Venturi Fouling and What Can Cause an Overestimate of the Flow Rate by One Percent
NURETH-14 - 2011 September 25-30


Presented at:
NURETH-14
2011 September 25-30
Location:
Toronto, Canada
Session Title:
F5-1 Instrumentation Technique

Authors:
Peter Hedberg (Remote CAE AB, Sweden)
Hans Nilsson (Ringhals AB, Sweden)
  

Abstract

In this study, we are looking for phenomena which can explain the effect that venturi fouling has on the measurement of feedwater flow rate in a PWR, Unit 3 of Ringhals Nuclear Power Plant, Sweden. When hydrazine is injected into the feedwater, it reduces the deposits of magnetite on the wetted surface of the venturi, and elsewhere at the given temperature ~200°C. This changes the reading from the flow measuring device and becomes closer to the originally calibrated data. Over time magnetite is rebuilt on the walls. We are searching for what can overestimate the mass flow rate in the order of 1%. Potential explanations are; changes in the venturi cross section area, change in properties of the fluid mixture, effects of suspended magnetite particles, changes in wall shear stress due to regular wall roughness, changes in the wall shear stress due to self-organized ripple wall roughness, changes in swirling flow due to wall roughness, separation in the diffuser part of venturi due to wall roughness, changes in the velocity profile (entering the venturi) due to wall roughness, and local buildup of deposits around the pressure taps.

Besides visual inspection of recently replaced venturi meters, numerical and theoretical estimates have been used to find the most likely explanation. We have derived a new wall function to introduce the self-organized ripple wall shear stress and used it in CFD (Computational Fluid Dynamics) simulations.

The first conclusion from the simulations is that the required regular wall roughness is not consistent with the observed thickness of the deposit on the wetted surfaces. Nor does the cross section area change sufficiently to make the flow rate deviate by 1%. The changes in fluid properties, due to the fluid mixing, are not significant. This is also true for a fluid with suspended magnetite particles. The only effect that is large enough to overestimate the flow by 1% is the self-organized wall ripple, for the observed deposit thickness. Also, we find that the deposit in the venturi has a larger effect on the mass flow rate reading, than the deposit in the pipe system upstream from the venturi. The wavelength of the self-organized ripple can be determined from the friction velocity, together with kinematic viscosity. That is, the wavelength is determined by fluid flow parameters.

Individual Conference-Paper Copies (Electronic Where Available):

  • For CNS members, the first 5 copies per calendar year are free, and additional copies are $10 each.
  • For non-members, the price is $25 for the first Conference-paper copy in a request, and $10 each for additional copies of papers in the same conference and in the same request.
  • Contact the CNS office to order reprints.