As part of its ISO 17025 accreditation process, the Council for Geosciences in Pretoria, South Africa, had a requirement to measure and log a large number of environmental parameters in various laboratories throughout their multi-story building.
As with any data logging project, various monitoring methods are available depending on factors such as the number of measurement points, their location, how often they need to be sampled and of course cost.
The challenges in providing a suitable monitoring solution to the Council for Geosciences were numerous. First, a variety of measurement types needed to be made, including simple room temperature, differential pressure, water conductivity and even the surface temperature of an instrument head that costs about the same as a medium family sedan.
Another problem was the layout of the eight-level building. Measurements had to be made at various locations on various levels from the basement to the service area in the roof, and all the data had to be sent to a central location on the second floor with easy integration into the Laboratory Information Management systems (LIMS).
In addition, the selected system had to be able to provide alarm notifications to building staff when certain limits were exceeded. While some parameters such as room temperature simply had to be logged several times throughout the day, others such as the instrument head temperature had critical levels for which alarm notifications needed to be triggered.
Finally, due to the age and design of the building, there was a requirement to monitor for flooding. Expensive and delicate equipment had sustained damage in the past when the water supply system in the roof had failed. Thus, some form of flood detection and event monitoring was a key requirement.
The requirement that the data needed to be sent to a central point for easy integration into the Council’s LIMS system made the use of multiple, traditional data loggers impractical. Each data logger would have needed its data downloaded manually by one or several staff members on a regular basis, which would be labor intensive. An additional problem with that approach was that the data would not be available in real-time.
Another option would have been a centrally-located data acquisition system with hard wired probes spread around the building. However, due to the large amount of measurement points, and the complex layout of the building, this solution was rejected due the expense of both hardware and installation labor costs. It would also have been impractical to move the data collection points if required at a later stage.
The type of data logging system to use was thus narrowed down to a wireless data logging system. Wireless technology held the promise of making installation simpler than a hard-wired system, eliminating the need for offloading individual data loggers throughout the facility, and allowing for easy relocation of the sensors.
Onset?? HOBO?? ZW Series data nodes ??were chosen due to several unique features, the power of the included software, and the relative low cost of each measurement point.
These wireless data logging nodes are battery-operated devices that transmit the recorded data back to a central receiver at user-specified intervals. They overcome one of the main problems of traditional wireless logging systems, in that each node does not have to be in direct communication with the receiver.
Each node can be set up in dual purpose mode, by connecting to AC power, to act as a data logging node as well as a router to pass on data from other nodes. Through the combination of routers and data logging nodes, the system forms a self-healing network to ensure that the data reaches the receiver via an alternate path should the existing path fail or be obstructed.
In the event of failure of the receiver or dedicated computer, each data node can store a large quantity of data in the on-board buffer memory and transmit it once the receiver is again available.
Although the receiver and the router are usually mains-powered, they also have an internal battery backup so that data can continue to be collected even during a power failure. Thus, the potential for collected data being lost is minimized due to the redundancy and back-up built into the system.
As previously mentioned, the Council for Geosciences had a number of building parameters to measure. Many of these, such as room temperature and relative humidity, were simply measured using nodes with integrated probes dedicated for these measurements.
Higher temperatures were measured using regular thermocouples and 4-20 mA transmitters that were connected to nodes with analog inputs. Differential pressure sensors measuring across air-conditioning filters were similarly connected to analog nodes.
Other parameters, such as water flow, were measured using third-party flow sensors connected to nodes with pulse inputs. Chilled water temperatures, as well as the temperature of the expensive instrument head, were measured with thermistor probes connected directly to analog inputs.
Beyond environmental monitoring, one of the primary requirements of the system was to detect potential flooding caused by breaches in the aged water storage system in the roof of the building. This flooding had caused expensive damage on several occasions in the past, particularly when it occurred over weekends when the building was generally empty of personnel.
To minimize the damage, a gutter system was installed to collect and re-route the flood waters. The data nodes act as a flood detection system with the installation of float switches in the gutters.
Data and Network Management
To help manage the wireless data node network and the flow of incoming data, accompanying??HOBOnode?? Manager network management software is used. The software, which is part of Onset???s standard HOBOware?? Pro software package, displays a list of all data nodes and measurement points in the network, along with current status and measured value.
Along with this view are other frames that show graphs of selected measurement points, and a frame for configuring and displaying alarms.
The software can be configured to allow for data to be automatically sent at regular intervals to remote locations through FTP and email, or to a local network folder.
In terms of real-time graphing, graphs can be displayed individually or by grouping graphs by measurement type. The timebase of the graphs can be easily switched between the last 4 hours, 1 day, 1 week and 1 month using preset buttons, while another button allows a graph to be printed to the default printer.
This feature is used by Geoscience to manually monitor process times and values, as well as trends and stability of long term measurements.
HOBOnode Manager also features a ???network map??? feature which enabled building staff to import a schematic diagram of the eight levels of the building. The locations of each node are superimposed upon it. The last measured values from each node may also be shown. For troubleshooting purposes, lines showing the connections between each node can also be enabled.
Alarm notifications are typically sent via email, although visual and audio indicators on the PC display may also be used. Any triggered alarm can be programmed to send a message to any email address and an ???alarm cleared??? message can also be sent.
A number of system alarms are available to building staff, including ???node missing??? and ???node battery low.??? Each measurement value can also be programmed with an alarm level to send a message if it trips and/or if the alarm clears.
Building staff primarily receives alarms notifications via email for non-critical measurements. However, critical alarms such as temperatures on the XRF machine as well as the flood detection switches will trigger text messages to mobile phones in order to minimize equipment damage and financial loss.
Since the wireless data node network has been installed, the Council of Geosciences has able to fulfill its measurement logging requirement for ISO 17025, as well as be alerted in the event of equipment failure or flooding.
A total of 28 wireless data nodes have been deployed, which measure eight different parameters at a total of 56 measurement points in 20 rooms. Placement of the nodes was distributed over seven floors of the building, with all the data being returned in real time back to one central collection point.
The data has also been made available via the integral web server to any user on the network, and is regularly exported for entry into the local LIMS system.
ED. NOTE: This is a syndicated post, which originally appeared at onsetcomp.com/rss/feed. View original post.