Theoretical and experimental study of a hydraulic-assisted turbocharger (HAT)
Publication Date
July 31, 1923
Creator
Abstract
Efficiency improvements for internal combustion engines are essential to counteract the harmful emissions that negatively contribute to the Climate Change issue. Hydraulic-Assisted Turbochargers (HAT) fall within these measures; according to APC roadmaps and retrofit considerations, heavy-duty vehicles are more likely to benefit from HAT as thermo-propulsion systems and hybrids are predicted to have a strong role in this transportation category. The research work carried out in this thesis aims to theoretically and experimentally investigate a HAT prototype and understand the feasibility of this solution as a turbo-lag mitigation strategy and efficiency enhancement technique. A bespoke experimental rig was designed and manufactured to mimic the HAT system in a hydraulic hybrid vehicle configuration. High-pressurised oil was stored in a 20L accumulator up to 280 bar and released through two automated solenoid valves and two nozzles into the rotating assembly. A HAT prototype inherited from a previous research project was tested; the unoptmised bearing casing caused hydraulic lock, high friction and low speeds. A free running Pelton was then used to inform the new design which included in its final configuration shrouding, open draining and ceramic ball bearings to minimise windage losses, hydraulic lock and bearing friction respectively. This final design was embedded into a standard Garrett GTX-2867r and tested in a bespoke engine-like setup. At the maximum available pressure of 250 bar and using 2x1.5mm nozzles, the maximum speed performance was 63k rpm in 0.64s producing a boost of 1.31 bar. The work at the compressor does not influence the maximum speed however it increases the time response by 10%. If compared to an Electric-Assisted Turbocharger, the HAT shows a lower time response by 42-57% on average, but its absolute maximum speed is lower by 23% on average. A 0D computational model was developed to mimic the HAT dynamic performance. The model can predict, adapting the windage losses equation with empirically deduced parameters, the maximum speeds of the assembly, however, it underestimates the time response of 12%.
Item Type
ethesis
Thesis Type
PhD
Supervisors
Subjects (LC)
Associated Schools / Departments
Department of Mechanical, Materials and Manufacturing Engineering (UK)
eprints ID
72944
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
PhD thesis Annalisa Ferrante v4.pdf
Type
Full-text
Description
Examined. Thesis after official corrections after VIVA
Size
5.05 MB
Format
Adobe PDF
Checksum (MD5)
7fca43e8d91ce0ca5dcf96bf662f9fa3