Saith Acoustic Monitoring Showcase

Overview

An M2M system was rapidly developed for real time location monitoring of maintenance equipment for a client working in the oil industry. This involved creating a system for conveying relatively large amounts of data over a mobile phone networks utilizing 3G and GPRS technologies.

Business Case

Our client works in a specialised industry which entails accurate location tracking of maintenance equipment in oil pipes which travels over a number of miles of pipeline. When a problem occurs during the maintenance process, the equipment requires to be rapidly located so that immediate action can be taken, thus resulting in minimal disruption to service.

Our client approached us with an idea for monitoring the location of the maintenance equipment. This idea entailed the deployment of time synchronised remote devices which communicate over GPRS, and the Internet. The remote devices would be positioned miles apart on a pipeline and send sampled data over the Internet in real time to a laptop where autocorrelation analysis could be performed in MATLAB. The timescale for developing and constructing a proof of concept system was just 6 weeks.

Solution

Raycomm designed and developed a solution within the short timescale imposed on the project which consisted of the following key parts:

• Hardware development for six remote monitoring units
• Development of a C# Windows application that receives and stores the data returned across the Internet from each of the remote monitoring units
• A robust communication protocol designed by Raycomm which ensured the data collected by the monitoring units could be transferred reliably over GPRS and the Internet
• Development of a C# Windows application that configures the remote monitoring units over a USB connection

Remote Monitoring Unit

Cost effective, off the shelf parts were adapted as necessary to fulfil the requirements of the remote monitoring unit prototypes. A GPRS modem, extended memory, USB and SD card interfaces were added to an Analogue Devices ADUC842 8052 microprocessor development board. The ADUC842 is particularly well suited to this application because it has vast external memory addressing capability used to store the sampled datasets.

Analogue circuitry was developed to detect and amplify a weak acoustic signal from a piezo sensor. An instrumentation amplifier was used to amplify the weak signal from the piezo sensor. The gain of the instrumentation amplifier could be changed with a digital potentiometer that was controlled by the microprocessor. The microprocessor adjusts the digital potentiometer as a consequence of receiving gain control messages over the GPRS Internet connection, thus providing remote control capability.

Careful consideration was given to the active filter design in order to remove noise. A bandpass filter of 4th order Butterworth response was designed, with cut-off frequencies at 106Hz and 884Hz. It was particularly important to attenuate 50Hz mains frequency which could be picked up from overhead power lines. This acoustic signal was subsequently amplified, driven through a cable to one of the ADC’s on the ADUC842 microprocessor, sampled, stored on the SD memory card and transmitted over the GPRS connection.

Firmware development for the remote monitoring units and Windows software applications progressed in parallel with hardware development. The firmware and hardware for the remote monitoring units contained the following key features:

• Embedded C firmware for the Analogue Devices ADUC842QS microprocessor.
• A driver to control a GPRS modem connected by serial port.
• A driver for read/write access to the 128MB SD memory card.
• Firmware for management of sampled data storage in an address mapped 1MB SRAM memory chip.
• A custom protocol for reliably transferring datasets over the GPRS connection and Internet.
• Time synchronisation with other remote monitoring units.
• The ability to accept new configuration settings from the application running on the centrally controlling laptop.
• A USB interface on the remote monitoring units which employed the FTDI FT2232L chip.

The client’s requirements stated that the remote monitoring unit sampling needed to be time synchronised in order for the MATLAB post processing algorithms to work properly. The remote monitoring unit were required to start sampling simultaneously, when located miles apart. Time synchronisation failure would result in inaccurate location of the maintenance equipment within the pipeline.

One of the main challenges encountered during the project was to devise a way to time synchronise the commencement of each monitoring units sampling. Due to network latencies it was out of the question to broadcast a “start sampling” message from the centrally controlling laptop to each remote monitoring unit. The design decision was made to time synchronise each of the remote monitoring unit processor clocks before the units were deployed along the pipeline.

The time synchronisation mechanism was realised with the use of CAT5 patch leads to daisy chain each of the remote monitoring units together. A switchbox was fabricated which provided the signal simultaneously to each remote monitoring unit which instructed time synchronisation. The CAT5 patch leads would then be removed so the remote monitoring units could be deployed in turn along the pipeline. A remote monitoring unit standby timer would be configured, typically for a number of hours to cater for deployment activities.

Data Collection Application

A Windows application was required to receive and store data from each of the remote monitoring units. The application was written in C# and used Windows Forms for the user interface. It encompassed the following key features:

• The ability to receive and maintain simultaneous incoming streams of data from all of the remote monitoring units.
• A custom communications protocol which ensured data received from the remote monitoring units could be reconstructed accurately, despite intermittent GPRS and 3G Internet connectivity used throughout the system.
• Visual indication as to the state of the M2M system for example configuration parameters and indication of which remote monitoring units are currently connected and returning data.
• The ability individually request and change the parameters of each remote monitoring unit.

Communications Protocol

A custom communications protocol was required to transfer sampled datasets from the remote monitoring units to the centrally controlling laptop. The protocol had to cater for transferring data over intermittent GPRS and 3G internet connections. Steps were taken to minimise the amount of data transferred over the mobile phone networks in order to reduce operating costs.

The communications protocol also catered for the various control messages sent between the maintenance control site laptop and the remote monitoring units.

A mechanism was also devised to overcome the dynamic IP addressing assigned by mobile network operators to devices which connect to the Internet via their network.

Technologies Involved

• Internet communications
• GPRS / 3G mobile phone network communications (Orange)
• .NET Framework 2.0
• C#
• Keil uVision IDE
• C
• Real time OS (Keil RTX Tiny)
• USB interface to 8052 processor using the FTDI FT2232L chip
• SD card interface to 8052 processor
• GPRS modem interface to 8052 processor
• Instrumentation amplifier
• Active filter design with OpAmps
• Analogue Devices ADUC842QS 8052 based microprocessor.

Raycomm are a UK based, custom software and electronics design and development service company creating bespoke electronic products, mobile device software applications, backend web server software and e-commerce database software.

For a no-obligation discussion with one of our senior consultants on your project needs:

• Call us on +44 (0) 1275 847553
• or fill in our Enquiry Form
• or email us using the link at the top of the page

We look forward to hearing from you.