Available Instrumentation for Online Measurements

The most widely used instruments that allow us to measure some of the above variables will be discussed in this section. We will focus on how to measure in order to get the best representation of the bed conditions, on describing the advantages and limitations of the specific instruments, their reliability, expected precision, and relative cost. It is important to remember that any probe located inside the solid bed of a stirred reactor will require special care to avoid damage from the solids movement during the agitation periods.

Temperature. Any SSF control system should include several temperature measurements to monitor the bed temperature distribution and to measure inlet and outlet air temperature so that energy and water balances are kept under control. Several inexpensive temperature sensors are commercially available, thermocouples (TC's) being the most widely used within industrial environments. Their low cost, wide measuring range, and fast and linear response explains their popularity among process engineers; however, thermocouples show poor precision and accuracy. Even though the performance of thermocouples in SSF bioreactors can be significantly improved if they are calibrated frequently, other temperature sensors are superior. Resistance Temperature Detectors (RTD's) are better suited for SSF processing needs. These devices are stable, precise, respond fast, and do not need periodic calibration, even though they are more expensive and fragile than thermocouples. For a complete description of these and other temperature sensors, the reader is encouraged to consult specialized literature (Liptak 1995a; Creus 1998a).

Bed water content or water activity. None of the traditional methods commonly used to measure the water content of solid samples have gained wide acceptance in the SSF field since they are time consuming (needing between 2 and 15 hours)

(Creus 1998b). In addition, capacitance or conductivity-based devices are too sensitive to the electrodes/sample contact area or the apparent density of the sample (Gimson 1989), which normally varies during the fermentation. If properly calibrated and temperature-compensated, such devices may be useful to measure the water content of the solids phase in static packed-bed reactors, but they are limited to a maximum of 50% water content (Creus 1998b) which is too low for the majority of SSF processes. New methods like IR analyzers or IR scales are preferred, since they give a quick and precise indication of the humidity of a sample taken from the process (Fernández et al. 1996; Durand et al. 1997). However, these methods are not appropriate for automatic control of the humidity of the bed of solids, since they require the intervention of an operator to handle the samples. Optic absorbance sensors are a good option for automation in this case, since they provide water activity measurements in 2-3 min with a precision of 0.3% (Bram-orski et al. 1998; Bellon-Maurel et al. 2003). Other commercially available humidity measuring devices such as infrared or neutron radiation sensors are not only expensive, but they are also impractical to use in SSF reactors and provide information regarding the surface of the solids only (Brodgesell and Lipták 1995; Creus 1998b). More promising are those based on the emission of radio frequency fields or on Time Domain Reflectometry (TDR) (Bellon-Maurel et al. 2003), since they compute a representative value of the water content in the 3-D zone covered by the electrodes (SCI 1996; SE 1998; Hillen 1999; Atkinson 2000).

Gas flow rate. Several methods exist for measuring both volumetric and mass gas flow rate. Those based in pressure drop (Pitot tube and annubar, Venturi, and orifice flow meters), variable area (rotameters), speed (anemometers, turbines), force (badge meter) and vortex (Von Karman effect) are used to measure volumetric flow rate. These techniques can be adapted to measure mass flow rates also, in which case pressure and temperature compensation is required. In addition, thermal methods (based on the temperature difference between two resistance probes) and Coriolis force (vibrating tube) are specific for mass flow measurements (Creus 1998c). Although there are many options, it is rather difficult to select an adequate flow meter, since not all of them are applicable in a given system due to space, cost, pressure drop, and precision constraints (Lomas and Lipták 1995). Moreover, in some cases a flow rectification device would have to be installed before the sensing probe (Siev et al. 1995). Hence, it is advisable to define the kind of flow meters that cover the measurement range of interest first and then to identify those that better suit the intended application, considering the maximum allowable error, the pressure and temperature that the sensor will be exposed to, and the type of flow (laminar, turbulent, or transition). The lowest cost sensor should then be chosen from this short list. In assessing costs, in addition to the purchase price, the costs of maintenance, spare parts, and sensor operation must be considered, since at times an appropriate instrument can be cheap to buy, but in the long term the other costs can make it impractical. For further details see (Lipták 2002). Table 26.1 (adapted from Cole-Palmer 2003a,b) can be useful in choosing a suitable flow meter.

Table 26.1. Flowmeter characteristics

Attribute^

Accuracy

Repeatability

Was this article helpful?

0 0

Post a comment