Quality Magazine

Monitoring Temperature and Humidity

February 4, 2008

Monitoring temperature and/or humidity conditions is an essential ingredient of a wide range of quality assurance applications. There are many common methodological errors, however, in ways that this task is approached that either compromise quality standards or add unnecessary time and expense to the monitoring task. Insufficient calibration of temperature and humidity monitoring instruments is high on the list of problematic areas. Mismatching technology to the monitoring task at hand is another problematic area. This article will revisit technology trends in monitoring instrumentation, provide tips on calibration and discuss common methodological errors that quality managers should avoid.

"Before" Data and Calibration

“Before” data refers to taking an extra step in calibration to determine whether an instrument is in or out of range and by exactly how much. This is an extra step and usually costs more than a simple recalibration, which is why some quality managers forego this calculation. However, such “before” data is the only thing that will help determine and demonstrate that an instrument was operating in acceptable ranges during the entirety of a manufacturing operation; it also can be used to expose possible temperature or humidity fluctuations that jeopardize product quality. In many applications, gathering “before” data is highly recommended.

Another aspect of recalibration to consider is whether an instrument is returned exactly to original specifications, which manufacturers can do but third party recalibration services cannot.

Best Fit Technology

Which is the better option-the $40 data logger in a bulk purchase or the $400 data logger with a graph-at-a-glance paperless chart recorder? Actually, both could be the best fit for a particular application and it behooves quality managers to first of all ensure that they are accessing instrument sources that provide a wide range of options. In recent years there has been a mushrooming of instrument features and models and there is absolutely no reason to make do with an instrument that is not exactly matched to a specific application.

A systematic approach can determine which instrument specifications and features will best do the monitoring job at hand. Variables to consider include the monitoring range that the instrument is tuned to, how data is displayed and handled, alert or alarm systems, the required durability of the instrument, whether remote probes are required, and cost.

The first question is whether a chart recorder or a data logger is preferred. Objectively speaking, either type of instrument can be best-match technology for an application and the choice between the two is strictly a matter of personal preferences. Some quality managers prefer the visual graphic that a chart recorder provides without any interaction or need to download data, as is required with a data logger. Others prefer the ease of saving and archiving electronic data from a data logger and may think that paper charts from chart recorders are too susceptible to loss or damage. These are strictly preferences and there is no one right way to keep track of data. In fact, the lines are now somewhat blurred between these two types of instruments. Many chart recorders now have built-in digital displays. And, there are now so-called paperless chart recorders that digitally provide the same graph-at-a-glance as paper chart recorders.

Alert features not only help ensure that out-of-range conditions are averted but also can significantly reduce the time spent by quality department personnel on downloading and handling data. The simplest alert is actually just large readable displays that make possible a quick check while walking past the logger or recorders in situ. There are now easier-to-read chart recorders that have 8-inch readable chart areas able to display data details, which also are often augmented with a digital display. Similarly, one can find data loggers with large digital displays. Usually digital displays have the option of showing either current conditions or min/max readings. Quick checks of min/max readings during routine inspections take negligible time and are helpful in determining if problems have occurred. (Note: Min/max readings will not tell how long conditions are out of range, and for that reason are usually ill-matched for critical applications where such data is needed.) Instruments with alerts often are the most efficient timesaving models in that they make it easier to determine when data does or does not need to be downloaded for analysis.

In many companies that have facility-wide alarm systems, the best match instruments are those with relays that enable feed of data into other automated systems such as HVAC controls. Some models can take this a step further by enabling e-mail alerts to a desktop or PDA of out-of-range conditions.

Streamlining Monitoring Tasks

Many quality departments could significantly cut the time they spend on monitoring tasks by both better consideration of real-world monitoring requirements and accessing technology that facilitates easier data handling.

There are two ways to get real-world monitoring requirements wrong-by underestimating the need for data or by overestimating the need for data. Because hot air rises, it is important to monitor vertical spaces at various heights. Problematic transition points need to be identified. These would include doorways, areas near HVAC outputs and exits to unconditioned spaces such as loading docks and staging areas. Usually simultaneously tracking outside temperatures is recommended. Likely local hot spots should be sought out near inventory racks, shelving and pallet storage areas due to constricted air circulation. Seasonal variations also need to be considered and in many climates there is a need to repeat mapping exercises with each season.

Often unrecognized by quality managers, excessive sampling also is a problem; it is both unwieldy and an unnecessary drain on productivity, tying up both computers and managers in processing too much data. What is too much data? Consider that in a large warehouse space temperature changes will likely happen very slowly over the course of several minutes, such that sampling every 10 to 15 minutes should adequately evaluate temperature trends.

Figure 1 shows the relative data processing loads of three different warehouses that are A) sampling every 15 minutes, B) sampling at one minute intervals and C) sampling every 2.5 minutes. The B warehouse that is sampling data at one-minute intervals is generating 15 times the data that warehouse A is, and warehouse C is generating 6 times as much data. Unless explosives are involved, or the very rare products that deteriorate nearly instantaneously with temperature variation, warehouses B and C are sampling at frequencies that exceed the likely rate of temperature change in their facilities.

Technology selection also has great bearing on the efficiency of monitoring, and instrument features that streamline data handling in data loggers are usually well worth considering. First, one should consider that the previously mentioned audio/visual alarms and min/max displays are useful in that they help determine when data downloading is actually necessary. If temperatures and humidity have remained in acceptable ranges, it probably will not be necessary to take the time to download data.

Second, it is often especially useful to look for data loggers that are capable of storing data on the same type flash memory cards used in digital cameras. This will enable staff to collect the data without moving data loggers around, a considerable time-saver.

Third, look for data loggers that use faster USB connections for downloading. These can be three times faster, and though we are only talking about minutes for each logger, if there are many loggers to monitor this can add up.

Fourth, consider sourcing data loggers that can plug into company-wide computer networks. These will enable one person to sit at their desktop and monitor all the data loggers throughout the company. They also provide the ability to check on the status of temperature and humidity conditions remotely via Internet connections and remote control software.

No One-Size-Fits-All Solution

A wide array of instruments and features are now available that will best fit the needs for quality control in tight and loose tolerance temperature and humidity conditions. It is quite timely to re-examine what is available in monitoring instrumentation and how to streamline monitoring tasks. This exercise can greatly impact the efficiency of quality departments and how they approach temperature and humidity monitoring.

Tech Tips

  • Because of a mushrooming of instrument features and models in recent years, it is usually possible to find an instrument matched to a specific application.

  • There are two ways to get real-world monitoring requirements wrong-by underestimating the need for data or by overestimating the need for data.

  • In many companies that have facility-wide alarm systems, the best match instruments are those with relays that enable feed of data into other automated systems such as HVAC controls.