Performance of reduced-scale Vortex amplifiers used to control glovebox dust

Abstract

Ventilation systems for a nuclear plant must have a very high reliability and effectiveness. In this application, fluidic devices have advantages which electro-mechanical and pneumatic devices lack. Fluidic devices will not easily wear out, they have a relatively fast response and in some cases they may be cheaper than an equivalent conventional device. Most importantly, they have fewer moving parts (usually none) so are inherently reliable, so long as the fluidic design is effective. So vortex amplifiers (VXA) are ideal for active ventilation systems where access for maintenance is problematic.
From 1995 to 2000, space limitations at Sellafield drove the desire to minimise VXA size and also glovebox size. Recently completed plant expansions use a smaller version of VXA produced by scaling geometrically the existing standard model. It is called the mini-VXA. Subsequent performance of the mini-VXA has been disappointing with high oxygen levels noted in the inerted gloveboxes; this required an expensive increase in the inert gas supply rate of gloveboxes to mitigate against fire risk.
After doing experiments using a mini-VXA and typical glovebox, the author has confirmed the high 02 levels. The 02 distribution in the glovebox indicates that oxygen is entering the glovebox by the VXA supply ports; against the general direction of flow.
The ultimate source of this back leakage is the control port (that is open to atmosphere) and smoke visualisation studies on the mock VXA indicate a mechanism. This is due to separated flow patterns with excessive control port momentum. A temporary solution using an orifice plate and spacing chamber has been shown to reduce essential nitrogen supply to one quarter that without the modification. Addition of the orifice plates enables further reduction in nitrogen use, and the smallest orifice tested performs best with no discernable cost in pressure drop and therefore fan power. The author also found the following points. The ratio of control port area to supply port area is a critical parameter affecting mixing of the two airstreams. Yet exit port area is unimportant. The ratio of supply port area to exit port area has no influence on discharge coefficient (at least within the scope of current work). It is also identified that the ratio of chamber height to exit port radius does not affect the discharge coefficient or two angle parameters. Doubling chamber height, supply port area and control port area at the same time has a slight effect on the discharge coefficient (attributed partly to a viscous effect), but no effect on the two angle parameters. The chamber height has little effect on Reynolds number. If the supply port area is not too small relative to the exit port, the supply port area will not significantly affect Reynolds number. The use of discharge coefficient and the two angle parameters to characterize VXA performance breaks with the traditional form of dimensionless characteristics that are used for the purpose. Testing these alternate characteristics has enabled the momentum
(which dominates control of VXA performance) to be more explicitly expressed in updated design rules.