13–18 Oct 2019
Chattanooga Convention Center
US/Eastern timezone

Manufacturing Process Improvements of Bubblers for Helium Gas-Injection in SNS Pulsed Mercury Neutron Source Target

17 Oct 2019, 16:30
2h
Meeting Room 4 & 5 (Chattanooga Convention Center)

Meeting Room 4 & 5

Chattanooga Convention Center

Poster Target/Moderator Poster

Speaker

Mr Isaac Waldschlager (Oak Ridge National Laboratory )

Description

The Spallation Neutron Source (SNS) at Oak Ridge National Laboratory (ORNL) currently produces spallation neutrons by the impingement of an accelerator-sourced 1.4-MW proton beam pulsed at 60-Hz onto a mercury source target. The subsequent deposition of energy within the liquid mercury target creates pressure waves of nonnegligible strength. This causes a rapid expansion and collapse of the mercury resulting in considerable material stresses and cavitation in the walls of the target vessel housing the mercury. Currently, the most effective way of mitigating this damage is by absorbing the expansion of the mercury within the mercury itself. To accomplish this, helium gas is injected into the mercury flow by arrays of orifice tubes in the mercury flow passages. These orifice arrays are commonly referred to as “bubblers.”
Many practical challenges arise in the manufacture of helium bubblers. These challenges are due almost entirely to the relatively small diameter (10-20 µm) and quantity of holes prescribed to produce helium bubbles of appropriate size and concentration within the bulk mercury flow. Initially, the manufacturing plan was to press the individual orifice tubes into the bubbler bodies. This proved difficult because the relatively large imperfections in the fit-up between tube and body would often result in excessive gas leakage around the orifice tube.
To avoid the problems with leakage around the press fit orifice tubes, the current design calls for GTAW welding of the thirty separate orifice tubes into bubbler bodies. This configuration facilitates the interchanging of tubes with different sized orifices; however, while shipping, handling, testing, and installing these bubblers, there is ample opportunity to foul the bores and permanently block or greatly reduce the flow through these bubblers. Various attempts to clean blocked orifice tubes have proven only marginally successful. It is often necessary to drill out and replace plugged orifice tubes discovered in bubblers prior to installation into the mercury target module.
Initially, the vast majority of tubes had to be removed and replaced to assemble a successfully functional bubbler. To avoid rework, strict cleanliness and handling protocols were put into place while fabricating, handling, and installing bubblers. The result was a reduction in the cases of rework; however, blockage of orifice tubes still plagues the current process.
In an effort to further reduce the chances of fouling the orifices, a bubbler has been designed such that the individual orifices are not welded as individual piece parts but instead the body of the bubbler encompasses all features on the assembly in one machined part. This design means the final operation in the manufacture of the bubbler subassembly is the laser drilling of the orifice holes themselves such that the chance of fouling the holes is reduced as much as possible. Because the design has square-tipped extensions into the mercury flow, this design is referred to as “toothed bubblers.” Four toothed bubblers have been produced to ensure a matched performance between the current design and this new design. A pair with 20 µm diameter holes and a pair with 18 µm diameter holes. Unfortunately, the 18 µm pair suffered from multiple clogged orifices and the flow data from them was unusable. The pair with 20-um diameter holes had a flow through all but one of a total of 60 holes and had total overall flow rates well matched between the two; however, the flow was significantly less than the targeted design flow. The reason for the shortfall in the achieved flow is not yet fully understood and is the subject of future study.

Primary author

Mr Isaac Waldschlager (Oak Ridge National Laboratory )

Presentation materials

There are no materials yet.