The design and analysis of a kerosene turbopump for a South African commercial launch vehicle.

• Main Author: Smyth, Jonathan.
• Other Authors: Brooks, Michael J.
• Language: en_ZA
• Published: 2014
• Subjects: Turbine pumps > South Africa.; Kerosene as fuel > South Africa.; Launch vehicles (Astronautics); Theses > Mechanical engineering.
• Online Access: http://hdl.handle.net/10413/11200

South Africa is one of the few developing countries able to design and build satellites; however it is reliant on other nations to launch them. This research addresses one of the main technological barriers currently limiting an indigenous launch capacity, namely the development of a locally designed liquid fuel turbopump. The turbopump is designed to function in an engine system for a commercial launch vehicle (CLV) with the capacity to launch 50-500 kg payloads to 500 km sun synchronous orbits (SSO) from a South African launch site. This work focuses on the hydrodynamic design of the impeller, vaneless diffuser and volute for a kerosene (RP-1) fuel pump. The design is based on performance analyses conducted using 1D meanline and quasi-3D multi-stream tube (MST) calculations, executed using PUMPAL and AxCent software respectively. Specific concerns that are dealt with include the suction performance, cavitation mitigation, efficiency and stability of the pump. The design is intended to be a relatively simple solution, appropriate for a South African CLV application. For this reason the pump utilises a single impeller stage without a separate inducer element, limiting the design speed. The pump is designed to run at 14500 rpm while generating 889 m of head at a flowrate of 103.3 kg/s and consuming 1127.8 kW of power. The impeller has six blades with an outer diameter of 186.7 mm and axial length of 84.6 mm. The impeller's high speed and power requirement make full scale testing in a laboratory impractical. As testing will be a critical component in the University of KwaZulu-Natal's turbopump research program, this work also addresses the scaling down of the impeller for testing. The revised performance and base dimensions of the scaled impeller are determined using the Buckingham-Pi based scaling rules. The test impeller is designed to run at 5000 rpm with a geometric reduction of 20%, using water as the testing medium. This gives an outer diameter of 147.8 mm and an axial length of 69.9 mm. At its design point the test impeller generates a total dynamic headrise of 67.7 m at a flow rate of 18 kg/s, with a power requirement of 15 kW. A method for maintaining a similar operating characteristic to the full scale design is proposed, whereby the scaled impeller's blade angle distribution is modified to maintain a similar diffusion characteristic and blade loading profile. This technique is validated by MST analysis for off-design conditions with respect to both speed and flowrate. === M.Sc.Eng. University of KwaZulu-Natal, Durban 2013.