Upgrade and Shakedown Test of a High Temperature Fluoride Salt Test Loop

Abstract

A FLiNaK high temperature test loop, which was designed to support the Thorium Molten Salt Reactor (TMSR) program, was constructed in 2012 and is the largest engineering-scale fluoride loop in the world. The loop is built of Hastelloy C276 and is capable of operating at the flow rate up to 25m3/h and at the temperature up to 650°C. It consists of an overhung impeller sump-type centrifugal pump, an electric heater, a heat exchanger, a freeze valve and a mechanical one, a storage tank, etc. Salt purification was conducted in batch mode before it was transferred to and then stored in the storage tank. The facility was upgraded in three ways last year, with aims of testing a 30kW electric heater and supporting the heat transfer experiment in heat exchanger. Firstly, an original 100kW electric heater was replaced with a 335kW one to compensate the overlarge heat loss in the radiator. A pressure transmitter was subsequently installed in the inlet pipe of this updated heater. Finally, a new 30kW electric heater was installed between the pump and radiator, the purpose of which was to verify the core’s convective heat transfer behavior of a simulator design of TMSR. Immediately after these above works, shakedown test of the loop was carried out step by step. At first the storage tank was gradually preheated to 500°C so as to melt the frozen salt. Afterwards, in order to make the operation of transferring salt from storage tank to loop achievable, the loop system was also preheated to a relatively higher temperature 530°C. Since the nickel-base alloy can be severely corroded by the FLiNaK salt once the moisture and oxygen concentration is high, vacuum pumping and argon purging of the entire system were alternatively performed throughout the preheating process, with the effect of controlling them to be lower than 100ppm. Once the salt was transferred into the loop, the pump was immediately put into service. At the very beginning of operation process, it was found that flow rate in the main piping could not be precisely measured by the ultrasonic flow meter. Ten days later, the pump’s dry running gas seal was out of order. As a result, the loop had to be closed down to resolve these issues.