← Survey overview · Case Study 06 · Updated June 2026

16.8 MW Data Center — Southeastern US

Generators beneath the chillers: a ground-level generator row sits directly under the west rooftop chiller array, and under a westerly wind the exhaust follows a short, direct path into the intakes above. Public-source inputs. Screening-level.

What this page covers

This page presents a screening-level exterior CFD analysis of a 16.8 MW data center in the Southeastern United States, developed by a major colocation operator and engineered by a national firm with a dedicated mission-critical practice. Heat rejection is a large rooftop air-cooled chiller array; the standby generators sit at grade in a row along the west side — directly beneath the west end of that array.

The result is a short, direct recirculation geometry. Under a westerly wind, the generator exhaust has very little distance to travel before it reaches the chiller intakes above it — and in a hot, humid summer climate, that re-ingested air erodes the capacity of an air-cooled plant.

Facility context

The facts below are derived entirely from cited public sources: published facility and provider materials, aerial and street-level imagery, and public terrain data. No proprietary drawings, specifications, or operational data are used. Identifying details are withheld; the facility is described by type and configuration only.

Region.

Southeastern United States (hot, humid design climate).

Owner-operator.

A major colocation operator. Not identified here.

Engineer of record.

A national engineering firm with a dedicated mission-critical practice. Not identified here.

Critical IT capacity.

16.8 MW critical (22 MW electrical service), N+1 redundancy.

Cooling.

36 rooftop air-cooled chillers in four rows of nine, vertical discharge.

Backup power.

10 diesel standby generators in a ground-level row along the west side, directly beneath the west chiller array.

Scenario presented

Wind and ambient.

Wind from the west at 10 mph, 92°F ambient — near the regional summer design condition. This vector carries the west-side generator exhaust up into the chiller array directly above it.

Operating scenario.

All generators and all chillers running at full load — the on-generator condition that maximizes both the exhaust source and the heat-rejection demand together.

This case was isolated by a directional sweep across the full compass — not the standard four-point (N/S/E/W) check. As with any generators-beneath-chillers layout, the governing variable is the vertical adjacency more than the wind: the exhaust source sits under the intakes.

What the model shows

The observations below are qualitative readings of the presented scenario. Quantitative outputs — per-unit intake temperatures, recirculation magnitudes — are shared with the owner's team on request.

1. A short, direct recirculation path. With the generator row directly beneath the west chiller array, exhaust is drawn straight up into the chillers above rather than dispersing — the shortest exhaust-to-intake path a layout can produce.
2. West-side intakes run markedly hotter. The west-side chiller intakes see local temperatures well above the downwind units, concentrated where the array sits over the generator enclosures.
3. Intake dry-bulb governs capacity. Air-cooled chiller performance tracks intake dry-bulb directly, so the recirculation translates immediately into lost heat-rejection headroom — a first-order effect, not a rounding error.
4. Best caught while layout choices are still open. This is the kind of exposure that is inexpensive to resolve on the drawing — a placement or orientation change — and expensive to remedy once built. The question worth putting to the engineering team: was the generators-beneath-chillers relationship modeled at the design ambient, or assumed adequate?

Methodology

The methodology applied here is the same standardized exterior CFD approach applied to every facility in the cohort — cylindrical far-field domain, logarithmic atmospheric boundary layer inlet, polyhedral mesh in Siemens STAR-CCM+, realizable k-ε RANS baseline. The presented case uses a 92°F ambient. Full domain setup, boundary conditions, solver choices, and stated limitations are documented at the Methodology page. Key terms are defined at the Key Terms and FAQ page.