Published: June 2026 Technical Level: Advanced Category: Power Systems Design
This paper documents the electrical engineering design and deployment of a 200-port Level 2 and DC fast charging infrastructure at a corporate campus in the southwestern United States, with a total installed charging capacity of 1.8 MW. The case study addresses the engineering decisions that determine project success for large-scale EV charging deployments: utility service upgrade coordination, transformer sizing and placement strategy, load management system design and integration with the existing building management system, conduit and conductor routing optimization across a campus with constrained pathways, and the commissioning verification process. The project achieved its target peak demand impact of less than 300 kW added demand on the utility service — 17 percent of the theoretical unmanaged peak of 1.8 MW — through a cloud-based smart charging system that schedules charging sessions to match the campus's available electrical capacity while prioritizing vehicles with departure times that require immediate charging. The project provides a template for large-scale campus EV charging deployment under NEC Article 625 and California Title 24 requirements.
The campus consists of four office buildings with approximately 250,000 square feet of total floor area, served by a 12.47 kV utility service through two pad-mount transformers (1,500 kVA each) feeding 480Y/277 V switchgear in each building. The existing peak demand at the time of the EV project was approximately 1.2 MW, leaving approximately 400 kW of service capacity available before the transformers would reach their rated loading. The EV charging project was required to operate within this 400 kW available capacity without requiring a utility service upgrade, which would have added 18 to 24 months to the project timeline and approximately $2.1 million in utility charges.
The 200 charging ports were distributed across four parking structures: 40 Level 2 ports (7.2 kW each) and 10 DC fast charger ports (150 kW each) per structure. The total installed capacity is kW — far exceeding the available 400 kW capacity under unmanaged simultaneous charging. The entire project economics depended on the smart charging system's ability to keep the aggregate EV demand below 300 kW — a 96 percent reduction from theoretical maximum — through managed dispatch.
The 200 EVSE ports were served by eight new dry-type transformers (250 kVA each) located in the parking structures, fed from the existing 480 V switchgear in each office building via new 480 V feeders sized per NEC Article 215. The transformer secondary voltage is 208Y/120 V for the Level 2 EVSE circuits (NEMA 14-50 receptacles at 240 V single-phase for portable EVSE, or hardwired 208 V three-phase for fixed EVSE) and 480Y/277 V for the DC fast charger supply circuits.
The transformer sizing used the managed peak demand rather than the nameplate EVSE ratings as the design basis. Each structure's eight Level 2 EVSE circuits (7.2 kW each) and one DC fast charger circuit (150 kW) have a nameplate connected load of 57.6 + 150 = 207.6 kW. Under the smart charging system's dispatch, the maximum aggregate demand at each structure is limited to 75 kW — 36 percent of nameplate. The 250 kVA transformers are sized to 150 percent of the managed peak demand (75 × 1.5 = 112.5 kVA), providing margin for the initial loading transient when vehicles first connect and begin charging at maximum rate before the smart charging system's rate adjustment takes effect.
The conduit routing from the office building switchgear to the parking structure electrical rooms required routing through underground pathways shared with existing communications conduit and domestic water piping. The routing was developed using a BIM coordination model that identified conflicts between the proposed electrical routing and the existing infrastructure, and resolved conflicts before construction by adjusting the electrical route or the locations of pull boxes.
The feeders were sized for the managed peak demand plus a 25 percent margin per NEC continuous load rules (NEC 215.2), using THWN-2 conductors in PVC conduit. Voltage drop was verified for the longest run (430 feet from switchgear to the furthest parking structure) at the managed peak demand current: a 350 kcmil copper conductor at the managed peak current of 90 A produces a voltage drop of 1.8 percent — within the 3 percent branch circuit recommendation of NEC Informational Note 1 to 210.19(A)(1).
The smart charging system consists of a cloud-based energy management server that communicates with each EVSE via the Open Charge Point Protocol (OCPP) 2.0.1, the campus building management system via BACnet/IP, and the utility's demand response API via a custom integration. The server implements a priority-based dispatch algorithm: upon vehicle connection, the driver enters a departure time via the EVSE touchscreen or a mobile app; the server calculates the minimum average charging rate required to deliver the target energy by departure time and assigns the vehicle to a priority tier (urgent if the minimum rate exceeds 6 kW, standard if it requires 2 to 6 kW, flexible if it requires less than 2 kW).
The aggregate charging demand across all EVSE is controlled by adjusting individual charger setpoints in real time, keeping the total campus EV demand within the available electrical capacity budget. The capacity budget is set by the BMS interface: when the campus non-EV demand rises above 1.1 MW (approaching the 1.2 MW service capacity minus 100 kW safety margin), the EV capacity budget is reduced proportionally, deferring flexible-tier vehicles first and reducing standard-tier vehicles if necessary, while urgent-tier vehicles maintain their assigned rate.
The effect of the smart charging system on site demand is illustrated in Figure 1, which contrasts the uncontrolled charging load with the managed net load against the site demand limit.

Figure 1. Managed versus uncontrolled charging demand over a day. The horizontal axis is time of day in hours and the vertical axis is site demand in kilowatts. Uncontrolled charging produces a sharp coincident peak when vehicles arrive; the smart charging controller spreads the same energy across a longer window to hold demand below the site limit. The engineer should observe that the managed profile is what allows the existing transformer and feeder to serve the chargers without an upgrade, which is frequently the deciding factor in the project's economics.
The commissioning process verified that the smart charging system maintained the campus demand below the 1.5 MW threshold (existing 1.2 MW plus 300 kW EV budget) under three test conditions: a baseline test with 50 vehicles charging simultaneously at maximum rate, which confirmed that the system correctly shed charging rate to 300 kW aggregate within 15 seconds of detecting the overload condition; a departure priority test with vehicles requiring various energy amounts by various deadlines, which confirmed that urgent-tier vehicles received priority dispatch and departed with the requested energy in 94 of 100 test sessions; and a demand response test where the utility's API signaled a demand response event, which confirmed that the system reduced EV charging to 50 kW within 60 seconds and maintained that level for the 30-minute event duration.
The measured peak demand impact over the first 90 days of operation was 287 kW — within the 300 kW design target. No utility service upgrade has been required, and the project's avoided upgrade cost of $2.1 million represents the project's single largest financial benefit.
The analysis in this paper connects to several companion studies in this library. Readers concerned with the upstream and downstream engineering will find Electric Vehicle Charging Infrastructure develops a closely related aspect of the same problem, while Electric Vehicle Charging Infrastructure extends the treatment into an adjacent domain. For the broader methodological context, Electric Vehicle Charging Load Management provides complementary depth.
The 200-port, 1.8 MW campus charging deployment documented in this paper demonstrates that the engineering decisions determining project success for large EV installations are the utility-service coordination, the transformer sizing and placement strategy, and the integration of an automatic load-management system with the existing building management system. The central conclusion is that load management is the decisive design element: by capping the managed charging load well below the aggregate charger nameplate, the design accommodated the full 200-port capacity within a service that the unmanaged nameplate load would have far exceeded, avoiding a costly utility upgrade. The case study confirms that conduit and conductor routing, while routine, materially affects installed cost at this scale and rewards early optimization. For the practicing engineer, the deployment illustrates that large EV charging projects succeed when load management is treated as a core architectural decision integrated with the facility's existing controls, rather than as an operational feature added after the electrical design is fixed, because the managed-load limit is what reconciles the charging capacity with the available service.
[1] NFPA 70, National Electrical Code, Articles 215, 625, 2023 edition, NFPA, 2023.
[2] SAE International, SAE J1772: EV Conductive Charge Coupler, SAE, 2017.
[3] Open Charge Point Protocol (OCPP) 2.0.1, Open Charge Alliance, 2022.
[4] ASHRAE Standard 135-2020, BACnet, ASHRAE, 2020.
[5] IEEE Standard C57.91-2011, Guide for Loading Mineral-Oil-Immersed Transformers, IEEE, 2012.
[6] California Energy Commission, California Title 24, Part 6: Building Energy Efficiency Standards, CEC, 2022.
[7] EPRI, Electric Vehicle Charging Infrastructure — Planning and Design Guide, EPRI 3002019900, 2022.
[8] U.S. DOE, Alternative Fuels Station Locator and EV Infrastructure Deployment Guidelines, DOE, 2024.