Pillar 1 drill-down

Mamre Road: grid readiness

706–752 Mamre Road, Kemps Creek NSW · SSD-92743706 · 1 GW · 52 ha
Proposed connection: three lead-ins from Transgrid's Sydney West 330 kV substation

⚠ The binding constraint

N-1 firm transformer capacity at the Sydney West Bulk Supply Point. Sydney West has been in firm-capacity deficit since 2025/26, before this project is added. Transgrid's own RIT-T puts the gap between forecast maximum demand and N-1 firm transformer capacity at 48 MVA in 2025/26, rising to 791 MVA by 2032/33. The committed remedy is a single 375 MVA transformer. The deficit is already larger than the fix.

A 1 GW load is roughly 1,050 MVA at 0.95 power factor, which is more than the entire forecast 2032/33 shortfall again.

The frequently cited Kemps Creek BSP does not resolve this. It is real, funded and committed, but it was scoped to relieve the Macarthur 132 kV BSP and serve Aerotropolis distribution growth. It is not a transmission connection asset for a gigawatt load at Sydney West.

The gap, to scale

Deficit versus remedy versus this site

Transgrid publishes the firm-capacity gap at two points in time. Intermediate years are not published, so only the two stated values are shown. The remedy and the site load are placed on the same axis for comparison.

Published firm-capacity gap Committed remedy This site's load

How to read this. The green bar is the only committed capacity addition at this connection point. It is smaller than the deficit that exists before this project, and roughly a third of what this project alone would draw. Site load is our conversion of 1 GW at 0.95 power factor and is an estimate, not a Transgrid figure.

Section 1

Thermal and N-1 firm capacity

From Transgrid's RIT-T Maintaining reliable supply to Western Sydney, commenced as a reliability corrective action under Schedule 5.1.4 of the National Electricity Rules.

Measure	Value	Grade
Demand vs N-1 firm transformer capacity	Forecast demand exceeds firm capacity from 2025/26, under both POE50 and POE10	A
Gap, 2025/26	48 MVA	A
Gap, 2032/33	791 MVA	A
Expected unserved energy, 2025/26	5 MWh	A
Expected unserved energy, 2031/32	approximately 2,036 MWh	A
Expected unserved energy, 2047	11,307 MWh	A
Transformer repair duration assumed	72 days	A
Consequence of a contingency	Load shedding required to hold load below the firm capacity of remaining in-service transformers	A
Supplying 330 kV lines	Seven. Three north (20, 26, 29), three west (32, 38, 39), one south (30)	A
Referenced operating limit	1,500 MVA at Sydney West BSP	B

Why grade A. Every figure is stated in Transgrid's own RIT-T consultation report, produced under a regulatory obligation and specific to this exact connection point. This is the best public evidence available short of a connection study.

Currency caveat. That RIT-T rests on 2023 demand forecasts and 2022-23 dollars. Forecasts have since risen, not fallen: Transgrid's June 2026 letter reports more than 8 GW of prospective data centre load in advanced discussions. Treat 791 MVA as a floor.

Worth noting. Transgrid's Western Sydney load composition chart lists data centres as a distinct load category alongside commercial, industrial and residential. In this network's planning documents, data centre demand is a named driver, not a rounding error.

Section 2

Planned augmentations, and which one actually matters

Two separate regulatory processes are in play. Conflating them is the most common error in this story, and it consistently makes the site look better than it is.

Addresses the binding constraint

Sydney West BSP reliability

Preferred option	New 375 MVA 330/132 kV transformer at existing Sydney West BSP A
Capital cost	$24.78m ±25% ($2022-23) A
Commissioning	2027/28 A
Alternative	New 330/132 kV BSP at Mt Druitt, convert lines 932 and 219 to 330 kV B
Net effect	Closes under half the existing deficit, before this project is added

Does not address it

Kemps Creek BSP

Identified need	Flows exceeding Macarthur 132 kV BSP capacity, potentially from 2026/27 under transformer outage A
Capital cost	$136.7m ($111.7m Transgrid + $25.0m Endeavour) A
Net market benefits	approximately $600m present value A
Delivery	Six years, commissioning 2029/30 A
Environmental approval	REF exhibited 11 Jun to 10 Jul 2026 A
Scope	Aerotropolis distribution growth, not a gigawatt transmission connection at Sydney West

Section 3

Queue position, timing and who pays

From Transgrid's letter to data centre proponents of 17 June 2026, signed by the Executive General Manager, Network.

Measure	Value	Grade
Prospective DC load in advanced discussions	More than 8 GW	A
NSW average daily demand, Transgrid's own comparator	7.5 to 10 GW	A
Connection agreements already signed, Western Sydney	approximately 1.5 GW, accommodable near-term	A
Enquiries within 12 km of Sydney West since late 2024	14 GW, equated by Transgrid to NSW peak winter load	B
Capacity position	Largely exhausted beyond 2033	A
Who funds augmentation	Proponents that trigger the need or benefit, funded upfront	A
Cost exposure for this site	Requires network study. Not inferable from public data	n/a
Indicative time to full 1 GW connection	Contingent on augmentation that is neither designed nor funded, and on major transmission projects Transgrid describes as uncertain in timing	C

Wording correction worth having in your pocket. Secondary reporting renders this as "largely full by 2033". The primary letter says capacity is largely exhausted beyond 2033. The distinction is material and slightly favourable near-term: headroom exists until then, and disappears after.

On cost. We will not put a number on this site's augmentation bill, because no public source supports one. What is A-grade evidence: the liability sits with the proponent, it is payable upfront, and comparable augmentations in this corridor have run between $24.78m and $263.1m. A gigawatt-scale augmentation sits at or beyond the top of that range.

Section 4

What public data cannot tell you

These fields are marked, not graded. Assigning even a D would imply a basis that does not exist, and a connections engineer would spot it immediately.

Requires network study

Each of these is a live question for a load of this size. None is answerable from a TAPR, a RIT-T or a hosting capacity map.

Fault level and short-circuit headroom	Requires the network model and a fault study at the specific connection busbar. Not published at any granularity.
System strength	Requires assessment against the NER system strength framework, considering the specific converter technology and its interaction with local synchronous plant.
Transient and voltage stability	Requires dynamic simulation of the actual connection configuration and load models.
Protection coordination	Requires the protection scheme design, settings and grading study for the lead-ins and the BSP.

This is where a network study earns its fee, and saying so is more credible than pretending otherwise. Any vendor claiming to derive fault level or system strength from open data is selling you a number with nothing behind it.

Section 5

What this changes for the score, and what is still missing

Nothing, and that is the point. The scorecard already grades Grid Headroom 1.1, available capacity in zone, as Severe, which holds the pillar in its bottom band under the Critical Sub-Indicator Rule and caps the site at 39, High-risk. This deep-dive does not move the score. It evidences it, at a level of specificity a connections engineer can argue with.

The scorecard says	The deep-dive says
"No spare capacity for a gigawatt-scale load"	Firm capacity deficit of 48 MVA in 2025/26 growing to 791 MVA by 2032/33, against a committed remedy of 375 MVA
"Requires major network augmentation"	Which BSP, which RIT-T, what it cost, when it commissions, and why the one everyone cites solves a different problem

Gaps in this analysis, stated plainly

No connection application or offer exists on the public record for this site. Everything site-specific is therefore inference from network-level evidence.
The Sydney West RIT-T rests on 2023 forecasts. Demand has grown since. Figures are a floor, not a ceiling.
Endeavour Energy's DTAPR has not been read in original, so distribution-level headroom at the relevant zone substations is not covered.
The NSW Network Hosting Capacity Opportunities Map has not been queried. Hosting capacity is a distribution concept and of limited use for a transmission-connected gigawatt load, but it should be checked before a paid report.
AEMO ESOO and ISP have not been mined for the NSW-wide reliability and transmission-development picture framing the major projects Transgrid refers to.
The Transgrid TAPR connection opportunities portal publishes N-1 utilisation data and would sharpen Section 1 considerably.

Items 3 to 6 are the highest-value next increment, and none requires new methodology. This is the shape of what a paid pilot buys.