Solar BOQ Margins — How EPCs Lose 6% on Mis-Costed Cables

Cable is the most consistently mis-costed line item in Indian solar project BOQs. Not because EPC founders do not know what cable costs — it is because the quantity calculation is wrong in five specific ways that compound on each other. A project where the cable BOQ is off by 18% in quantity and the AC feeder is one size undersized means actual cable cost runs 30–34% above the quoted figure. On a ₹3.5 Cr project where cable represents 8–12% of CAPEX, that is ₹84,000–₹1.42 lakhs in unrecovered cost. Multiply across 20 projects per year and you have a ₹1.7–₹2.8 Cr annual margin leak that never shows up as a single identifiable line item — just as a project that always finishes over budget.

Direct answer. Indian solar EPCs lose an average of 5.5–7% of project margin through five systematic cable BOQ errors: DC string cable undercount from routing deviation, AC feeder undersized due to wrong voltage drop assumption, earth cable specification below IS 3043 requirements, conduit and tray quantity calculated at straight-line distance without bends, and termination hardware excluded from the BOQ. The BOQ Leak Audit — Heaven Designs’ five-category proprietary framework — identifies and corrects all five, typically recovering ₹0.8–₹1.5 lakhs per MW in margin that would otherwise be lost post-commissioning.

This analysis is for Rohan — the Indian EPC founder who discovers the cable variance at material delivery, not at bid. Every number in this article is India-specific; currency is ₹.

Why Cable Is the BOQ’s Most Dangerous Line Item

Cable mis-costing is insidious because cables are invisible in the final installation — they run inside conduit, within cable trays, underground. Nobody physically inspects cable quantity the way they inspect modules or inverters. The quantity error only surfaces at material delivery when the electrician reports “we need three more reels” and the project manager has to buy them at spot rate with zero margin — and often at a higher specification than the original BOQ price because the spot vendor only carries one grade.

The five cable errors are independent. Each exists in EPCs that have the other four right. Correcting all five requires a systematic BOQ protocol, not just more careful estimation. An EPC that fixes only one or two errors will still bleed margin on the remaining three.

According to Bridge to India’s EPC Margin Trends analysis (2025), solar EPC gross margins in India declined from 12–15% in 2022 to 7–10% in 2025 due to equipment price normalization and competitive bidding. In a 7–10% margin environment, a 5.5–7% cable leak is not a rounding error — it is a structural threat to project profitability.

Error 1 — DC String Cable: The Routing Deviation Problem

The most common DC cable error is calculating string cable quantities from the module layout drawing using straight-line distances from the module row to the string combiner box. Real cable routing is never straight-line — it follows mounting rails, drops to the rail edge, routes along purlins, and runs up a conduit to the combiner box. The deviation factor between straight-line distance and actual routing distance is typically 1.18–1.32 depending on the mounting system design.

For a 300 kW rooftop with 600 modules in 50 strings of 12 modules, if the average straight-line string distance is 8 meters but the actual routing distance is 10 meters (a 1.25x factor):

• Straight-line BOQ: 50 strings × 8 m × 2 (positive + negative) = 800 meters of 4 mm² PV cable
• Actual requirement: 50 strings × 10 m × 2 = 1,000 meters
• Under-estimate: 200 meters of 4 mm² XLPO cable
• Cost impact at ₹65/meter: ₹13,000 — bought at spot rate, typically at ₹75–₹85/meter

Before the connector hardware error compounds it, the DC cable routing error alone costs ₹13,000–₹17,000 per 300 kW system. On a 20-project year with 300 kW average, that is ₹2.6–₹3.4 lakhs/year in unrecovered DC cable cost.

Error 2 — AC Feeder Cable: Wrong Voltage Drop Assumption

AC feeder cable from the inverter output to the DISCOM metering point is sized on two independent criteria: ampacity (current-carrying capacity per IS 1554) and voltage drop (limited to 1.0–1.5% per most DISCOM technical specifications). Most BOQs size AC cable on ampacity alone and ignore voltage drop entirely — or calculate voltage drop using a generic cable resistance value rather than actual cable reactance from the cable datasheet.

Consider a 3-phase feeder running 150 meters from a 100 kW inverter output to the LT panel at 0.9 power factor:

• Load current at 230V (phase to neutral): 181 A
• Ampacity requirement: 4-core 70 mm² cable (200 A capacity per IS 1554, ambient 40°C)
• Voltage drop at 1.0% target over 150 m (full-load): requires 4-core 95 mm² (at 0.9 PF including reactance)

An EPC that sizes on ampacity alone specifies 70 mm² at 150 meters and delivers 2.1% voltage drop — above the DISCOM 1.5% limit. The cable must be replaced with 95 mm² at project completion, adding ₹22,000–₹32,000 in unplanned cable cost, a 5–7 day rework period, and in some cases a DISCOM technical inspection failure that delays commissioning.

Feeder Length	Load (100 kW inverter)	Ampacity-Only Size	Correct VD-Limited Size	Voltage Drop Error
50 m	181 A	35 mm²	50 mm² (1.0% VD)	1.8% — over limit
100 m	181 A	35 mm²	70 mm² (1.2% VD)	3.4% — far over
150 m	181 A	50 mm²	95 mm² (1.1% VD)	2.1% — over limit
200 m	181 A	70 mm²	120 mm² (1.0% VD)	1.8% — over limit

The voltage drop calculation must use the actual cable reactance at the operating power factor — the total impedance includes both the resistive (R) and inductive (X) components. A generic assumption of pure resistance underestimates voltage drop by 8–15% depending on the cable type. The correct calculation is VD = I × L × (R cos φ + X sin φ) / 1000, where R and X come from the cable manufacturer’s datasheet.

For the complete AC cable sizing methodology including power factor and reactance, see the solar engineering workflow for Indian EPCs.

Error 3 — Earth Cable: IS 3043 Under-Specification

Earth cable is the most under-specified item in Indian solar BOQs. IS 3043 (Code of Practice for Earthing) and the CEA General Safety Requirements specify minimum earthing conductor sizes based on the phase conductor size — not based on a generic “looks adequate” assessment. Most BOQs specify 16 mm² copper earth cable throughout because it appears adequate and costs 35% less than the correctly specified 25–50 mm² copper.

For systems above 50 kW:

• Module frame bonding (rail-to-rail): 35 mm² copper per IS 3043 Table 5 (based on 10 mm² DC conductor size)
• Structure-to-earth pit: 50 mm² copper (from mounting structure to dedicated earth electrode)
• Inverter protective earth: minimum 25 mm² copper to earth bus (based on 70 mm² AC phase conductor)
• Earth bus to DB/ACDB: 25 mm² copper (minimum per IS 3043)
• Lightning protection earth conductor: 50 mm² copper per IS 2309

Total earthing cable on a 300 kW roof system: 850–1,200 meters depending on roof configuration and stringing layout. Specifying 16 mm² throughout saves ₹18,000–₹24,000 on the BOQ — but creates a liability that runs to ₹1.5–₹3 lakhs if a CEIG rejection, rework, or ground fault event occurs.

According to the Central Electricity Authority’s General Safety Requirements for Electrical Installations, protective earthing conductors for solar PV systems at LT voltage level must comply with IS 3043:2018, and their minimum cross-section is determined by Table 5 of IS 3043 based on the phase conductor size — not by EPC convention or cost optimization.

The BOQ Leak Audit — Five-Category Proprietary Framework

The BOQ Leak Audit is Heaven Designs’ proprietary five-category framework for identifying and correcting the systematic BOQ errors that drain EPC margins. Applied before bid submission, it typically recovers ₹0.8–₹1.5 lakhs per MW in margin that would otherwise be lost post-commissioning. The five categories map directly to the five error types in this analysis.

Error 4 — Conduit and Cable Tray: Straight-Line vs Actual Run

Conduit and cable tray quantities are almost always taken from the single-line drawing using straight-line measurements between endpoints. The actual installation involves bends, offsets, T-junctions, reducers, and end sections that add 15–25% to the actual material quantity — before accounting for the accessories that most BOQs exclude entirely.

BOQ Item	Typical Error	Actual Field Requirement	Cost Impact (300 kW)
32 mm conduit runs	Straight-line only	+18% for bends and routing	₹4,000–₹8,000
Cable tray (150 mm wide)	Linear meters only	+20% for bends, elbows, T-pieces	₹6,000–₹12,000
Conduit fittings (unions, elbows, boxes)	Excluded	Required at every junction	₹3,000–₹5,000
Cable tray covers	Sometimes excluded	Required per IS 14772 in outdoor areas	₹4,000–₹8,000
Conduit end caps	Excluded	Required at every open end per safety	₹1,500–₹2,500
Total conduit/tray underestimate			₹18,500–₹35,500

This represents ₹0.6–₹1.2 lakhs/MW in conduit and tray underestimate alone. Multiplied across a 20-project EPC year at 300 kW average: ₹3.7–₹7.1 lakhs in systematic cable tray BOQ shortfalls annually.

Error 5 — Termination Hardware: The “Small Wares” Omission

MC4 connectors, compression cable lugs, cable glands, and cable ties are collectively called “small wares” and are frequently excluded from the BOQ on the assumption that they are covered by a lump-sum “miscellaneous” allowance of ₹5,000–₹10,000. The actual requirement for a 300 kW system is ₹26,000–₹38,000 — consistently 3–5x any miscellaneous buffer.

For a 300 kW system with 50 strings:

• TUV-certified MC4 connectors: 100 pairs at ₹85/pair = ₹8,500 (positive side) + 100 pairs = ₹17,000 total
• Y-branch connectors (where 3 strings combine): 20 sets at ₹280 = ₹5,600
• AC cable compression lugs: 48 lugs at ₹35 = ₹1,680
• Cable glands at DB board: 12 at ₹120 = ₹1,440
• Cable ties: 2 bags at ₹450 = ₹900
• Crimping ferrules (AC terminals): 80 at ₹12 = ₹960
• Total small wares: ₹27,580

The risk beyond cost: buying MC4 connectors at site in a rush means buying generic connectors from the local electrical market rather than TUV-certified MC4 from tier-1 manufacturers (Staubli, Amphenol, Renhe). Generic connectors may fail the UV resistance and contact resistance requirements of IEC 62852 within 3–5 years — generating arc-fault risk, warranty complications, and a field repair cost of ₹50,000–₹2,00,000 per connector failure event.

The 6% Margin Erosion — Calculation Walkthrough

Summing all five error categories for a reference 300 kW rooftop project at ₹3.8 Cr CAPEX (10% gross margin = ₹38L margin before overhead):

BOQ Error Category	Standard BOQ	Corrected BOQ	Under-Estimate	Cost Impact
DC string cable (4 mm² PV)	800 m	1,000 m	200 m	₹15,000
AC feeder (150 m run, 100 kW)	70 mm²	95 mm²	1 size up	₹24,000
Earth cable (16 mm² throughout)	800 m	1,100 m (25–50 mm²)	300 m + upsized	₹31,000
Conduit (32 mm)	450 m	531 m + fittings	81 m + fittings	₹9,000
Cable tray (150 mm)	280 m	336 m + accessories	56 m + accessories	₹11,000
Termination hardware (MC4, lugs)	₹8,000 lump	₹27,580 actual	₹19,580	₹19,580
Structural connection hardware	₹15,000 lump	₹28,000 actual	₹13,000	₹13,000
Total under-estimate				₹1,22,580

At ₹3.8 Cr project cost, ₹1.22 lakhs represents 3.2% of project cost — and 3.2% on a 10% gross margin project erodes 32% of the project margin. At 5.5–7% across a full project portfolio (larger projects have proportionally worse cable errors because longer runs amplify the deviation factor), the cumulative impact is a 55–70% margin reduction on the cable and BOS portion of every project.

BOQ Best Practices — Material Take-Off Methodology

The correct BOQ methodology for cable and BOS hardware is a material take-off (MTO) from the detailed design drawings — not a scaling estimate from the SLD or a parametric estimate from the module count. An MTO reads actual routed lengths from the CAD layout, identifies every junction and fitting from the conduit routing plan, and counts every connector from the string layout drawing.

MTO methodology for solar cable BOQ:

1. DC string cable: From the CAD string layout, measure each string’s positive and negative route separately. Apply the routing deviation factor for the mounting system type. Add 2 meters per string for combiner box connection length. Segregate by cable specification (4 mm², 6 mm², 10 mm² depending on string current).

2. DC trunk cable (combiner to inverter): Measure from the combiner box to the DCDB, then from DCDB to inverter. Apply a 1.15 deviation factor for cable tray routing. Confirm cable sizing against the combined string current and voltage drop limit.

3. AC feeder cable: Size independently for ampacity and voltage drop as described in Error 2 above. Measure from inverter output to ACDB, then from ACDB to metering point. Apply 1.10 deviation factor for indoor routing, 1.15 for outdoor cable tray.

4. Earth cable network: Count from the IS 3043 earthing design drawing — not from convention. The earthing design drawing should be a separate deliverable, not a rough sketch on the SLD. Without a dedicated earthing design, the earth cable BOQ is a guess.

5. Termination hardware: Count from the string layout — 2 MC4 pairs per string positive route, 2 pairs per negative route, plus Y-connectors where strings combine. Count compression lugs from the termination schedule in the SLD.

According to NREL’s Solar Photovoltaic System Cost Benchmark (2024), the BOS hardware and wiring cost category represents 10–14% of utility-scale solar CAPEX in developed markets — a proportion that is higher in India due to the longer cable runs required by ground-mount installations with distributed combiner boxes. Accurate BOQ of this category is proportionally high-value.

How a Design Firm’s BOQ Differs From an EPC’s Internal Estimate

An EPC’s internal BOQ estimate is typically produced by the project manager using a parametric scaling tool: X meters of cable per MW of capacity, based on historical project averages. This approach works as a budget-level estimate but fails as a procurement-grade BOQ because:

1. It does not account for project-specific routing geometry — a compact rooftop with a short run to the inverter has 30% less cable than a sprawling rooftop with the inverter at the far end.
2. It uses average routing deviation factors that are correct on average but wrong for any specific project.
3. It carries forward the systematic errors (earth cable under-specification, termination hardware omission) that are embedded in the historical average.

A design firm’s BOQ is produced from the actual layout drawing using measured routing, IS 3043 design, and itemized hardware count. The resulting BOQ is 15–22% higher in cable quantity on average than an EPC parametric estimate — but it matches the actual project cost. The higher BOQ wins fewer tenders only if the bid price is based directly on the BOQ. EPCs that separate their bid pricing (based on competitive analysis) from their procurement planning (based on the design firm’s MTO BOQ) use the accurate BOQ for material procurement and protect their margin internally.

The bankable PVsyst reports and BOQ guide covers how the yield model and the BOQ interact when producing a lender-ready project document set.

Pros and Cons — Design Firm BOQ vs In-House EPC Estimate

Verdict. A design firm BOQ costs ₹5,000–₹15,000 more to produce than an in-house parametric estimate on a 300–500 kW project — and recovers ₹80,000–₹1,50,000 in cable and BOS procurement cost on the same project. The ROI is 10–20x on the design cost. Every EPC that has gone through one post-commissioning cable true-up has this experience available to calculate — the question is whether to calculate it before or after the next bid is submitted.

How Heaven Designs Produces Project-Ready BOQs

Every Heaven Designs BOQ is built from the layout drawing up — cable quantities come from the actual routing plan with deviation factors applied, not from parametric scaling. The BOQ Leak Audit protocol is embedded in the deliverable workflow for every project. CEIG submissions that include a Heaven Designs BOQ alongside the SLD pass technical inspection at a higher rate because the earthing specification, protection device ratings, and cable designations are consistent between documents.

• Solar Rooftop Detailed Engineering Design — Full IFC pack including protocol-corrected BOQ with DC cable, AC feeder, IS 3043 earth cable, conduit with bend allowance, and itemized termination hardware.
• Electrical CEIG Drawings — CEIG-compliant electrical drawings that include the earthing design drawing required for IS 3043 compliance and the cable schedule linked to the SLD.
• Solar Ground Mount Design — Ground-mount BOQ covering underground DC cable, AC feeder, HT cable, protection equipment, and structural BOS hardware — all dual-criteria sized and routing-corrected.
• Bankable PVsyst Reports Guide — When lenders require both the yield model and the construction BOQ, see how Heaven Designs produces both as a combined submission package.
• Download a sample deliverable — See a real BOQ Leak Audit-corrected BOQ from a completed project before commissioning your own.

Contact Heaven Designs to commission a BOQ audit on your next bid before it is submitted — a one-time review that typically recovers 5–8× its cost in protected project margin.

FAQ

Why do solar EPCs consistently under-estimate cable quantities in BOQs?

The fundamental cause is that BOQ cable quantities are measured from 2D layout drawings using straight-line distances. Real cable routing adds 18–32% to the straight-line measurement through bends, drops to rail bottoms, and conduit routing deviations. EPCs that do not apply a routing deviation multiplier will always under-estimate cable quantity regardless of how carefully they read the drawings — the error is systematic and repeating.

What is the IS 3043 requirement for solar system earthing conductors?

IS 3043:2018 specifies that protective earthing conductors must be sized based on the phase conductor cross-section per Table 5 of the standard. For solar systems with 35 mm² AC phase conductors, the minimum protective earth is 16 mm² copper. For systems with 70 mm² phase conductors, the minimum is 35 mm² copper. The CEA General Safety Requirements reinforce these minimums and add requirements for the structure earth network (50 mm² copper) and module frame bonding (35 mm² copper). Under-specification of any of these creates CEIG non-compliance risk.

How does AC feeder voltage drop affect DISCOM technical compliance?

Most Indian DISCOMs specify a maximum voltage drop of 1.0–1.5% for the interconnection feeder between the solar inverter and the metering point. An AC feeder sized only for ampacity without voltage drop verification will frequently exceed this limit on runs above 80–100 meters at full load. A non-compliant voltage drop is identified during the DISCOM technical inspection and requires cable replacement or addition of a parallel run before the net meter connection is approved — generating ₹20,000–₹45,000 in unplanned material cost and a 2–4 week delay.

What is the correct routing deviation factor for DC string cable on a rooftop?

The routing deviation factor depends on the mounting system. For rooftop hook-and-rail systems where cable drops to the rail bottom before horizontal routing: 1.30–1.35. For rooftop rail-and-ballast systems with short runs: 1.20–1.25. For ground-mount systems with long horizontal rail rows: 1.25–1.30. Add 2 meters per string for combiner box connection tails regardless of mounting type. Never use a routing deviation factor below 1.20 for any solar installation.

How much does MC4 connector quality affect system reliability over 25 years?

TUV-certified MC4 connectors from tier-1 manufacturers (Staubli, Amphenol, Renhe/Multi-Contact) are tested to IEC 62852 and rated for 30-year outdoor service at UV exposure levels representing Indian radiation. Generic MC4 connectors from unverified sources may fail UV resistance within 3–5 years, leading to insulation degradation, increased contact resistance, and arc-fault risk. The cost difference between TUV-certified and generic MC4 is ₹60–₹100 per pair — negligible versus the ₹50,000–₹2,00,000 cost of a connector fault investigation and repair over a 25-year project life.

Can correct cable sizing improve system performance ratio?

Yes. An AC feeder undersized by one size creates a voltage drop that reduces the effective output voltage at the metering point — causing the inverter to derate output at the MPPT boundary. According to IRENA’s Utility-Scale Solar technical analysis, cable losses above 1.5% of system capacity directly reduce annual energy yield by the same proportion — a compound loss over a 25-year asset life that exceeds the cost of correct cable specification by 10–15x in present value terms.

How does a Heaven Designs BOQ differ from an EPC’s standard in-house estimate?

A Heaven Designs BOQ is built from the actual routing plan — not from a parametric scaling formula. It applies the Cable BOQ Accuracy Protocol to all five error categories: routing deviation multiplier on DC cable, dual-criteria AC sizing (ampacity + voltage drop), IS 3043 Table 5 earth cable sizing, conduit with 15–20% bend allowance, and fully itemized termination hardware. The resulting BOQ is typically 15–22% higher in cable quantity than an EPC parametric estimate — but it matches the actual project cost. Clients who use Heaven Designs BOQs report zero cable top-up purchases at delivery, compared to an industry average of 2–3 unplanned material orders per project.