Issue link: http://read.dmtmag.com/i/550313
58 tobaccoasia the digital interface and analog outputs," accord- ing to Greg Brown, marketing executive at Mary- land-based firm Process Sensors Corporation, an- other major global supplier that has been active for many years. While the suitability and accuracy of NIR wave length sensors and other detection/analyz- ing methods have been long documented, the art is in robust housings and the flexibility to meet the customer's expanding needs for data collec- tion and remote diagnostics via digital interface. In other words: Industry customers expect that modern moisture analyzers can be deployed at virtually any location, in any environment, and be used by any personnel without compromising their functionality. Sophisticated technology for a diverse industry The specifications and requirements of tobacco manufacturers generally are extremely diverse, and moisture meters of any brand must be able to han- dle these as comprehensibly as possible. Moisture analyzers may be deployed at-line (online), tak- ing real-time measurements through high-volume sampling, or may be used offline in a lab setting for more thorough analysis, quality control and es- tablishing or monitoring trends. Two principal technologies are firmly estab- lished today and are regarded as most reliable in terms of measurement accuracy, although admit- tedly none of them can universally address the whole spectrum of potential measuring require- ments. Individual tobacco companies must there- fore choose carefully and weigh all pros and cons, the advantages and disadvantages, and determine in which environments they primarily want to use the machines before deciding, which model or brand best serves their particular needs. Dry if off, get great results! The more straightforward technology is, per- haps, the aforementioned loss-on-drying method, whereby samples are heated in an oven compart- ment (either by coils or, nowadays more common- ly, by microwaves) at a specific temperature and for a specific time period. In simplest terms, the weight loss after heating is calculated into mois- ture content. The advantage of the methodology obviously is that both surface and core moisture can be captured. Device manufacturers like Bra- bender claim a measurement accuracy of as little as +/- 0.1%, depending on the model. The downside is that under normal circumstances there is practi- cally no possibility to distinguish between the pro- portions of actual surface moisture and core mois- ture. Also, other volatile substances contained in the sample are evaporated, adding to the weight loss and thus potentially falsifying the moisture Process Sensor's MCT466-T Quik- Check Process Sensor's MCT460-T in action at conveyor belt sugars, and even fluctuate wildly in color, ranging from light lemon to dark brown. "These variations between types are huge, chemically speaking, and will impact a poorly designed algorithm, necessi- tating frequent re-calibration and apparent instru- ment drift. The simple Beer Lambert absorption laws used by the majority of moisture gauges in the market just do not get to grips with this com- plexity." The sign of a good chemo-metric model, he said, is independence to tobacco type and grade with the same calibration slope over the typical wide moisture range in tobacco processing. "The TM710e uniquely offers exactly this capability, but we are also smart enough to realize the learning is not over and that we must continue to evolve our tobacco model further." Design is equally important Yet potential hurdles do not only present them- selves in the practical application areas, but com- monly already start at the design and development stage. "[Some of] the main challenges in designing a moisture meter are the incorporation of robust electronics…, ease of use for both high-level en- gineers and line operators, and design flexibility of