understanding the physical system, but they should not be interpreted as rigid categories.

A particular project may involve all three simultaneously.

An existing transmission network may provide the starting point. A new circuit may then be constructed to connect the development. At the same time, an existing substation or transformer may need to be upgraded to accommodate the additional demand.

The resulting electricity supply is therefore better understood as an interaction between the new customer and an existing infrastructure system.

This is an important conceptual point for the MDCO series.

A data centre does not necessarily “plug into” an existing electricity supply in the same way that a household connects to an established local distribution network.

A very large connection can involve:

existing infrastructure + dedicated connection infrastructure + network reinforcement

The proportions differ from project to project.

At this point, however, another question naturally arises:

Who pays for each component?

That question is deliberately outside the scope of this article.

It requires examination of TNB’s connection-charge mechanisms, tariff arrangements, regulatory treatment and the distinction between customer-specific assets and the wider regulated network.

Those questions will be examined in the subsequent MDCO articles.

MDCO Insight: Connecting a large data centre to the grid can involve existing infrastructure, new customer-related infrastructure and reinforcement of the wider network.

The Physical Relationship Between a Data Centre and the Wider Grid

The distinction between these different forms of infrastructure becomes even more important when considering the physical relationship between a data centre and the electricity system.

A useful way to visualise this relationship is to imagine three concentric layers.

Infrastructure that already exists

The first layer is the electricity infrastructure that was already present before the data centre was proposed.

This may include:

• transmission lines;
• transmission substations;
• transformers;
• distribution substations;
• distribution feeders; and
• associated land, buildings, protection and control infrastructure.

Such infrastructure may have been developed at different times and for different purposes as the electricity system expanded.

Some facilities may serve a relatively small geographical area, while others form part of the wider transmission system serving numerous consumers.

Infrastructure constructed for the new connection

The second layer consists of infrastructure introduced specifically to connect the new development.

Depending on the project, this may include new transmission circuits, cable systems, connection bays, transformers, substations or other facilities associated with the supply arrangement.

These facilities physically establish the connection between the wider network and the new customer.

Infrastructure reinforced because of the new demand

The third layer is more subtle.

The new customer may not simply require a new connection. Its demand may also affect the adequacy of infrastructure elsewhere in the system.

For example, an existing transformer may have sufficient capacity for existing consumers but insufficient spare capacity after a very large new load is added.

Similarly, a transmission corridor may have adequate capacity under existing conditions but require strengthening when a substantial additional demand is connected.

In such circumstances, the electricity connection becomes partly a question of network reinforcement.

This is why it is not always sufficient to ask:

“How much infrastructure does the data centre need?”

A more complete question is:

“What changes to the electricity system are required to accommodate the data centre’s demand while maintaining the required level of system performance and security?”

That distinction will become particularly important when MDCO examines cost allocation.

The physical infrastructure associated with a connection may have different characteristics:

Infrastructure category	Physical relationship
Existing network	Already forms part of the electricity system
Dedicated connection infrastructure	Constructed specifically to establish or serve the new connection
Network reinforcement	Modifies or expands existing infrastructure to accommodate additional system demand

In an actual project, the boundaries may not always be obvious.

An asset may be physically associated with one customer while also forming part of the broader electricity network. Conversely, infrastructure that appears to be “new for the project” may eventually become part of a wider network arrangement.

This is precisely why physical infrastructure and financial responsibility should not be assumed to be the same thing.

The physical question is:

What infrastructure is required?

The financial question is:

Who bears the cost?

And the regulatory question is:

How is that cost treated within the regulated electricity system?

E07 addresses only the first question.

The subsequent articles will examine the other two.

MDCO Insight: The physical relationship between a data centre and the grid can involve infrastructure that is existing, dedicated to the connection, or reinforced to accommodate the new demand.

Why Location Matters to Electricity Supply

The electricity connection requirement of a data centre cannot be understood by looking at its load alone.

Two data centres with identical electrical demand may present very different challenges if they are located in different parts of the country.

Consider two hypothetical projects, each requiring 100 MW.

The first may be located close to a major transmission substation with substantial spare capacity and suitable transmission corridors.

The second may be located in an area where the existing network is already heavily utilised and where additional transmission infrastructure would be required.

The two projects have the same demand.

But they do not have the same electricity-system requirements.

This is why location is an engineering variable as well as a planning variable.

A site may be attractive from an electricity-supply perspective because it is:

• close to existing high-voltage transmission infrastructure;
• near a suitable transmission substation;
• located within an established industrial or development corridor;
• close to infrastructure with available capacity; or
• situated where a suitable connection can be established without extensive network modification.

Conversely, a site may require substantially more infrastructure if adequate grid capacity is not available nearby.

This creates an interesting relationship between land-use planning and electricity planning.

A location may be attractive to a developer because of land availability, fibre connectivity, road access or proximity to customers.

At the same time, the electricity utility must consider whether the location is appropriate from the perspective of network capacity, system security and long-term grid development.

TNB’s connection guidance recognises the importance of location and system capability in determining supply arrangements.

Planning approval and electricity-system feasibility are different questions

This distinction also connects directly with →E06 — How Data Centres Are Approved in Malaysia.

Planning approval asks a question such as:

Is this development acceptable at this location under the applicable planning and regulatory framework?

Electricity-system assessment asks a different question:

Can the required electricity supply be provided at this location under the technical and security requirements of the electricity system?

Both questions matter.

A project may be acceptable from a land-use planning perspective but require substantial electricity infrastructure before it can receive the required supply.

Conversely, a location may be attractive from a grid perspective but face other planning, environmental or community constraints.

This reinforces a broader MDCO observation developed throughout the earlier series: major data-centre development is not governed by a single system.

Land, electricity, water, telecommunications, transport, environment and community considerations interact.

The electricity system is one of these systems—but for a large data centre, it can become one of the most important physical constraints.

MDCO Insight: Where a data centre is located can be almost as important as how much electricity it requires, because grid capacity and network configuration vary from place to place.

The Observatory Perspective

A data centre receives electricity through an electricity system that has been developed over many decades.

The electricity reaching the facility may pass through infrastructure that predates the project by many years.

At the same time, the new connection may require infrastructure that has never previously existed.

And somewhere between those two extremes, existing infrastructure may need to be upgraded, reinforced or reconfigured to accommodate the new demand.

The physical system can therefore contain several different layers:

Infrastructure that already exists

This forms part of the wider electricity network before the data centre arrives.

Infrastructure constructed for the new connection

This establishes the physical link between the network and the new customer.

Infrastructure reinforced or expanded

This enables the wider system to accommodate the additional demand while maintaining required performance and security.

There is another important distinction.

Some infrastructure may effectively serve a single customer.

Other infrastructure may serve multiple customers.

And some infrastructure may form part of a much larger interconnected system whose benefits extend well beyond the particular customer whose connection triggered a particular investment.

This leads naturally to a question that is central to the next stage of MDCO’s research:

When a very large customer connects to an existing electricity system, which infrastructure is paid for directly by that customer, and which infrastructure forms part of the wider electricity system?

There is no need to answer that question in E07.

Indeed, answering it properly requires us to distinguish between several different mechanisms, including electricity-supply application requirements, connection charges, tariffs, regulated network investment and the treatment of infrastructure under Malaysia’s electricity regulatory framework.

That is why the MDCO series will approach the issue in stages.

→E07.01 — The Electricity Supply Application Process for Data Centres will examine how a large customer applies for supply and what charges and infrastructure requirements arise before energisation.

→E07.02 — Understanding Electricity Tariffs in Malaysia will examine the tariff structure, including how electricity charges are constructed and how different customer categories are treated.

The A10 series will then move from explanation to analysis, examining who ultimately bears the costs associated with the electricity infrastructure serving very large data-centre loads.

The objective is not to begin with an assumption about who should pay.

It is to first understand what infrastructure exists, what infrastructure is required, how the electricity system is regulated, and how the different cost mechanisms operate.

That sequence is important.

Only after the physical system is understood can the financial and regulatory questions be examined properly.

MDCO Insight: Understanding the physical electricity system is the first step towards understanding how the costs of that system are ultimately allocated.

Selected References

Electricity System and Connection Infrastructure

• Tenaga Nasional Berhad (TNB) – Electricity System and Supply Application: Official information on Malaysia’s electricity system, including transmission and distribution voltage levels, system characteristics and the Electricity Supply Application Handbook (ESAH). https://www.tnb.com.my/esah/supply-application
• Tenaga Nasional Berhad (TNB) – Electricity Supply Application Handbook (ESAH): Technical guidance on electricity supply applications, planning criteria, supply schemes, connection arrangements and requirements for different customer demand levels. https://www.tnb.com.my/esah/
• Tenaga Nasional Berhad (TNB) – Connection Guidelines: Guidance on planning and design criteria, demand estimation, supply schemes and the development of connection infrastructure. https://www.tnb.com.my/esah/connection-guidelines

Electricity Grid and Distribution Regulation

• Suruhanjaya Tenaga – Grid Code for Peninsular Malaysia: The principal technical framework governing the planning, development and operation of the Peninsular Malaysia electricity grid, including the current 2025 Main Code and Additional Code. https://www.st.gov.my/resources/grid-code-peninsular-malaysia-main-code-2025
• Suruhanjaya Tenaga – Distribution Code: Technical requirements governing the planning, development and operation of distribution systems and the interface between distribution-system operators and users, including large power consumers. https://www.st.gov.my/resources/distribution-code-peninsular-malaysia-sabah-ft-labuan-amendments-2017
• Suruhanjaya Tenaga – Electricity Supply Industry and Regulatory Framework: Official information on the statutory and regulatory framework governing electricity generation, transmission, distribution and supply in Malaysia. https://www.st.gov.my/about-us

Electricity Infrastructure and System Planning

• Suruhanjaya Tenaga – Grid System Operation: Official information on grid-system operation, including supply-demand adequacy, transmission-network constraints, generation and system performance. https://www.st.gov.my/security/grid-system-operation
• Suruhanjaya Tenaga – Electricity Supply Industry Regulatory Documents: Official repository of legislation, regulations, codes, standards, guidelines and other regulatory instruments governing Malaysia’s electricity sector. https://www.st.gov.my/ms/tentang-kami/perundangan

Citation

Malaysia Data Centre Observatory (MDCO). E07 The Electricity System Supplying Malaysia’s Data Centres. MDCO Explain Series.

MDCO Note

This article forms part of the Malaysia Data Centre Observatory (MDCO) Explain Series, which aims to improve public understanding of data centre development through evidence-based, accessible and balanced analysis. It is intended for educational and informational purposes only and does not constitute legal, engineering, planning, environmental or professional advice.

Malaysia’s rapidly evolving data centre ecosystem includes facilities developed, owned or operated by organisations such as AirTrunk, Amazon Web Services (AWS), Bridge Data Centres, DayOne, EdgeConneX, Google, K2 Data Centres, Microsoft, NTT Global Data Centers, Princeton Digital Group (PDG), ST Telemedia Global Data Centres (STT GDC), STACK Infrastructure, Vantage Data Centers, YTL Data Centre Park and many others. MDCO is independent of these organisations, as well as governments, regulators, utilities and advocacy groups. Its role is to facilitate transparency, structured understanding and equal access to information by presenting publicly verifiable evidence, relevant context and multiple stakeholder perspectives. MDCO does not endorse, oppose or advocate for any particular organisation, project or policy position.