Premium
Error Analysis of Heat Pulse Method for Measuring Soil Heat Capacity, Diffusivity, and Conductivity
Author(s) -
Kluitenberg G. J.,
Bristow K. L.,
Das B. S.
Publication year - 1995
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1995.03615995005900030013x
Subject(s) - thermal diffusivity , thermal conductivity , thermal conduction , line source , volumetric heat capacity , observational error , laser flash analysis , heat capacity , work (physics) , thermal , materials science , heat equation , mechanics , thermodynamics , mathematics , geometry , heat transfer , physics , heat flux , optics , mathematical analysis , statistics
A dual‐probe heat pulse (DPHP) method was developed recently that allows for the simultaneous, automated measurement of soil thermal diffusivity (κ), volumetric heat capacity (ρ c ), and thermal conductivity (λ). Estimation of thermal properties is based on theory for the conduction of heat away from an infinite line source (ILS) that is heated for a short period of time. In this study, we examined possible sources of error in the use of the ILS theory by comparing it with other models that explicitly account for finite length and cylindrical shape of the actual heater. For probe geometry and heating times typical of our experimental work, the analysis of model error showed that assuming an infinite length for a heat source of finite length caused errors <2% in the estimated thermal properties. Assuming the cylindrically shaped heater to be a line heat source caused errors of <0.6% in the estimated thermal properties. Thus, the ILS theory appears to be appropriate for use in the DPHP method if probe geometry is considered carefully. However, small changes in probe geometry can lead to large model errors. First‐order error analysis also was used to predict how thermal property estimates will be affected by experimental errors in the measured inputs to the ILS model. The analysis shows that κ and ρ c estimates are sensitive to measurement error in probe spacing ( r ), but λ is unaffected by error in r . Estimates of κ and λ were shown to be sensitive to measurement error in the time to the temperature maximum ( t m ), whereas ρ c was affected only slightly by such error.