Development and evaluation of a self-cleaning custom-built auto sampler controlled by a low-cost RaspberryPi microcomputer for online enzymatic activity measurements.

A fully automated on-site device (SAMP-FIL) that enables water sampling with simultaneous filtration and effective cleaning procedures of the device's components was developed and field-tested. The SAMP-FIL was custom-built using commercially available components and was controlled by a RaspberryPi single-board computer operating open-source software. SAMP-FIL was designed for sample pre-treatment with minimal sample alteration to meet the requirements of on-site measurement devices that cannot handle coarse suspended solids within the measurement procedure or cycle. A highly effective cleaning procedure provides a fresh and minimally altered sample for the connected measurement device. The construction and programmed software facilitates the use of SAMP-FIL for different connected measurement devices. The SAMP-FIL sample pretreatment was tested for over one year for rapid and on-site enzymatic activity (beta-d-glucuronidase, GLUC) determination (BACTcontrol) in sediment-laden stream water. The formerly used proprietary sampling set-up was assumed to lead to significant damping of the measurement signal due to its susceptibility to clogging, debris accumulation and bio-film accumulation. The implementation of SAMP-FIL considerably increased the error-free running time and measurement accuracy of BACTcontrol devices. This paper describes how low-cost microcomputers, such as the RaspberryPi, can be used by operators to substantially improve established measuring systems via effective sampling devices. Furthermore, the results of this study highlight the importance of adequate sample pretreatment for the quality of on-site measurements.