Solar System Not Producing Enough Power: Diagnostic Guide

If your solar system is not producing enough power, do not diagnose it from one low day, one high bill, or a peak-power number below the panel nameplate. First prove that the production data are complete. Then compare actual energy with an appropriate expectation for the same dates, weather, shade, and operating limits. The shape and timing of the shortfall will tell a qualified professional what to investigate.

Teamsun publishes a solar maintenance and repair service, but serviceability depends on the system owner, address, equipment, monitoring access, warranties, and problem. This guide does not promise remote diagnosis, warranty approval, parts availability, or support for every brand. It gives an existing solar owner a safe way to assemble evidence and request the right diagnostic scope.

The short answer: Compare energy with energy—usually actual kilowatt-hours against the original monthly production estimate or a weather-aware model—not instantaneous kilowatts against the array’s DC nameplate. Rule out missing monitoring intervals, recent weather, new shade, snow, soiling, and known export controls. Preserve screenshots and logs. Do not open equipment, climb on the roof, test energized conductors, or repeatedly reset an unknown system.

Is the solar system actually underproducing?

A system is underproducing only when reliable measured energy is materially below a suitable expectation after accounting for the comparison period and operating conditions. A disappointing screen or electric bill is a reason to investigate, not proof of an equipment failure.

Start by separating four quantities that owners often mix together:

Quantity	Unit	What it tells you	What it does not prove
Array nameplate	kW DC	Rated module capacity under laboratory standard test conditions	The AC peak you should see at your home
Instantaneous power	kW	Output at one moment	The day’s or month’s total production
Energy produced	kWh	Electricity generated over a period	How much of that energy the home used or exported
Utility net usage	kWh imported or exported	Energy exchanged with the grid after on-site use	Total solar generation behind the meter

The U.S. Department of Energy explains that PV module ratings use standard test conditions of 1,000 watts per square meter of irradiance and a 25°C cell temperature. Actual sunlight is usually different, and operating cells are often hotter, so nameplate capacity is not a promise that the monitoring graph will reach that number (DOE performance guidance).

The utility bill creates another common false alarm. A standard bidirectional utility meter generally measures net imports and exports at the grid connection. It does not see solar energy consumed instantly inside the house. A rising bill can result from higher home consumption, a changed rate, an estimated meter read, a billing-period difference, lower solar production, or several of those at once.

Use this first-pass test:

1. Choose a complete period: at least seven days for a sudden failure pattern, or a complete billing month for an energy comparison.
2. Export solar production in kWh from the inverter or monitoring platform.
3. Mark missing intervals, device communication warnings, time-zone changes, and gateway outages.
4. Find the proposal’s monthly production estimate, a prior-year production export, or a current model for the same location and system.
5. Record major differences in weather, tree shade, roof work, snow cover, construction, utility controls, and system availability.
6. Compare the same dates and the same energy unit.

DOE’s homeowner guidance says a year-over-year production drop greater than 10% can be a screening signal for a maintenance issue (DOE guide for homes with solar). That is not a universal failure threshold, performance-guarantee rule, or warranty standard. A contract may use a different method, and annual weather can vary. Use the number to decide whether to investigate—not to select a replacement part.

What should expected solar production be compared against?

The strongest expectation is a model or contractual baseline that matches the installed design and the actual comparison period. A generic online estimate is useful for reasonableness, but it cannot by itself prove that an installed system breached a guarantee or that a particular component failed.

Use baselines in this order:

Baseline	Best use	Necessary adjustment	Main limitation
Contract or proposal monthly estimate	Checking whether the installed system broadly tracks the sold design	Confirm final as-built size, orientations, shade assumptions, loss factors, and contract measurement rules	Sales models may use typical-year weather, not the weather that occurred
Same month in prior years	Detecting a change in a stable operating system	Account for weather, new shade, snow, downtime, equipment changes, and normal aging	Last year may also have incomplete data or unusual weather
NREL PVWatts estimate	Independent reasonableness screen	Enter the installed size, array type, tilt, azimuth, losses, inverter efficiency, and DC-to-AC ratio as accurately as possible	Uses modeled assumptions and historical weather; it is not a site inspection
Weather-aware performance model	High-confidence performance investigation	Use coincident irradiance, temperature, equipment state, curtailment, and availability	Requires reliable sensors/data and competent analysis
Neighbor or online screenshot	Almost none	Systems would need the same size, geometry, shade, equipment, weather, controls, and data completeness	Those conditions rarely match

NREL’s current PVWatts Version 8 adds improved thermal and inverter models plus optional monthly soiling and snow inputs. It also warns that results depend on assumptions and historical weather. PVWatts supplies a range based on many years of weather variability; it does not reconstruct the clouds over your roof during the week you are disputing.

For a simple screening calculation, use:

Production gap (%) = (adjusted expected kWh − actual measured kWh) ÷ adjusted expected kWh × 100

“Adjusted expected” is the important part. Suppose the original estimate says 900 kWh for May, the monitoring platform contains every interval, and a weather-aware review estimates that the actual May solar resource should have produced 810 kWh. If the system measured 760 kWh, compare 760 with 810, not automatically with 900. The 50 kWh difference still deserves classification, but the 140 kWh sales-model gap does not all prove equipment loss.

DOE’s existing-system guidance defines a performance ratio as actual production divided by model production when the model uses the same measured solar resource, temperature, and age (DOE existing PV operations and maintenance). That is why a weather-matched analysis is stronger than comparing two monthly totals without context.

Before paying for analysis, request these original inputs from the seller or system owner:

• Final installed DC size and inverter AC rating.
• Module count, model, orientation, tilt, and roof-plane grouping.
• Shade study or shading loss used in the proposal.
• Monthly and annual production estimate.
• Assumed soiling, snow, mismatch, wiring, availability, degradation, and clipping losses.
• Performance-guarantee language, exclusions, measurement source, and claim deadline.
• As-built electrical one-line and permission to operate.

If the final system differs from the proposal, the expected model should be rerun with the as-built configuration. Otherwise, a comparison may be precise but wrong.

Is the production reading real, delayed, or measured at the wrong point?

An app can show low or zero production even while the array is generating. Communication loss, missing intervals, an incorrect consumption-meter configuration, site-transfer problems, and time-zone errors can distort the screen without creating the same loss at the production meter.

Follow the energy through the measurement layers:

Layer	What to collect	A discrepancy may mean
Inverter or module-level production	Daily/monthly kWh, device map, event log, last-update timestamp	Equipment output, device reporting, or portal-data issue
Gateway or data logger	Connection status, firmware, last contact, internet changes	Missing telemetry rather than stopped generation
Dedicated production meter, if installed	Register readings and interval export	Independent production evidence or meter/configuration issue
Consumption monitor	Home load, grid import/export, CT configuration	Reversed/misassigned current transformers or calculation error
Utility net meter and bill	Interval imports, exports, read dates, rate	Grid exchange after self-consumption—not gross PV production

Enphase’s current troubleshooting guidance says a gateway that is not reporting can prevent data from reaching the cloud even while microinverters may still produce. It also says abnormal production or consumption displays can be caused by current-transformer or meter installation and should be handled by a certified professional. A “not reporting” device and a nonproducing device are therefore different hypotheses.

SolarEdge’s owner support similarly separates production problems from communication problems. Its guidance uses status indicators and the monitoring portal to narrow the category, but tells owners not to repair inverters or optimizers themselves.

Tesla distinguishes app production from utility-meter data and notes that a gateway communication interruption can stop data reporting. Manufacturer screens, indicators, and authorized homeowner steps vary. Follow the manual for the exact model rather than applying a reset sequence found for another inverter.

Document these data-quality checks before treating missing telemetry as missing energy:

• Does the app show a current “last updated” time?
• Are there blank hours or days in the export?
• Did the internet router, network name, password, modem, or cellular service change?
• Does the inverter’s owner-visible local screen or status indicator show production?
• Does the production meter advance during daylight, if the system includes an owner-readable meter?
• Do utility exports or daytime imports change in a way consistent with generation?
• Does one device fail to report, or does the entire gateway stop reporting?
• Did the problem start after monitoring ownership or installer administration changed?
• Are production and consumption values swapped, negative, duplicated, or physically impossible?

Do not “repair” a data gap by estimating missing production and saving it as actual. Keep measured, reconstructed, and modeled values in separate columns. That protects a warranty claim and lets the technician state how much uncertainty remains.

This page intentionally does not become a monitoring-app troubleshooting guide. B167 owns app connectivity and reporting failures. Here, the question is narrower: is the low-production symptom supported by an independent measurement, or is the evidence incomplete?

Which weather and site conditions can explain lower output?

Weather, season, shade, snow, soiling, and heat can reduce output without proving a failed component. The diagnostic clue is whether the timing and shape of the reduction match the environmental condition and whether that condition was included in the original model.

Weather and seasonality

Daily production changes with available sunlight, day length, sun angle, cloud cover, rain, fog, temperature, and snow. A clear March day can briefly show high power because cells are cool, while a hot clear summer day may produce more daily energy but a lower peak. Compare complete energy periods, not isolated peaks.

NREL’s National Solar Radiation Database provides solar-radiation and meteorological data used for resource analysis. A professional performance review can align the system’s measured intervals with coincident irradiance and temperature. A typical meteorological year remains useful for design, but actual-time weather is better for diagnosing a specific period.

New or underestimated shading

Shade can change after installation. Trees grow and leaf out; nearby construction adds an obstruction; roof equipment moves; snow or debris affects a roof plane. Shade also moves by hour and season, so one ground photo is not a complete study.

Look for a repeatable reduction at similar times on clear days, differences between roof planes, or a module-level cluster that follows an obstruction. Do not assume module-level power electronics eliminate all shade loss. NREL’s research on shaded residential systems emphasizes that detailed shade modeling matters when estimating performance (NREL shaded-system study).

Soiling

Dust, pollen, bird debris, leaves, and other deposits can reduce light reaching modules. The loss varies by climate, rainfall, array angle, nearby sources, and cleaning history. NREL maps soiling variability and defines an insolation-weighted soiling ratio for performance assessment (NREL soiling research). That variability is why a universal “dirty panels lose X%” claim is not a diagnosis.

Before buying cleaning, compare production before and after rainfall, visually inspect from the ground, check whether the original model included soiling, and ask what measured evidence suggests cleaning is economic. Avoid roof access or abrasive tools. Follow the module, racking, roof, and system-owner instructions.

Snow

Snow cover can cause partial or total short-term loss. It may slide or melt unevenly, causing different roof planes to recover at different times. Do not climb onto a snow-covered roof or use a hard-edged tool. Tesla’s owner guidance specifically says never shovel snow from panels or use a hard-edged tool (Tesla performance support). The exact safe practice comes from the equipment manufacturer and roof conditions.

Use this decision table:

Pattern	Environmental explanation worth checking	Evidence that strengthens it	Evidence that points elsewhere
Low only on cloudy days	Available irradiance	Weather record and similar behavior across all devices	One device remains low on clear days
Seasonal morning/afternoon loss	Sun angle and shade	Repeats by time and roof plane	Abrupt onset without a site change
Loss after nearby construction	New obstruction	Site photos and shade analysis	No affected time window or roof plane
Uniform gradual loss in a dry/dirty period	Soiling	Ground-visible deposits, rain response, irradiance-corrected trend	One string or device missing
Zero/low while snow-covered	Snow	Photos from a safe location and recovery after melt	Persistent loss after full clearing
Lower midday peak on hot clear days	Temperature response	Weather-aligned power curve	Sudden step change at all temperatures

Environmental causes can coexist with faults. A weather-aware model should remove what the weather explains and leave the unexplained residual for diagnosis.

Are clipping and curtailment normal—or evidence of a problem?

Clipping and curtailment both limit output, but for different reasons. Clipping is an inverter power ceiling designed into many systems. Curtailment is an external or configured instruction to reduce production. Neither should be labeled a failed panel without examining the design, settings, and event history.

Inverter clipping appears when available DC power exceeds the inverter’s AC output capability. The graph often has a repeatable flat top around the inverter limit during high-resource periods. Designers may intentionally use more DC module capacity than inverter AC capacity because the array reaches nameplate conditions infrequently. The important buying question is whether the original annual model included the chosen DC-to-AC ratio and expected clipping.

NREL’s clipping research explains that DC capacity above inverter AC capacity can cause output limitation and that clipping can interact with performance analysis (NREL inverter-clipping study). A flat-topped graph is a clue, not a sufficient design review; export limits, device settings, and measurement caps can produce similar shapes.

Curtailment can occur when a utility or control system restricts export, the grid is outside permitted voltage or frequency, an export-control setting is active, or a battery is full in a no-export configuration. Tesla’s performance support describes cases in which production is reduced after a battery fills when export is not allowed. Other brands and utilities implement controls differently.

Ask for these records:

• Inverter AC rating and array DC rating.
• Original clipping/loss assumption.
• Interconnection approval and approved export capacity.
• Export-control or zero-export configuration.
• Battery operating mode, state of charge, and reserve setting.
• Inverter/event log with timestamps.
• Utility voltage or frequency events documented by qualified testing or utility records.
• Firmware or control-setting changes near the symptom onset.

Graph shape	Possible category	What proves it
Smooth, repeatable flat top on clear days	Designed clipping	Inverter limit, configuration, modeled clipping, and no fault events
Abrupt reductions when battery becomes full	Export-control curtailment	Battery/control log and approved export setting
Intermittent stop/start around the same grid conditions	Grid-related protective response	Time-aligned inverter events and qualified voltage/frequency evidence
New lower cap after service or firmware/configuration change	Changed setting or component	Before/after configuration and commissioning record
One string/device stays low while system cap is not reached	Local equipment, connection, shade, or data issue	Electrical and device-level diagnostic evidence

Do not ask an owner to change grid-protection, export-control, battery, or inverter commissioning settings. Those settings may be part of the approved interconnection and require authorized access, a licensed electrician, the manufacturer, or the utility.

What does the shape of low solar output suggest?

The pattern narrows the inspection scope, but it does not identify a replaceable part by itself. A technician should connect the pattern to measurements, equipment logs, the one-line diagram, and the manufacturer’s current procedure.

Use this symptom-led decision tree:

1. Is there an immediate safety condition? If yes, keep away, use emergency or utility help as appropriate, and stop normal troubleshooting.
2. Is the data complete and current? If no, preserve the gap and resolve monitoring or find another measurement point.
3. Is the comparison valid? Match kWh, dates, system configuration, actual weather, shade, snow, soiling, and operating limits.
4. Does the whole system or only part of it look low? Whole-system and device-specific problems need different evidence.
5. Is the pattern gradual, abrupt, intermittent, or capped? The transition time often matters more than the current daily total.
6. Who owns the equipment and warranty path? Obtain authorization before intrusive work.
7. What professional test can confirm or reject each hypothesis? Require a written diagnostic deliverable before accepting a replacement quote.

The pattern-to-scope table is a starting point:

Observed pattern	Plausible categories—not a diagnosis	Professional checks to quote
Entire system at zero	Grid outage, open protection/device, inverter shutdown, communication-only outage, ownership/platform issue, fault	Independent production evidence, inverter status/events, approved shutdown state, AC/DC checks by qualified person, monitoring status
Whole system consistently low	Weather, soiling, new shade, changed configuration, inverter derating, metering error, export control	Weather-corrected performance, site/shade review, event logs, configuration, calibrated meter comparison
One roof plane or string low	Shade, snow/debris, string wiring/connector/fuse issue, array-map error, module group fault	String/branch measurements, plan-to-field mapping, safe roof/thermal inspection where justified, equipment-specific tests
One module-level device low	Local shade/debris, module, microinverter/optimizer, wiring, map or communication issue	Device history, adjacent-device comparison, manufacturer diagnostic, electrical/roof inspection
Output falls when hot	Normal temperature response or equipment thermal derating	Weather-aligned curves, ventilation/site condition, inverter temperature/events, manufacturer limits
Midday flat top	Clipping, export cap, control setting, or data cap	DC-to-AC ratio, approved export, settings, model, measured power
Abrupt gaps or stop/start	Grid protection, curtailment, fault, gateway/data gap, intermittent connection	Time-aligned event logs, utility data, independent measurement, qualified electrical testing
Bill rises but solar graph looks normal	Higher consumption, rate/billing change, meter/CT error, reduced self-consumption, or production issue hidden by reporting	Production export, utility intervals, consumption data, bill read dates/rates, meter configuration

DOE’s monitoring guidance describes electrical measurements at the system, inverter, string, and module levels, together with weather inputs, modeled performance, alarms, and curtailment signals (DOE solar monitoring platforms). That layered approach is more defensible than replacing the component named in an app alert.

A qualified diagnostic may include calibrated production measurement, inverter and gateway logs, comparison with the as-built one-line, string or branch current/voltage testing, insulation or ground-fault procedures, CT orientation and configuration review, controlled thermal inspection, shade analysis, and manufacturer case support. The scope depends on the architecture. The owner should not perform these electrical or rooftop procedures.

This distinction prevents cannibalization with related guides. B168 will address exact error codes; B169 covers storm-triggered inspections and insurance evidence; B157 covers planned maintenance; B173 covers systems whose original installer is gone. This page owns the evidence path from a low-production symptom to a defensible diagnostic scope.

What can a homeowner safely check before calling for service?

Homeowner checks should stop at records, ground-level observation, owner-visible status, and the exact user actions published for the installed model. Solar modules can generate DC electricity whenever exposed to light, so turning off an AC breaker does not make every solar conductor safe.

OSHA identifies shock, electrocution, burns, arc flash, falls, and overhead-line contact among solar electrical hazards (OSHA solar electrical safety). Its page is written for worker protection, but it explains why internet-guided electrical troubleshooting is inappropriate for a homeowner.

Safe evidence collection

• Save screenshots of daily, monthly, annual, device, alert, and last-update screens.
• Export raw production data when the platform allows it.
• Photograph equipment labels and visible status indicators without opening covers.
• Photograph the array only from a safe ground location.
• Record storms, outages, internet changes, tree work, roof work, construction, and electrical work.
• Download utility interval data and bills with actual read dates.
• Gather the proposal, final contract, warranties, permit set, one-line, interconnection approval, and permission to operate.
• Record exactly which owner-manual steps were taken and when.

Stop and escalate

Do not touch or approach equipment if you observe smoke, flame, arcing, a burning odor, shock or tingling, exposed conductors, severe physical damage, water inside electrical equipment, a displaced module, or a damaged/overheating battery. Keep people away and contact emergency services, the utility, and the equipment owner or manufacturer as appropriate.

Do not:

• Climb on the roof to inspect, wash, brush, or clear snow.
• Open an inverter, disconnect, combiner, panel, gateway, battery, or junction box.
• Unplug module, optimizer, microinverter, string, or battery connectors.
• Use a handheld meter on solar or service conductors.
• Change commissioning, grid, export, battery, or protection settings.
• Repeatedly cycle breakers or disconnects to clear a trip.
• Bypass an alert, interlock, rapid-shutdown function, or protective device.

DOE’s existing-system guidance cautions that simply power-cycling after a trip is not a substitute for investigating ground-fault conditions. Tesla’s owner support likewise tells owners not to open panels, disconnect wiring, or attempt solar repairs because of high voltage. The safe boundary is observation and documented, model-specific owner guidance—not improvised electrical work.

State licensing adds another boundary. Connecticut assigns limited PV work to PV-1/PV-2 licensees but reserves connection to existing building panels, circuits, wiring, junction boxes, and the utility meter for E-1/E-2 licensees (Connecticut DCP solar trade scope). Massachusetts’ electrical board guidance treats PV components as electrical work requiring licensed electricians and electrical permits (Massachusetts Board PV interpretation). Rhode Island’s renewable-energy professional policy says licensed electricians are required for renewable electrical work, while its REP route covers ancillary non-electrical work (Rhode Island DLT renewable-energy policy). The exact contractor, permit, and inspection requirements remain address- and scope-specific.

What documents should go into a solar diagnostic request?

A complete diagnostic packet reduces paid discovery time and helps the service company decide whether it can support the system. It also preserves evidence before monitoring history, equipment state, or ownership changes.

Create six folders:

1. System identity

• Property address and utility account service location.
• Final module count and DC size.
• Inverter/microinverter/optimizer, gateway, production meter, battery, and module models.
• Serial numbers shown in context.
• As-built one-line and array layout.
• Commissioning date, permission to operate, and interconnection approval.

2. Ownership and authorization

• Purchase, loan, lease, or PPA agreement.
• Home-sale or estate transfer documents if relevant.
• Monitoring-platform owner and administrator.
• System-owner authorization for third-party inspection.
• Insurance claim number when storm or property damage may be involved.

3. Production evidence

• Raw interval export, not only screenshots.
• Daily, monthly, annual, and device-level views.
• First low date and last known normal date.
• Missing-data periods and last-update timestamps.
• Original monthly estimate and performance guarantee.
• Prior-year production for matching dates.

4. Context

• Utility interval imports and exports.
• Bills before and after the change.
• Actual-weather record or weather-aware analysis if available.
• New trees, construction, roof equipment, snow, debris, or soiling.
• Router, firmware, electrical, roof, battery, or utility changes.

5. Events and actions

• Alert text and error codes copied exactly.
• Inverter/gateway event logs.
• Photos of owner-visible lights or displays.
• Date/time of outages or abnormal operation.
• Every reset or support step already attempted.
• Manufacturer or installer case numbers and written responses.

6. Warranty and service history

• Installer workmanship warranty.
• Module, inverter, optimizer/microinverter, gateway, meter, and battery warranties.
• Product registrations and receipts.
• Previous service tickets, RMA numbers, invoices, and replaced parts.
• Labor protection plan or service-contract terms.

Use a one-page symptom summary:

Field	Example observation—not a conclusion
First noticed	“July 14 monthly view looked lower than proposal”
Last known normal	“Complete daily data through June 28”
Measurement	“Gateway export: 412 kWh July 1–31”
Comparison	“Proposal: 620 kWh; PVWatts TMY: 590 kWh; actual-weather model not yet run”
Data gap	“No portal intervals July 6–8; utility exports continued”
Pattern	“East roof devices recover by 11 a.m.; west group uniformly lower after 2 p.m.”
Recent change	“Tree leaf-out and new neighboring dormer; no electrical work”
Safety observation	“No smoke, odor, water, exposed conductor, impact, or visible displacement from ground”

Do not label the cause in the intake form unless it was confirmed. “Inverter bad” can steer the quote prematurely; “system-wide production fell abruptly at 12:42 p.m. and event X appeared” preserves useful evidence.

If you want Teamsun to evaluate a service path, request solar diagnostics with this packet. Teamsun still must confirm service territory, ownership authorization, equipment and platform compatibility, licensed scope, safe site access, warranty path, and availability before accepting work.

How should a low-output repair quote be written?

A good quote separates diagnosis from a presumed repair. It defines the questions, measurements, deliverables, exclusions, and stop conditions before naming replacement parts.

Require these sections:

Quote section	Minimum detail
Intake assumptions	Address, ownership, equipment, symptom, available records, and data gaps
Diagnostic scope	Monitoring review, model/baseline, site/roof observations, event logs, electrical tests, meter/CT checks, and manufacturer support included
Safety and licensed roles	Contractor and individual licenses, roof-access method, shutdown/energization plan, and battery or structural coordination where applicable
Deliverable	Findings, measurements, photos, logs, fault tree, unresolved uncertainty, and recommended options
Diagnostic fee	Remote review, truck roll, labor hours, travel, instruments, manufacturer case work, and no-fault outcome
Repair authorization	No parts ordered or intrusive repair performed without written approval, unless specifically authorized
Warranty route	Claimant, evidence, RMA responsibility, covered part, labor contribution, shipping, and denial path
Parts and compatibility	Exact model, approved substitute, availability, lead time, firmware, commissioning, and system consequences
Permit and utility	Who determines whether repair, alteration, inverter change, storage work, or export-setting work triggers review
Acceptance test	Post-repair measured production, device reporting, event status, commissioning record, updated one-line/map, and owner access

Ask the contractor to state which hypothesis each test addresses. “Full system inspection” is not sufficiently precise if the complaint is underproduction. A useful scope might say: verify portal completeness; compare June output with the as-built model and coincident weather; inspect the repeated afternoon loss on the west string; record inverter events; measure string performance according to manufacturer procedure; verify production-meter and CT configuration; and issue a findings report with repair options.

Watch for these red flags:

• A replacement is quoted from a screenshot without checking data completeness.
• Cleaning is sold without quantifying or even observing soiling.
• The company compares AC output directly with DC nameplate and calls the difference a defect.
• A flat top is automatically labeled inverter failure without checking design clipping or export limits.
• The quote says “all brands supported” but does not identify the models, portal access, commissioning ability, or manufacturer claim path.
• A warranty replacement is described as free without allocating diagnosis, labor, shipping, access, permitting, and recommissioning.
• The technician proposes unlicensed electrical work or cannot identify the contractor of record.
• The scope omits a written findings report and acceptance test.
• The company promises a production percentage without defining weather, measurement, exclusions, or baseline.
• An upgrade is recommended before repair feasibility and compatible warranty options are documented.

There are legitimate cases in which an upgrade is better than a component-level repair: a discontinued proprietary platform, unavailable compatible parts, repeated failure, unsafe or noncompliant conditions, or a replacement that requires a redesign. Ask for at least two written paths—repair/restoration and upgrade/redesign—when both are technically feasible. Compare total scope, compatibility, expected performance, new warranties, permits, utility approval, monitoring continuity, and disruption.

Use the solar repair company selection guide for contractor compatibility, ownership, manufacturer access, and cross-brand questions. B156 chooses who can accept the system; this guide determines what the low-output diagnostic must prove.

When is maintenance, repair, monitoring work, or an upgrade the right path?

Choose the service path that matches the verified cause, not the symptom label. “Low output” may end with no repair, a monitoring correction, site maintenance, an electrical repair, a warranty claim, or a redesigned system.

Finding	Appropriate next path	Do not assume
Production normal; consumption increased	Consumption and bill review	Solar hardware replacement will lower the new load
App data missing; independent production normal	Monitoring/gateway/network/platform work	Array stopped generating
Output aligns with actual weather and model	No equipment repair; improve expectation/reporting	Proposal’s typical-year monthly figure is a guaranteed actual-month result
Quantified soiling or removable obstruction	Manufacturer/roof-safe maintenance plan	Rooftop DIY or universal cleaning schedule
New persistent shade	Shade/design analysis; vegetation or redesign options	Module electronics eliminate all shade loss
Confirmed string/module/inverter fault	Warranty or paid component repair	Error name alone proves the failed part
Export or grid constraint	Installer/manufacturer/utility coordination	Owner can change protection or interconnection settings
Discontinued/incompatible equipment	Repair feasibility and upgrade comparison	Any “equivalent” component can be substituted
Roof damage or leak	Coordinated roof and electrical assessment	A solar electrician alone owns every roof scope
Storm/impact evidence	Safety, insurance, roof/structural/electrical path	Routine maintenance article covers claim evidence

Scheduled maintenance belongs in B157. App-only failure belongs in B167. A specific manufacturer code belongs in B168. Storm inspection belongs in B169. An abandoned or orphaned-system service path belongs in B173. Keeping those intents separate avoids sending an owner through a generic checklist that does not answer the immediate decision.

Teamsun also publishes system upgrade services and a solar-plus-roofing service, but those pages do not prove an upgrade or roof project is needed for your system. Require diagnostic evidence first. Then confirm Teamsun’s current service territory, equipment compatibility, licenses, warranty route, schedule, and quoted scope directly.

Frequently asked questions about solar underproduction

Why is my solar system producing less than the installer estimate?

The estimate may use typical-year weather and design assumptions, while the actual period had different clouds, snow, shade, temperature, downtime, soiling, or operating controls. Confirm the final as-built system, model assumptions, complete production data, and actual weather before treating the difference as a fault.

How much should a solar system produce each day?

There is no reliable universal daily number per kilowatt. Production depends on location, date, weather, orientation, tilt, shade, temperature, system losses, inverter sizing, availability, and export controls. Use the original monthly model or a correctly configured PVWatts estimate for a range, then use weather-aware analysis for a disputed period.

Why does my system never reach its advertised kW rating?

Panel nameplate capacity is DC power measured at standard test conditions. Your monitoring may show AC power after temperature, wiring, conversion, mismatch, shade, soiling, and inverter limits. A lower AC peak alone does not show underproduction; compare modeled and actual kWh over time.

Does a high electricity bill prove low solar production?

No. The bill can rise because home consumption increased, rates or bill timing changed, the meter read was estimated, exports changed, or solar production fell. Compare the solar production export, utility interval imports/exports, billing dates, and consumption data.

Can solar monitoring be wrong?

Yes. A gateway can lose communication, intervals can be missing, and production/consumption sensors can be configured incorrectly. Check last-update times and independent measurement layers. Do not assume the array stopped solely because the cloud app is blank.

Is one underperforming panel worth repairing?

It depends on the verified energy loss, access cost, component and labor coverage, parts availability, effect on the rest of the architecture, safety, and remaining system life. First confirm that the module-level screen is mapped correctly and that shade or communication is not being mistaken for a hardware fault.

Does a flat production graph mean the inverter is too small?

It can indicate designed inverter clipping, but export limits, curtailment, data caps, or changed settings can look similar. Compare the DC-to-AC ratio, inverter rating, interconnection approval, configuration, event logs, and original annual clipping model.

Should I clean my solar panels if production is low?

Only after evidence suggests soiling is material and a safe, manufacturer-compatible method is available. Compare weather-corrected trends, rainfall response, and ground-level observations. Do not climb on the roof or use abrasive tools to test the theory.

Can snow damage production permanently?

Snow cover normally creates a temporary production loss, but sliding snow, ice, physical damage, water entry, or displaced equipment can create a separate issue. Do not climb or shovel panels. If low output persists after safe natural clearing, document the pattern and request a qualified assessment.

Should I reset the inverter when solar output is low?

Only follow the exact homeowner procedure in the current manual or a verified manufacturer support instruction for that model. Do not repeatedly cycle unknown equipment. A trip can represent an electrical fault, and resets can obscure useful event history.

Who pays for a solar underproduction repair?

Ownership and written contracts control. A lease/PPA owner may be responsible under its service terms. An owned system may have product, workmanship, labor, or service-plan coverage, each with different exclusions. A replacement part warranty does not automatically pay diagnosis, labor, shipping, roof access, permits, or lost production. Review the solar warranty comparison before authorizing work.

Can a new solar company service a system it did not install?

Sometimes. It must confirm ownership authorization, equipment and portal access, licensed scope, manufacturer support, parts, documentation, roof safety, and warranties. No contractor should promise universal cross-brand support without reviewing the exact system.

What should a solar diagnostic report include?

It should identify the measurement sources and data gaps, expected-production method, actual-weather adjustment, observed pattern, tests performed, measurements, event logs, photos, confirmed findings, unresolved hypotheses, warranty/parts path, repair and upgrade options, and a post-work acceptance test.

When should I call emergency services instead of a solar repair company?

Call 911 for fire, smoke, active arcing, or immediate danger. Keep away from shock hazards, exposed conductors, damaged batteries, severely displaced equipment, and water inside electrical equipment. Use the utility and manufacturer emergency instructions as appropriate; routine diagnostic booking comes after the site is safe.

Sources and methodology

This guide was researched and checked on August 10, 2026. It uses current federal safety and performance guidance, NREL modeling and research, manufacturer support documentation, New England licensing sources, consumer-warranty guidance, current search results, and forum questions. Forums were used only to identify buyer language such as “30% below estimate,” “app shows zero,” and “one panel is low”; they were not used as evidence for performance, safety, warranties, or repair scope.

The research method separates four evidence classes: measured production, modeled expectation, observed condition, and confirmed fault. It does not create a universal acceptable-loss threshold. The original contract, actual weather, exact equipment, current manufacturer instructions, state licensing, local permitting, and utility requirements control each real project.

Key primary sources include:

• DOE: operate and maintain an existing PV system
• DOE: optimize solar PV performance and longevity
• DOE: solar monitoring platforms
• NREL: PVWatts Version 8
• NREL: National Solar Radiation Database
• NREL: solar soiling research
• OSHA: solar electrical hazards
• FTC: written warranties and service contracts
• Enphase: homeowner troubleshooting and support
• SolarEdge: production and system issues
• Tesla: understanding system performance

Teamsun-specific service tickets, monitoring screenshots, supported-brand lists, response times, diagnostic fees, warranty outcomes, technician credentials, and measured repair results were not available as verified first-party evidence for this article. The guide therefore makes no claims about those items.

If the evidence shows unexplained low production and you want a company to evaluate the next step, request solar diagnostics from Teamsun. Include the system identity, ownership, production export, expected-production model, alerts, warranties, utility data, and symptom timeline. Teamsun must confirm whether the address, equipment, platform access, warranty route, licensed scope, and requested service fit its current capabilities.