Wireless sensors for hydrocarbon gas detection

Readers of the INSIDER Newsletter for November will know that in a report from the Invensys OpsManage11 user conference in Paris, the development of a wireless flammable gas detector by GasSecure of Oslo was mentioned. The background to this development was fascinating, but maybe more to instrumentation engineers than automation and control professionals, so it is described separately here.

Niels Aakvaag is a senior systems architect at GasSecure, www.gassecure.com, from Oslo, and their gas detector, the GS01, is a development that was undertaken with input from Statoil and ConocoPhillips, to create sensors that could be distributed around hazardous plant locations, on or off-shore, and work in a wireless mesh network. The specification, amongst other factors, called for the system to be designed to SIL2 standard and to operate from a recognised wireless standard, with the intrinsically safe sensor to have a 5 second response time to gas, with the sensors being battery powered for 2 years.

Possibly the project was started to take advantage of a recently developed MEMS-based IR/optical sensor, which gave the opportunity of approaching the cost of a conventional wired detector installation, where cable installation would typically be twice the actual sensor cost. The wireless sensor would obviously be far more convenient. The MEMS sensor uses a single optical beam, but analyses two wavelengths, one sensitive to hydrocarbons. This gives a path zero attenuation calibration at the same time as the active wavelength measurement, so giving good zero stability.

It was realized from the start that it would not be possible to run even the MEMS sensor continuously within the available 5mW power budget, so the sensor was designed as a two stage unit, with a continuous ultrasonic speed of sound monitor to detect coarse changes in gas composition, as a “pre-warning”. On detecting some sort of gas change, the optical sensor is switched on, and monitors just for hydrocarbons. Only then is a gas alarm signalled, typically for gas levels above 5000ppm. The expected use of the optical sensor is around 100 times per day.

The sensors are interrogated every 20 seconds, to conserve power, but can still achieve an overall 5 second response system time by delaying their “Safe” condition response for approx 17 seconds of the 20, before a response transmission. This is of course, unless a high gas level is detected within this waiting period, when the alarm response is triggered relatively immediately.

The wireless interface was designed by Nivis, and can be configured to use either WirelessHART or ISA100.11a. The initial units were set up to use ISA100, with a digital output from the gateway conforming to Profisafe over Profinet: the system integration used an ABB controller. With no experience of ISA100 in operation on a hazardous site, GasSecure and Statoil wanted to check the reliability of such a wireless system on site, so they did this by installing ten Yokogawa wireless temperature transmitters spread around the Statoil Kårstø gas processing plant in June 2011, with transmission distances of up to 100m. Of these, 8 worked well, with only a 1.09% packet error rate: the other two transmitters had been deliberately placed in obviously poor wireless locations, and gave more problems. But the efficacy of the ISA100 system was sufficiently proven.

The GS01 development continues, and in December 2011 the first completed units will be installed for trials in the Statoil Kårstø plant. The schedule after this is that in April 2012 a further test will be carried out on the Statoil Grane platform, offshore in the North Sea. The major first production project is for a new ConocoPhillips installation commencing in June 2012.

The GS01 is a fascinating development, and it will be interesting to see the performance proved offshore and in Scandinavian winter conditions! I am sure the developers will have an answer to ice and other hazards. One operator question from the audience was interesting: it is fine that new sensors introduced to the network join automatically and so on, but he had experienced problems of not knowing where each sensor was located, both with new sensors and when sensors were moved on the plant. While the wireless system cannot really help, his plant maintenance systems would seem to need to highlight this as a potential problem!

Advertisements

2 Responses

  1. […] Wireless sensors for hydrocarbon gas detection (Nick Denbow’s blog 9/11/2011) […]

  2. Very Nice Posting..And good topic for security of our business.I’m very encourage to read it, and i’m looking forward for your next post. Thanks for shared.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: