Skip to content
Technical

Solar Inverter Wiring Diagram: Panels to Meter Board

A practical solar inverter wiring diagram guide covering the DC run, cable sizing, AC protection, earthing, metering and commissioning checks.

Keyur Rakholiya
Keyur Rakholiya CTO
July 23, 2026 Updated October 14, 2026 13 min read
Conceptual grid-connected solar path from PV strings through DC protection, inverter, AC protection and the consumer board to the service meter and grid
Connection blocks only. Meter position, protection, earthing and terminal wiring must follow the exact equipment documents and DISCOM requirements.

A solar inverter wiring diagram is four diagrams stacked together. The DC run from the array. The inverter terminal layout. The AC run to your distribution board. And the earthing that ties every metal part to one point. Add the utility metering and you have the complete circuit.

Most diagrams online draw the boxes and skip the parts that fail. Connector crimps. Isolator ratings. Residual-current device type. Torque on a DC terminal.

This guide covers an on-grid system from the panels to the meter and the grid. If your system has a battery, wire the battery side from the inverter and battery connection diagram as well. This guide walks the run in order, from the first connector pair on the roof to the meter at your service entrance: a numbered procedure, a worked cable-sizing calculation, single-phase versus three-phase neutral handling, the tests that happen before energising, and the errors that cause the most callbacks.

One condition applies throughout. This work must be carried out by a qualified, licensed electrician. The installation manual for your exact model and your local wiring rules take precedence over anything here. In India that means the Central Electricity Authority (Measures relating to Safety and Electric Supply) Regulations, 2023, IS 732:2019, and your DISCOM connection conditions. Where this page and your manual disagree, the manual wins.

Short version. A solar inverter wiring diagram runs in one direction. PV strings feed a DC surge protection device and a DC isolator, then the inverter. The output passes an AC isolator and a correctly rated breaker with the right residual-current device, then a generation meter where the DISCOM requires one, then the consumer board, then the bidirectional net meter and the grid. Every metal part shares one earth.

The complete wiring run, in order

Read the system as a single line from the roof to the street. Each stage has one job, one protective device, and one place it can be isolated. Order matters more than any individual component, because a device fitted on the wrong side of a junction protects nothing.

The complete on-grid wiring run in order, with what each component is sized against
StageComponentSized against
1PV modules in seriesModule Voc at lowest site temperature
2Connectors and DC cableIsc, run length, voltage-drop target
3DC surge protection deviceArray voltage class and exposure
4String fuses, where requiredModule maximum series fuse rating
5DC isolatorArray Voc and 1.25 times Isc, DC-rated
6Inverter DC inputMPPT window and per-input current limit
7AC isolatorInverter maximum continuous AC current
8AC breaker and RCD1.25 times inverter AC output current
9Generation meter, where requiredDISCOM specification
10Consumer distribution boardBoard rating and a spare way
11Bidirectional net meterDISCOM specification

The installation sequence follows from that order.

  1. Check the string, cable and protective-device schedules against the inverter datasheet.
  2. Mount the inverter to the clearances in the manual, out of afternoon sun.
  3. Run DC cable in UV-resistant conduit, with every roof penetration sealed.
  4. Crimp connectors with the matched die, one brand throughout, then tug-test each joint.
  5. Bond module frames and mounting structure to the earthing system first.
  6. Land DC conductors on the isolator and surge device, polarity marked at both ends.
  7. Verify polarity and open-circuit voltage at the inverter end, isolator open.
  8. Connect DC into the inverter, respecting maximum input current per MPPT.
  9. Run AC cable to the AC isolator, then to the breaker and RCD in the board.
  10. Complete earthing and bonding, and measure continuity end to end.
  11. Label every isolator, breaker and meter, and fix the shutdown placard.
  12. Run the commissioning tests, record them, then energise in the manual’s sequence.

DC side: strings, polarity, connectors and isolation

The DC side carries the highest voltage in a residential system and the least forgiving current.

Every DC decision starts from two module figures: open-circuit voltage at the coldest expected temperature, and short-circuit current.

String polarity is checked with a meter at the inverter end before the conductors go near the inverter. A reversed pair often reads zero rather than flagging itself, and the installer starts hunting a phantom module fault.

Use one manufacturer of MC4-style connector throughout, with the matching crimp die. Cross-mated pairs from two brands seat correctly and look identical, but the contact geometry differs enough to raise resistance. Never disconnect a PV connector under load.

The DC isolator must be DC-rated, not an AC switch pressed into service. It needs a voltage rating above array Voc at the lowest site temperature, a current rating at or above 1.25 times array Isc, both poles broken, and a position within reach of the inverter.

String fuses are a design decision, not a default. With one or two strings on an input, no fault combination can push reverse current past a module rating. At three or more parallel strings they become necessary, rated above string operating current and below the module maximum series fuse rating. The string sizing and overcurrent protection guide covers the arithmetic, and the string sizing calculator screens the series count.

A surge protection device belongs within about 10 metres of the inverter DC input; beyond that, fit a second at the array end. Array earthing is separate: frames and rails bond to the main earth, which gives earth-fault detection a reference.

Respect the published per-input limits. Qbits lists 20 A maximum DC input current per MPPT on the single-phase TLS and TLD on-grid models, and 180 V to 1000 V MPPT with 1100 V maximum DC for the 6 to 17 kW three-phase TLC family, according to the Qbits product specification. Other families list different windows, so use the numbers for your exact model and download the model datasheet before the design is frozen.

Worked example: sizing the DC cable to a voltage-drop limit

Cable size has two tests. First, current-carrying capacity under real conditions, including conduit grouping and ambient temperature. Second, voltage drop. Passing the first and failing the second gives a system that is safe and quietly lossy for its whole life.

For a two-wire DC run:

L is the one-way run in metres, I is operating current in amperes, ρ is resistivity in ohm sq mm per metre, and A is cross-section in sq mm. Copper sits near 0.0172 at 20 degrees C, but a rooftop conduit runs hot, so this example uses 0.022.

Worked example inputs. One string, single-phase system. Maximum-power current 11 A, maximum-power voltage 380 V, one-way run 25 m, copper, 4 sq mm.

Worked voltage-drop calculation for a 25 m, 11 A, 4 sq mm copper DC run
StepCalculationResult
Loop length2 × 25 m50 m
Numerator50 × 11 × 0.02212.1
Voltage drop12.1 / 43.03 V
As a percentage3.03 / 3800.80%
Power lost0.80% of 4,180 Wabout 33 W

At 4 sq mm the drop is 0.80%, inside the 1% designers usually target on a DC run and well inside the general limits IS 732:2019 sets for an installation. Step to 6 sq mm and the same run falls to 2.02 V, or 0.53%, which buys roughly 11 W at full output. A design sheet should show that tradeoff, not hide it.

The AC side uses the same formula with the AC current. A 5 kW inverter at 230 V draws about 21.7 A; over 12 m in 6 sq mm that gives 1.91 V, or 0.83%. For three-phase the multiplier changes from 2 to the square root of 3, because the return path is shared.

AC side: isolator, breaker rating, RCD type and the board

The AC side is where a rooftop system meets an installation that was never designed for a second source. Three components decide how that goes: a dedicated AC isolator, a breaker sized against the inverter rather than the load, and the correct type of residual-current device.

Size the breaker at about 1.25 times the inverter maximum continuous AC output current, rounded up to the next standard rating. A 5 kW single-phase inverter at roughly 22.7 A gives 28.4 A, which lands on a 32 A miniature circuit breaker. The cable feeding it must carry at least the breaker rating under its installation conditions.

A transformerless inverter has no galvanic isolation between array and grid, so an earth fault can produce residual current with a smooth DC component. That decides the RCD type.

Residual-current device types and their suitability on a transformerless solar AC circuit
RCD typeDetectsUse on a solar AC circuit
Type ACSinusoidal AC residual current onlyNever acceptable
Type AAC plus pulsating DC residual currentAcceptable where the manual says so, typically when the inverter has integrated DC residual-current monitoring tested to IS/IEC 62109-2
Type BAC, pulsating DC and smooth DC residual currentRequired where the inverter has no integrated DC fault-current detection, or where the manual specifies it

The AC isolator belongs next to the inverter, lockable, and the final connection should use a dedicated way in the board.

Earthing and equipotential bonding

Earthing does two jobs. It gives fault current a low-impedance path back to the source so protective devices operate. And it holds every exposed metal part at the same potential, so nobody takes a shock from touching two things at once. That covers frames, rails, the inverter enclosure, isolator enclosures and metal conduit.

The rule people skip is the important one. There is one earthing system, not two. A separate earth pit for the array, unbonded to the building main earthing terminal, creates a potential difference that discharges through whatever bridges the two systems.

  • Bond module frames to the rails, and the rails through to the main earth, at the specified size.
  • Bond to clean, un-anodised metal with a washer that bites through the coating. An anodised frame is an insulator on its surface.
  • Run a continuous protective conductor from the inverter earth terminal to the board, not to a local rod.
  • Bond any building lightning protection system to the same main earthing terminal.
  • Measure and record earth continuity at commissioning rather than assuming it.

On a transformerless inverter the array is not galvanically separated from the grid, so functional earthing of a live PV conductor is generally not permitted. The solar inverter grounding guide covers the measurement method.

Where the generation meter and net meter sit

Metering position is set by the DISCOM, so getting it wrong means a failed inspection rather than a technical fault. Requirements vary by state and by DISCOM, and the threshold for a separate generation meter is not uniform.

The bidirectional net meter replaces the existing service meter at the service entrance, upstream of the whole installation including the point where the solar circuit joins the consumer board. That is what lets it record import and export separately on one connection.

The generation meter, where a state requires one, sits between the inverter AC isolator and the consumer board. Being on the PV branch only, it records gross PV output regardless of how much the premises consumed.

A current transformer for export control goes on the incoming supply conductor, on the grid side of the PV connection point, so it sees net flow. On the load side, or reversed, the control logic reads the wrong sign. The net metering application guide covers the approval sequence.

Single-phase versus three-phase wiring, and the neutral

The DC side of a single-phase and a three-phase system looks nearly identical. The AC side does not. Three things change the drawing: conductor count, neutral handling and phase rotation.

Single-phase versus three-phase AC wiring differences
ItemSingle-phaseThree-phase
AC conductorsLine, neutral, protective earthThree lines, earth, neutral where required
Nominal voltage230 V415 V
NeutralAlways connected as a voltage referenceModel-dependent; some units are three-wire plus earth
Phase rotationNot applicableWrong rotation can block synchronisation
Voltage rise at the connection pointHigher, all current in one lineLower, current split across three lines
Qbits on-grid seriesTLS and TLDTLC

Neutral handling trips installers moving between brands. A three-phase inverter that needs a neutral will not run correctly without it, and one that does not may flag a fault if a neutral is landed on the wrong terminal.

Voltage rise is the practical reason large single-phase systems get pushed to three-phase. Every exported ampere lifts the voltage at the connection point slightly, and on a weak feeder that rise can hit the inverter overvoltage limit and trip it. The Qbits on-grid inverter range spans both configurations.

Terminations, torque, and why a loose DC joint starts a fire

A termination that is not tight has contact resistance. Current through contact resistance produces heat. Heat oxidises copper, oxide raises resistance further, and the fault accelerates. Left alone it ends in a glowing connection or an arc.

  1. Use the torque value printed in the manual for that terminal. Do not guess, and do not borrow a figure from another model.
  2. Use a calibrated torque screwdriver, not feel. Torque values differ between models and between terminals on the same model, so only the manual figure counts.
  3. Match the ferrule or lug to the conductor size and use the correct crimp die. A crimp made with the wrong die is loose inside a joint that looks finished.
  4. Mark each completed termination with a torque stripe, so a later inspection can see whether it has moved.
  5. Re-check accessible terminations at the first service visit. Thermal cycling in Indian summers loosens joints that were correct on day one.

Labelling and signage at the isolator and the board

Labelling exists for the person who arrives when you are not there: a technician, an inspector, or a fire crew isolating a building. An unlabelled installation hands them a main switch that does not de-energise everything. It is also among the first things a DISCOM inspector checks.

If the model changed between quotation and installation and the drawing still shows the original, every future service visit starts from wrong information.

Commissioning checks before you energise

Commissioning is a test sequence with a written record, not a switch-on. IS/IEC 62446-1 sets out what a grid-connected PV commissioning report contains, and most manuals follow the same structure. These checks happen with the DC isolator open and the AC breaker off.

  1. Visual inspection of the mechanical installation, cable support, conduit sealing and enclosure closure.
  2. Continuity of the protective earthing conductor and bonding, measured end to end.
  3. String polarity at each connector pair, confirmed with a meter.
  4. Open-circuit voltage per string, compared against the calculation and the other strings. A string low by roughly one module means a wrong count or a failed joint.
  5. String short-circuit or operating current, measured with an appropriate tester.
  6. Insulation resistance of the DC conductors to earth, at the specified test voltage.
  7. AC checks: supply voltage, neutral continuity, and phase rotation on three-phase.
  8. Residual-current device tested for trip current and trip time.
  9. Breaker and RCD ratings confirmed against the design schedule and the labels.
  10. Torque verification on every accessible termination.
  11. Inverter configuration: correct grid profile, anti-islanding confirmed, export behaviour set to what the DISCOM approved.
  12. Energisation in the manual’s order, then a functional check of output, monitoring and shutdown.

Results go on the commissioning sheet and are handed over with the drawings. The commissioning walkthrough covers the documentation set.

Common wiring errors, and the myths behind them

Most wiring failures trace back to a shortcut that sounded reasonable at the time. The table pairs each error with the failure it causes.

Common wiring errors and the failure each one causes
Wiring errorWhat it causes
Reversed string polarity at an inputNo DC reading, a polarity fault, or internal damage
Unequal module counts paralleled on one MPPTThe weaker string drags the pair off its maximum power point
Mixing connector brands on one runRaised contact resistance at every cross-mated joint
AC breaker oversized for the cableFault current the breaker will not clear fast enough
Type AC RCD on a transformerless circuitSmooth DC residual current can blind the device
Earth landed on an anodised frameHigh-resistance bond, so a real fault may go undetected
Separate, unbonded earth pit for the arrayPotential difference between array and building earth
Export CT reversed or on the load sideThe system exports when it should throttle
DC isolator out of reach of the inverterNo safe way to isolate for service
Unsealed penetration or non-UV conduitInsulation failure and earth faults within two monsoons

Three myths are worth naming.

A standard recipe works. “Use 4 sq mm and one fuse” is the most repeated line in rooftop solar and it is not a design. Cable size follows current, route and voltage drop. Fuse presence follows the number of parallel strings.

Switching off the AC makes the system safe. It does not touch the DC side. A rooftop array can sit at several hundred volts on an overcast day.

A bigger breaker is a safer breaker. A breaker protects the cable, not the inverter. Oversizing it removes the protection the cable relied on.

The Bottom Line

A solar inverter wiring diagram is a sequence with a protective device at every stage and one shared earth running through all of it. Internal power electronics sit in the inverter circuit diagram guide, and battery bank wiring in the battery connection diagram.

  • Ask for the cable-sizing calculation, string schedule and protective-device schedule before work starts, and check the RCD type against the inverter manual.
  • Insist on an as-built single-line diagram, the commissioning record and the full installation manual at handover.
  • Confirm the specifications you are designing around, then talk to the Qbits technical team about installation documentation for your exact inverter.
Qbits models

Specs below come from each model's datasheet. Confirm the exact model and current written terms with Qbits before you buy.

Browse all on-grid inverters →

FAQ

Frequently Asked Questions

Can I wire a solar inverter myself using this diagram?
No. Electrical installation work on a grid-connected system must be carried out by a licensed, competent electrician, and the Central Electricity Authority (Measures relating to Safety and Electric Supply) Regulations, 2010 set that expectation in India. A PV array produces voltage whenever it is illuminated, so switching off the AC supply does not make the DC side safe. Your DISCOM will also require the installation to be certified before a net meter is fitted. Use this guide to understand the design and to ask your installer better questions.
What size DC cable does a rooftop solar inverter need?
There is no single answer, because the size depends on string current, run length, the voltage-drop target and the installation conditions. Size for current-carrying capacity first, then check voltage drop with Vdrop = (2 × L × I × ρ) / A for a two-wire DC run. Indian rooftop designs commonly land on 4 sq mm or 6 sq mm copper for single-string residential runs, but a long run or a hot conduit can push that up. The designer must confirm the number against the module and inverter datasheets.
Do solar strings need fuses?
Fuses protect a string against reverse current fed back from the other strings paralleled with it. With one or two strings on an input there is no combination that can exceed a module's reverse-current rating, so fuses are normally omitted. Once three or more strings are paralleled on the same input, string fuses are generally required. The fuse rating must sit above the string operating current and below the module's maximum series fuse rating printed on the module datasheet.
What type of RCD should a solar inverter circuit use?
Never Type AC. A transformerless inverter has no galvanic isolation between the array and the grid, so it can produce a residual current with a smooth DC component that a Type AC device cannot detect. The installation manual states whether a Type A device is acceptable, which is usually the case when the inverter has integrated DC residual-current monitoring, or whether an external Type B device is required. Follow the manual, because it reflects how that specific model was tested to IS/IEC 62109-2.
Where does the net meter go in the wiring diagram?
The bidirectional net meter replaces the existing service meter at the service entrance, upstream of the entire installation. Everything else, including the point where the solar AC circuit joins the consumer board, sits downstream of it. A separate generation meter, where your state requires one, sits between the inverter AC isolator and the consumer distribution board so it records only PV output. Position and sealing are decided by the DISCOM, not by the installer.
Is three-phase wiring different from single-phase for a solar inverter?
Yes, in three ways. Phase rotation matters on a three-phase inverter, and the wrong rotation can prevent the unit from synchronising. Neutral handling differs by model, because some three-phase inverters need a neutral for voltage measurement while others run three-wire plus earth. Three-phase also spreads output current across three conductors, which reduces the local voltage rise that causes grid-overvoltage trips on large single-phase systems.
Why do loose DC terminations cause fires?
A loose joint has contact resistance, and contact resistance under current produces heat. Heat oxidises the copper, which raises the resistance further, so the fault accelerates. If the joint separates under load, a DC arc forms, and unlike an AC arc it has no current zero crossing to extinguish it. That is why every DC termination must be torqued to the figure in the manual and re-checked at commissioning.
What has to be labelled on a solar installation?
The DC isolator needs a warning that live DC is present whenever the modules are illuminated. The main distribution board needs a dual-supply warning so anyone isolating the mains knows a second source exists. Each isolator, breaker and meter needs a durable identification label that matches the as-built single-line diagram. A shutdown procedure placard should be fixed near the inverter for emergency responders.
Does bad wiring void an inverter warranty?
It can. Most manufacturers exclude damage caused by incorrect installation, reversed polarity, exceeding the rated input voltage or current, inadequate earthing, and water ingress through an unsealed gland. Keep the commissioning records, torque checks and as-built drawings, because they are the evidence that the installation followed the manual. Obtain the current written warranty terms for the exact quoted model before purchase rather than relying on a summary.
Keyur Rakholiya
Written by

Keyur Rakholiya CTO, Qbits Energy

Listed by Qbits as its CTO. Current role, technical responsibilities and credentials require confirmation by the company owner.

Related Reading
Guide 12 min

Solar Inverter Grounding, India EPC Safety Guide

Proper grounding protects equipment, people, and warranty validity in every solar PV installation. Here is the complete grounding framework aligned to IS 3043 and CEA Regulations 2013 - covering earthing fundamentals, conductor sizing, electrode design, SPD integration, hybrid system requirements, and the 8 mistakes that void warranties.

Get Started

Check the exact inverter before finalising a design

Use the current model datasheet and installation instructions to verify voltage, current, phase and application limits.