Premium
A Simple Conductimetric CO 2 Analyzer with Automatic Recalibration: II. Factors Affecting Calibration
Author(s) -
Wall Gerard W.,
Acock Basil,
Milliken G. A.
Publication year - 1995
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1995.00021962008700010013x
Subject(s) - calibration , analytical chemistry (journal) , volumetric flow rate , chemistry , calibration curve , carbon dioxide , conductivity , dissolution , electrical resistivity and conductivity , water flow , environmental science , soil science , mathematics , thermodynamics , chromatography , detection limit , statistics , physics , electrical engineering , organic chemistry , engineering
Conductimetric CO 2 analyzers provide an economical means of monitoring and controlling carbon dioxide concentration ([CO 2 ]) in enclosed plant growth cabinets. They rely on dissolving CO 2 from an air sample in demineralized water and measuring the resulting change in electrical conductivity. Regressing the analog output from the conductivity meter against [CO 2 ] values over the range 200 to 1000 μL L −1 yields a quadratic response. Calibration curves for conductimetric CO 2 analyzers are affected primarily by temperature and air and water flow rates. Air and water flow rates are relatively fixed, whereas temperature varies over time in normal operation. A covariate analysis for a three‐way treatment structure in a randomized complete block design determined how the intercept ( B 0 ) and the linear and quadratic parameter estimates ( B 1 and B 2 ) varied with temperatures of 10, 20, 30, and 40°C, air flow rates of 540, 1880, and 3700 mm 3 s −1 , and water flow rates of 1.0, 0.66, and 0.33 mL s −1 . Temperature effects were greatest, with air flow rate causing minor changes, while water flow rate had no effect. Temperature and air flow rate significantly altered B 0 , with only minor changes in B 1 and B 2 . Variations in B 0 were sufficient enough to alter the [CO 2 ] around the desired set point by 70 μL L −1 for all treatments evaluated. Deviations in [CO 2 ] control due to variation in B 1 and B 2 were within 10 to 15 μL L −1 ; only the adjustment in B 0 must be made every sampling period (every 900 s) to maintain [CO 2 ] control within 10 to 15 μL L −1 of the set point.