Anville data logger measurement uncertainty
Anville systems differ from the majority of data loggers by the technology used to switch the analogue inputs and in the design of the thermocouple cold junction compensation in the 425/TC system. The Anville Series 425/TC system minimises the cold junction compensation error by using correctly compensated miniature type T thermocouple connectors with matching class 1 type T thermocouple wiring to the cold junction temperature sensor connection.
The temperature of the thermocouple cold junction is measured with a 1/3 DIN Pt100 sensor which has an accuracy of ±0.1°C at 0°C. Design of the physical layout of the cold junction compensation reduces any thermal scatter to a typical deviation of ±0.05°C.
The majority of systems use solid-state semiconductor multiplexers. Anville systems use Coto reed relays, which have many advantages including 250VDC isolation between inputs, 400VDC isolation to ground, minimal thermo-electric EMF generation (which is ideal for thermocouple sensor switching), no resistive offsets, long life and total reliability.
A 16 bit high accuracy, dual slope, integrating A/D converter with a programmable gain instrumentation amplifier minimises noise, time and temperature drift errors. Digital linearization for thermocouple at Pt100 sensors is employed for signal conversion to °C.
For type T thermocouples, using the -100°C to 200°C temperature range, the resolution of Anville system is 0.33µV, which is equivalent to 0.01°C for type T. The system measurement accuracy is ±0.25°C over the range from -100°C to 200°C. This accuracy figure includes all the sources of instrument error including the digital linearization of the type T thermocouple voltage output into °C, cold junction compensation measurement, A/D converter errors and instrumentation amplifier errors. The system accuracy figure of ±0.25°C also includes the errors due to instrument ambient temperature changes over an operating range from 0°C to 40°C.
Using an Anville system will allow you to comply with the requirements of HTM2010 that the repeatability of measurement should be ±0.25°C or better and that the limit of error of the complete measurement system (including sensors) should be no more that ±0.5°C.
The table below shows the calculation of measurement uncertainty of the Anville Series 415/425/500 data loggers and that this measurement uncertainty of ±0.11°C cannot be removed by calibration and qualifies all measurements taken with the system.
| Source of uncertainty | From data sheet | Probability distribution | Divisor | Calculated uncertainty |
|---|---|---|---|---|
| Resolution | 0.01°C | Rectangular | 1.732051 | 0.0000 |
| Drift of unit and thermocouples | 0.10°C | Rectangular | 1.732051 | 0.0034 |
| Range accuracy | 0.25°C | Rectangular | 1.732051 | 0.0210 |
| Stability vs. ambient temperature (0°C - 40°C) | 0.10°C | Rectangular | 1.732051 | 0.0034 |
| Drift with use | 0.10°C | Rectangular | 1.732051 | 0.0034 |
| Cold junction error | 0.20°C | Rectangular | 1.732051 | 0.0135 |
| Repeatability | 0.10°C | Normal | 1 | 0.0100 |
| Standard uncertainty (60% confidence level) | 0.05461 |
| Expanded uncertainty (95% confidence level) | 0.11°C |
The contents of this site are © Anville Instruments Ltd 2007 and may not be reproduced without written permission. "EaziVal" is a trademark of Anville Instruments Ltd. All other brand and product names are trademarks or registered trademarks of their respective owners.
