Rooftop Solar Feasibility Data for an FDI Factory in Vietnam

A factory should not start a rooftop solar tender with only a target such as “1 MWp” and a recent electricity bill. A credible feasibility decision needs four connected evidence sets: the usable roof and its structural limits; interval electricity demand and operating calendar; the electrical interface at the proposed point of connection; and the legal, utility, fire-safety and contractual pathway applicable to the site. Until those datasets agree, capacity, yield, self-consumption and price remain scenarios rather than a design.

This guide is written for an FDI factory owner, project director or procurement team preparing a Vietnam rooftop PV feasibility review. It defines the information to request, the calculations to reproduce, the boundaries to assign and the evidence expected before issuing an EPC request for quotation. It does not certify a roof, approve a grid connection or state that a particular project qualifies for a legal mechanism.

1. What decision should the feasibility study support?

The first page of the study should state the decision in plain language. A useful decision is not merely “solar is feasible.” It should answer whether a defined part of the factory roof can host a bounded DC array, whether the proposed AC export or zero-export mode is compatible with the facility electrical system, how much energy is expected to be consumed behind the meter, what must be upgraded, and which uncertainties can change the investment case.

Separate three levels of answer:

• Screening: identify fatal constraints and a capacity range using available drawings and bills.
• Pre-feasibility: use interval load data, surveyed roof zones, preliminary electrical architecture and stated loss assumptions to compare scenarios.
• Investment-grade design basis: use verified surveys, structural calculations, utility and authority requirements, selected equipment data and an agreed commercial model.

A supplier should not present a screening result as an approved design. The owner should also avoid asking every bidder to invent different missing inputs; bids produced from different assumptions are not comparable.

2. Minimum data pack before the site survey

For a full year, 15-minute data contains 35,040 intervals before daylight-saving or missing-data adjustments. The study should identify gaps, time-zone errors, meter multiplication factors and whether the series represents whole-site demand or only one board. Monthly bills can validate energy totals, but they cannot show the noon minimum demand that controls export or curtailment.

3. Roof feasibility is more than gross square metres

Begin with a zone register, not one gross roof area. Each zone should record slope, orientation, roof sheet or slab type, purlin arrangement, corrosion, drainage, obstructions, shaded areas, maintenance routes, emergency access and ownership constraints. A roof plan should distinguish measured facts from assumptions.

The structural consultant needs actual module, rail, clamp and ballast data; distributed dead load alone is not enough. Wind uplift, local edge and corner zones, point loads, fastener pull-out, roof-sheet condition, purlin capacity and construction loads can govern. Existing roof warranties and water-tightness responsibilities must be assigned before drilling or clamping. The solar contractor should provide equipment reactions and layout inputs; the qualified structural party should decide adequacy and required strengthening.

A preliminary packing factor can screen capacity, but it is not a code rule. For example, assume a not-yet-selected 580 W module occupies 2.58 m². A nominal 1,000 kWp target would require 1,725 modules and produce 1,000.5 kWp DC. Module face area is about 4,450.5 m². If the screening model adds 25% for spacing, access and exclusions, the preliminary roof envelope becomes about 5,563 m². These are illustrative assumptions only: final quantities must use the selected module dimensions, surveyed exclusions and the approved fire and maintenance routes.

4. Convert energy data into an operating case

The owner should ask for an interval model that aligns factory load and simulated PV output at the same timestamp. At each interval, direct consumption is the smaller of facility demand and PV production, while surplus is export, storage input or curtailment according to the agreed control mode. The model should report at least annual PV generation, direct self-consumption, surplus, curtailment, site import reduction and the maximum reverse power at the point of connection.

Solar resource data from the World Bank/ESMAP Global Solar Atlas is appropriate for early screening, provided the study records the coordinates, dataset version, plane-of-array assumptions and loss chain. Investment decisions should compare the model with local measurements or a bankable dataset appropriate to the commercial requirement. A specific-yield value copied from another province or another roof is not evidence for the present project.

Consider this transparent hypothetical example:

• Installed DC capacity: 1,000 kWp.
• Screening specific yield: 1,350 kWh/kWp-year.
• Calculated annual PV energy: 1,350,000 kWh.
• Assumed direct self-consumption from interval simulation: 85%, or 1,147,500 kWh.
• Assumed surplus requiring export, storage or curtailment: 202,500 kWh.
• Factory annual electricity use: 4,800,000 kWh.
• PV share of annual factory use: 1,147,500 / 4,800,000 = 23.9%.

The 85% value is not a benchmark and should never be applied without an interval model. If a weekend, holiday or seasonal production change reduces daytime demand, the surplus may be materially higher even when annual factory consumption is unchanged.

5. Define the AC connection and the PCC boundary

The electrical feasibility section should draw a functional single-line diagram from PV strings through DC collection, inverters, AC collection boards, transformer if applicable, the factory switchboard and the point of common coupling. Mark ownership, isolation, metering, protection and control boundaries. State whether the study is DC-kWp-led, AC-kW-limited, export-limited or roof-limited.

Check available switchboard capacity, busbar ratings, breaker duties, cable routes, transformer loading in minimum and maximum demand conditions, fault level, protection coordination, anti-islanding requirements, earthing and lightning interfaces. A module certificate does not prove that the installed system, switchboard modification or protection scheme complies as an integrated plant.

For another illustrative check, 900 kW AC at 400 V and power factor 0.99 corresponds to approximately:

I = 900,000 / (√3 × 400 × 0.99) = 1,312 A.

This is only a balanced three-phase current estimate. It does not select a breaker, busbar or cable because installation method, temperature, grouping, harmonics, short-circuit duty, voltage drop, coordination and manufacturer data remain unresolved. A large rooftop system may be collected or connected at medium voltage instead; the topology must follow the actual factory network.

If the factory minimum daytime load is 650 kW while the PV plant could produce 900 kW AC, the instantaneous surplus is 250 kW. A zero-export scheme therefore needs a control and fail-safe basis capable of managing that condition; alternatively the owner must establish an approved export or storage route. Writing “zero export” in the proposal is not a protection and control design.

6. Keep legal, technical and commercial scopes separate

Vietnam’s Decree 58/2025/ND-CP implements parts of the Electricity Law for renewable and new energy development. The project team should verify the current self-production/self-consumption, notification or registration, sale/export, planning, construction, fire-safety and utility procedures that apply to the exact capacity, voltage, site and commercial arrangement. This article does not interpret a threshold or grant eligibility. The applicable procedure must be confirmed against the official text and competent authorities at the time of the project.

7. Feasibility outputs the owner should require

A practical pre-feasibility package should include:

1. A site and roof-zone register with measured, documented and assumed fields separated.
2. A roof layout concept showing exclusions, access, drainage and equipment areas without pretending to be an approved construction layout.
3. A load-data quality report and aligned interval energy model.
4. At least three capacity scenarios, including the limiting constraint and sensitivity to yield, self-consumption and curtailment.
5. A functional single-line diagram and PCC boundary matrix.
6. A preliminary equipment and cable-route concept, including space and shutdown implications.
7. A structural design-input schedule for the qualified structural consultant.
8. A legal/utility/authority action register with responsible parties and decision dates.
9. A risk register covering roof condition, corrosion, water ingress, fire access, production disruption, export control, data gaps and future loads.
10. A Class-appropriate cost estimate with exclusions, not an undisclosed fixed-price promise.

The report should identify what evidence would change the recommendation. For example, a survey may reduce usable area; interval data may expose weekend surplus; a fault study may require switchgear work; or the structural review may require reinforcement. These are decision variables, not inconvenient details to hide until after award.

8. A bid-comparison matrix for the RFQ

9. Common red flags

• The proposal derives capacity only from gross roof area.
• Monthly bills are used to claim a self-consumption percentage.
• Specific yield is presented without coordinates, dataset, losses or uncertainty.
• The point of connection and ownership boundary are missing from the SLD.
• “Zero export” is stated without meter, controller, protection and failure-mode requirements.
• Module or inverter certificates are presented as proof that the whole plant is compliant.
• Structural approval, fire routes, roof warranty and water-tightness responsibilities are excluded from the responsibility matrix.
• A budget estimate is described as a generation guarantee or firm EPC price.
• The bidder cannot provide interval-model inputs and outputs in an owner-readable format.

10. Before requesting a budget or EPC proposal

Owners can first review QuangAnhcons’ industrial rooftop solar electrical scope and the Solar Electrical BoP integration page. For project-wide electrical boundaries, see what an electrical EPC tender is buying and the factory electrical due-diligence guide.

To prepare a review, send the site coordinates, roof drawings, current SLD, transformer and switchboard data, 12 months of bills, interval load data, operating calendar and planned new loads. QuangAnhcons can then identify which assumptions are still open and define an engineering and commercial survey scope. Any proposal remains subject to site verification, qualified structural and electrical design, applicable authority and utility requirements, and agreed contractual boundaries.

Frequently asked questions

Can annual electricity bills size rooftop solar?

No. Bills help reconcile annual and monthly consumption, but interval demand determines direct use, export and curtailment. Use timestamp-aligned load and PV data.

Does a 1,000 kWp target mean the roof can carry it?

No. Capacity depends on surveyed usable zones, module geometry, access and fire routes, structural reactions, wind, roof condition and the selected mounting system.

Is a preliminary yield estimate a guarantee?

No. A guarantee requires an agreed weather basis, loss model, availability rules, measurement system, exclusions and contractual test method.

Can module IEC certificates prove the project is compliant?

No. Component qualification is only one evidence layer. The integrated array, electrical installation, protection, structure, monitoring and commissioning records must be assessed for the actual project.

No quick estimate yet. You can still submit a project quotation request.