Hukseflux Thermal Sensors Outdoor Pyranometer Calibration User Manual
- June 1, 2024
- Hukseflux
Table of Contents
Hukseflux Thermal Sensors Outdoor Pyranometer Calibration
Outdoor pyranometer calibration by comparison to a reference pyranometer: not recommended “high-risk”, requiring high operator competence levels, variable uncertainty The ISO 9847 standard describes the calibration of pyranometers relative to a reference pyranometer. The method can be performed on-site. However, the outdoor calibration is “high risk” in terms of ISO 9001 risk- based thinking. The resulting calibration accuracy is not a constant and will only, under exceptionally stable conditions, be sufficient to meet requirements for utility-scale PV system performance monitoring.
Introduction
The outdoor calibration method of ISO 9847, type B, and its equivalent in ASTM E824, transfers the sensitivity of the calibration reference sensor to a preferably lower-class test sensor under the outdoor natural sun. Outdoor calibration of pyranometers has the advantage that it can be done on-site. In addition, an advantage to meteorologists is local calibration at realistic solar angles and temperatures including the unknown directional error and temperature response in the calibration.
However, although the procedure may appear simple, working under variable atmospheric stability and unknown solar position is inherently
- in terms of quality management – “high-risk”.
- It is therefore seldom used, and discouraged by Hukseflux.
- the preferred method is to perform indoor calibration according to ISO 9847 type A
- an alternative, but more time-consuming and expensive method is outdoor calibration against a pyrheliometer and shaded pyranometer according to ISO 9846
ISO 9001:2015 – Risk-based thinking
In outdoor calibration, instrument installation, data analysis, data
rejection, and uncertainty evaluation are complicated procedures. Because of
variable environmental conditions during outdoor calibration, the process
cannot be automated and is “high risk” in terms of the risk-based thinking
required by ISO 9001:2015 – quality management systems – requirements
paragraph 0.3.3. Installation is critical, and the competence of personnel is
critical and must be reviewed according to ISO 9001 paragraph 8.4 – control of
externally provided processes, products, and services.
Calibration reference
The reference pyranometer is preferably of a higher accuracy class and preferably calibrated against a pyrheliometer (ISO 9847 paragraph 7.2) according to ISO 9846.
Data analysis by experts
ISO 9847 paragraph 7.4.5 requires data analysis for each measurement series,
in the following steps
- calculation of input ratios
- outlier rejection in case of more than 2 % deviation
- statistical analysis
- calculation of sensitivity
- individual uncertainty evaluation, including the standard deviation of the measurement
Uncertainty evaluation by experts
ISO 9847 paragraph 7.4.7 requires that the uncertainty of the calibration has
to be determined on a case-by-case basis. The result depends on measurement
conditions and measurement uncertainty of both the reference and the field
instrument. We recommend using the standard ASTM G217 standard guide for
evaluating uncertainty in calibration and field measurements of broadband
irradiance with pyranometers and pyrheliometers.
Time-consuming, typically 2 days with sun ISO 9847 7.4.2 requires outdoor horizontal calibration type B1 as well as for type B2, tilted and B3, normal incidence:
- minimum 2-day period
- solar zenith angles < 70 °
- minimum of 15 data series each 10 to 20 minutes long with 20 or more records in each series
- at least 240 records passing filters for data rejection
- 30 % of the records were taken within ± 2 hr around solar noon
Not all-season
In a lot of locations, only a limited number of months in the year are
suitable for calibration with a reasonable level of accuracy, either due to
low solar zenith angles (for example in winter) or low atmospheric stability.
Although the standard leaves the possibility open, we consider days with a lot
of clouds not suitable.
Low calibration accuracy
Typical contributions to the uncertainty budget are:
- the uncertainty of the calibration reference
- the uncertainty of the method
- instrument-related uncertainties, depending on the instrument class
Most users forget to take into account that the outdoor measurement
uncertainty of a calibration reference instrument is higher than its
calibration uncertainty. The best Class A calibration reference pyranometers,
calibrated against a primary standard pyrheliometer have a 0.8 % calibration
uncertainty. When using such the instrument, as a calibration reference under
conditions potentially different from the reference conditions of their
calibration, at least the uncertainty due to its directional response must
be added. This is in the range of 1 %. ASTM E824 paragraph 11.1.3 mentions 2 %
as an expected within-laboratory precision, using the same calibration
reference. We translate this to repeatability or “uncertainty of the method”
of ± 1 %.
Taking the RMS of the above, an outdoor calibration uncertainty of 1.6 % (k=2) seems realistically attainable. However, the resulting calibration is valid only for the solar azimuth and zenith angles during the calibration. The calibration reference condition for pyranometers typically (in particular for PV monitoring in Plane of Array) is normal incidence (at a zenith angle of 0 °) solar radiation. For the uncertainty of the transfer from the solar angles during calibration to normal incidence, we take 1 % for class A pyranometers and 2 % for class B pyranometers. The overall calibration uncertainty under the above, very favorable, conditions are in the order of 2 % for class A instruments and 3 % for class B instruments. By contrast, indoor calibration may reach uncertainties of 1.2 % and 1.5 % respectively.
- Table 1 Requirements for outdoor calibration according to ISO 9847, recommended analysis.
- Table 2 Hukseflux’s estimate of the best attainable calibration- and measurement uncertainty following ISO 9847.
- Table 3 Outdoor calibration of pyranometers according to ISO 9847 is not the right solution for utility-scale PV system performance monitoring.
The measurement uncertainty of pyranometers is higher than the calibration
uncertainty
Measurement uncertainties with pyranometers are a function of:
- calibration uncertainty
- instrument class
- environmental conditions including maintenance
- re-calibration time interval (non-sta ability)
- Taking a 1 % margin for instrument fouling, and 0.5 % margin for instrument non-stability, for Class A pyranometers the measurement uncertainty under optimal conditions is of the order of 2 % higher than the calibration uncertainty.
- If this instrument has been calibrated outdoors, this means a measurement uncertainty on the order of 4 %.
For Class B pyranometers, the measurement uncertainty is of the order of calibration uncertainty plus 3 %. If this instrument has been calibrated outdoors, this means a measurement uncertainty on the order of 6 %. These estimates apply to measurements at relatively low angles of incidence. The difference between hourly and daily totals is low.
Suitability for PV monitoring
As Table 3 shows, the uncertainty in outdoor calibration is larger than
required for a class A system of the IEC 61724-1 standard. It may, at best,
comply with class B. By contrast, calibration indoors or outdoors against a
pyrheliometer is more accurate and typically suitable for a class-A system.
Standards
ISO 9847: Solar Energy – Calibration of field pyranometers by comparison to
a reference pyranometer ASTM E824 – 05 Standard test method for transfer of
calibration from reference to field radiometersASTM G217 Standard guide for
evaluating uncertainty in calibration and field measurements of broadband
irradiance with pyranometers and pyrheliometers.
Calibration reference conditions
Reference conditions are not standardized, but the main manufacturers use:
- irradiance level 1000 W/m2
- normal incidence irradiance
- instrument temperature 20 °C
- horizontal instrument position
- spectrum: solar irradiance on a clear day
- When to employ outdoor calibration
- Hukseflux recommends indoor calibration. Outdoor calibration according to ISO 9847 is recommended only: as a check of instrument functionality using guarded rejection.
- For example, sending the instrument away for an accurate calibration if the outdoor calibration shows a deviation of > 5 % under near perfect conditions, at an ISO 17025 accredited laboratory, at normal incidence
- Why indoor calibration is preferred Modern instruments are produced within known and narrow performance limits. For these sensors, indoor calibration is best.
Advantages are
calibration at normal incidence, which is the reference condition for
directional response calibration at 20 °C which is the reference condition for
instrument temperature change of sensitivity is directly traceable to
sensor/coating degradation reference conditions comply with IEC Standard Test
Conditions for Solar Energy testing (STC), as applied in Photovoltaic (PV)
module and system testing
- fast, independent of weather
- independent of the day of the year and local latitude
- known and fixed temperature
- known and fixed uncertainty
- accuracy for class A pyranometer calibration is sufficient for utility-scale PV system performance monitoring, class A of IEC 61724-1
About this review
This review intends to provide objective information about preferred
calibration methods. We appreciate suggestions for improvement in this review.
About Hukseflux
Hukseflux is the leading expert in the measurement of energy transfer. We
design and manufacture sensors and measuring systems that support the energy
transition. We are market leaders in solar radiation and heat flux
measurement. Customers are served through our headquarters in the Netherlands,
and locally owned representative sales offices in the USA, Brazil, India,
China, Southeast Asia, and Japan.
Interested in our products and services?
E-mail us at: [email protected]
Copyright by Hukseflux. Version 2403. We reserve the right to change
specifications without prior notice Page 4/4.
For Hukseflux Thermal Sensors go to
www.hukseflux.com or e-mail us: at
[email protected]