Quick answer
A stated retained-output multiplier for specified PV losses, whose meaning depends on the starting quantity, output boundary and modeled conditions.
Answer: A PV derate factor is a retained-output multiplier for a defined set of losses. A factor of 0.95 means that 95% of the stated starting quantity remains after those losses. Specify the input, output boundary, time period and included mechanisms. One generic factor cannot independently predict site energy, inverter capacity or every module operating condition.
Define the factor before using it
“Panel derate,” “system losses” and “DC-to-AC derate” can refer to different boundaries. A factor might represent one module-temperature condition, selected DC-energy deductions or a simplified conversion to an AC-energy estimate. The label alone does not tell you which.
Write its meaning explicitly, for example: “0.96 retained fraction after these assumed sequential wiring and mismatch deductions, before inverter conversion.” Identify whether the calculation starts from power in W or kW, or energy in kWh. A nameplate rating is power under reference test conditions; it is not an annual energy quantity.
A 400 W reference power multiplied by 0.82 is 328 W arithmetically. It only predicts a defined operating point if irradiance, temperature and the factor’s boundaries support that interpretation. It does not mean the panel always produces 328 W or that an inverter should be rated at 82% of the array’s nameplate.
Calculate sequential retained fractions
For specified conditional losses expressed as fractions, the combined retained fraction is:
D = (1 − L1) × (1 − L2) × … × (1 − Ln)
The total loss fraction is 1 − D. Convert percentage inputs to fractions before calculating. This is the approach described for aggregated loss inputs in the historical PVWatts V5 manual. It does not authorize multiplying unrelated percentages from different report denominators or counting the same mechanism twice.
Hypothetical two-stage worksheet
Assume 10,000 kWh before two declared sequential deductions: 2% and 3%.
| Stage | Calculation | Energy remaining |
|---|---|---|
| Starting quantity | Declared illustrative input | 10,000 kWh |
| First deduction | 10,000 × 0.98 | 9,800 kWh |
| Second deduction | 9,800 × 0.97 | 9,506 kWh |
The combined factor is 0.9506 and the combined loss is 4.94%, or 494 kWh. Simply adding 2% and 3% gives a different result. The worksheet contains no assumption about an actual location, module technology, tariff or annual generation forecast.
Changing an assumed retained fraction from 0.80 to 0.85 changes this illustrative 10,000 kWh starting quantity from 8,000 to 8,500 kWh. The difference is 500 kWh, five percentage points of the starting quantity, or 6.25% relative to the first result. It does not prove that physically changing a system will recover that energy.
PVWatts: a loss input is not the entire DC-to-AC conversion
The historical V5 manual distinguishes its 14% default system-loss input from inverter efficiency, which is modeled separately. Its thermal calculation is also explicit. Therefore 1 − 0.14 = 0.86 is not a universal whole-system conversion factor. The earlier V1 default of 0.77 and the later V5 loss input describe different model conventions.
For a present integration, check the current PVWatts V8 API documentation. It lists system losses as a required percentage input without an API default, and inverter efficiency as a separate parameter. It also exposes DC/AC ratio, array type and monthly soiling inputs. Record the actual request, software version and result boundary instead of assuming an old default is automatically applied.
Do not add another inverter or temperature deduction to a result that already includes it. Check whether a separate monthly soiling input overlaps a soiling allowance embedded in the aggregate loss input.
Which effects need their own evidence?
| Effect | Evidence or model input to identify |
|---|---|
| Module temperature | Selected coefficient, weather, cell-temperature method and mounting assumptions |
| Soiling and snow | Site observations, seasonal treatment and maintenance assumptions |
| Shading | Obstruction geometry, irradiance treatment and electrical mismatch model |
| Wiring and connections | Actual circuit design and the model’s electrical-loss boundary |
| Inverter conversion and clipping | Selected model, operating curve, capacity and dispatch conditions |
| Degradation and availability | Time basis, exclusions and method; distinguish aging from outage time |
There is no defensible universal 0.77–0.86 target or climate-only prescription for all these effects. An arbitrarily conservative number is not evidence of lender acceptance or measurement accuracy.
For thermal assumptions, use the temperature-derating guide and temperature-coefficient definition. Wiring-loss assumptions do not establish compliance with an adopted electrical code or equipment instructions. The former universal “NEC limits DC voltage drop to 2%” claim is removed; project electrical requirements must be checked independently.
PTC is another reference condition, not a complete loss budget
California DGStats defines PVUSA Test Conditions using 1,000 W/m² irradiance, 20°C air temperature and 1 m/s wind. The previous 45°C ambient description was incorrect. Its guidance distinguishes the module PTC reference from further field losses.
Do not substitute a PTC rating for a site weather model or multiply it by thermal effects already implicit in the rating without examining the boundary. Likewise, a system’s CEC-AC reporting convention is not the same as a guarantee of energy at the delivery meter.
Review a factor in a proposal or simulation
Ask for the loss register, rather than only the final percentage:
- State the starting power or energy and its reference conditions.
- Name every included deduction and its supporting source or assumption.
- Identify mechanisms handled explicitly elsewhere in the simulation.
- State the output meter or power boundary and time period.
- Compare scenarios with other inputs held fixed and retain the calculation record.
When evaluating a generation and financial workflow, request this traceability for the actual demonstration. A generic claim of automatic loss modeling is not a substitute for inspecting an output. For a complete report, the PVsyst review checklist explains related reproducibility questions.
Questions about derate factors
What is a good derate factor?
One that is justified for the stated boundary and inputs. A default can support a declared preliminary scenario, but it does not validate the actual site or establish a universal pass threshold.
Is a factor of 0.86 the same as 86% inverter efficiency?
No. A retained fraction for an aggregate loss input and an inverter conversion efficiency describe different mechanisms. Check whether either is already modeled before applying another deduction.
Can I calculate annual energy from nameplate capacity alone?
No. Annual energy also requires a time-based solar-resource and system-performance model. A single power multiplier does not provide operating hours, irradiance history or the applicable annual boundary.
Source review: September 30, 2026. Numerical worksheets are illustrative, not field measurements or design approvals.
Related Glossary Terms
About the Contributors
CEO & Co-Founder · SurgePV
Keyur Rakholiya is identified by SurgePV as its CEO and a company co-founder. His SurgePV author page lists only role information that can be tied to the public profile below; credentials, project totals, testing claims, media appearances, and speaking engagements are not asserted without retained evidence.
Editorial contributor · SurgePV
Rainer Neumann is credited as an editorial contributor on SurgePV content. This profile does not assert engineering credentials, project totals, software-testing experience, education, speaking engagements, or media citations because independent verification evidence is not retained in the publication record.
