Email Document Reference

Enter your email address below and the reference for this document will be sent to shortly from webmaster@ceesi.com.

Title: Calibration Errors Of Ultrasonic Meters In The Bernoulli Laboratory Due To Non-Isothermal Flow Conditions.
Author: Aernout Van Den Heuvel, Frans Doorman, Piet Van Den Herik, Arjan Stehouwer, Robert Kruithof,
Source: 2009 North Sea Flow Measurement Workshop
Year Published: 2009
Abstract: The Bernoulli laboratory (Westerbork) in the Netherlands, jointly operated by Gasunie, NMi and KEMA, holds a record of over 30 years in calibrating very large gas meters at high pressure with natural gas. NAM, a major producer of natural gas in the Netherlands, utilises a large number of ultrasonic meters. Elster-Instromet is a world wide operating manufacturer of multi-path ultrasonic meters. Gastransport Services is the national gas transporting company of the Netherlands, who is connected to the network of NAM via several delivery stations utilizing ultrasonic meters. Due to the results of five calibrations of 24-inch ultrasonic meters in the Bernoulli laboratory early 2008, a quality check procedure (number QC-11) was started by the Bernoulli laboratory. Four out of five ultrasonic meter showed strongly non-linear behaviour at low flow rates, with a maximum at about 560m3/h (approximately 2% of Qmax) with errors peaking as high as +2%. Taking into account, the demands set in ISO/FDIS 17089-1: 2009(E), all four meters would have been rejected had the standard had already been ratified. Together, the manufacturer, the user and the calibration facility decided to give high priority to quickly identify and resolve this problem. After excluding all straightforward errors, two possible causes remained on the short list: a disturbed flow profile at low flow rates, typically below 2 m/s, and meter problems. A series of experiments and improvements to the calibration facility were executed in the course of 2008: an additional temperature measurement at the bottom dozens of calibrations of one specific 24-inch ultrasonic meter, made available by the user (this meter was even calibrated in the up side down position) thermal lagging was improved and finally, in January 2009, the full measurement section was insulated. Applying full thermal lagging with 15C temperature difference between gas and ambient, temperature differences were largely reduced, as was the error of the ultrasonic meter at 800m3/h. Also the ultrasonic meter diagnostics showed a large improvement of the flow profile. Although the thermal lagging had insufficient heat transmission resistance, which prevented a 100% success, the experiment showed, without doubt, that non-isothermal flow conditions were the root cause of the calibration errors. Non-isothermal flow conditions have previously been addressed at the 2005 Flomeko conference 3, however, the discussion at that time concentrated on the temperature measurement error and the effect was characterised as stratification. In that paper, other effects of non-isothermal flow: non-axial velocity components and an asymmetric flow profile, were not recognised as a problem. The majority of gas meters calibrated in Bernoulli laboratory, turbine meters, are not sensitive to non-isothermal flow conditions. Ultrasonic meters however, measure velocity components and may be very sensitive to a combination of non-isothermal flow, depending on their path configuration. Clearly, the Q.Sonic-4C of Elster-Instromet is sensitive to these non-isothermal flow conditions present in Westerbork.

In order to prevent spam and automated file downloads for documents within the Measurement Library, please follow the instructions below and then you will be able to email a reference to this article.