Improvement of tool utilization when hard turning with cBN tools at varying process parameters
(2021) In Wear 477.- Abstract
Hard turning of helical gear hubs produced from low-carbon alloyed steel with a bulk hardness of HRC 35, and case hardened up to HRC 60–63 with PCBN tools was considered in this study. The workpieces after heat treatment are characterized by a non-uniform thickness of the hardened surface layer which results in a high variation of measured HRC hardness. The deviation of surface microhardness is even greater (HV 314–742) due to the presence of an oxidized layer that should be removed before measurements. When also considering workpiece run-out and deviations in depth-of-cut due to the distribution of the workpiece diameter, all those factors result in a non-uniform development of tool wear that leads to a large deviation in tool life. To... (More)
Hard turning of helical gear hubs produced from low-carbon alloyed steel with a bulk hardness of HRC 35, and case hardened up to HRC 60–63 with PCBN tools was considered in this study. The workpieces after heat treatment are characterized by a non-uniform thickness of the hardened surface layer which results in a high variation of measured HRC hardness. The deviation of surface microhardness is even greater (HV 314–742) due to the presence of an oxidized layer that should be removed before measurements. When also considering workpiece run-out and deviations in depth-of-cut due to the distribution of the workpiece diameter, all those factors result in a non-uniform development of tool wear that leads to a large deviation in tool life. To uphold quality and avoid scrap, industry uses a fixed conservative tool life (number of parts machined), which leads to many inserts being underutilized, which in turn, means extra expenses considering the price of PCBN tools. The present study addresses the development of different strategies for monitoring of tool wear and prediction of tool life aimed to increase productivity. Laboratory machining tests are performed to optimize the type and number of sensors necessary for data acquisition, to estimate the significance of the data used for the development of prediction model, as well as the efficiency of machine learning algorithms for the selected data and strategies to increase the algorithm's accuracy.
(Less)
- author
- Gutnichenko, O. LU ; Nilsson, M. LU ; Lindvall, R. LU ; Bushlya, V. LU and Andersson, M. LU
- organization
- publishing date
- 2021-07-18
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Hard-turning, Machine learning, PCBN, Tool condition monitoring, Tool utilization, Tool wear
- in
- Wear
- volume
- 477
- article number
- 203900
- publisher
- Elsevier
- external identifiers
-
- scopus:85104338416
- ISSN
- 0043-1648
- DOI
- 10.1016/j.wear.2021.203900
- language
- English
- LU publication?
- yes
- id
- 669d7923-3e12-4668-b7f8-969ddb5ff23d
- date added to LUP
- 2021-04-27 08:33:57
- date last changed
- 2023-11-08 13:28:10
@article{669d7923-3e12-4668-b7f8-969ddb5ff23d,
abstract = {{<p>Hard turning of helical gear hubs produced from low-carbon alloyed steel with a bulk hardness of HRC 35, and case hardened up to HRC 60–63 with PCBN tools was considered in this study. The workpieces after heat treatment are characterized by a non-uniform thickness of the hardened surface layer which results in a high variation of measured HRC hardness. The deviation of surface microhardness is even greater (HV 314–742) due to the presence of an oxidized layer that should be removed before measurements. When also considering workpiece run-out and deviations in depth-of-cut due to the distribution of the workpiece diameter, all those factors result in a non-uniform development of tool wear that leads to a large deviation in tool life. To uphold quality and avoid scrap, industry uses a fixed conservative tool life (number of parts machined), which leads to many inserts being underutilized, which in turn, means extra expenses considering the price of PCBN tools. The present study addresses the development of different strategies for monitoring of tool wear and prediction of tool life aimed to increase productivity. Laboratory machining tests are performed to optimize the type and number of sensors necessary for data acquisition, to estimate the significance of the data used for the development of prediction model, as well as the efficiency of machine learning algorithms for the selected data and strategies to increase the algorithm's accuracy.</p>}},
author = {{Gutnichenko, O. and Nilsson, M. and Lindvall, R. and Bushlya, V. and Andersson, M.}},
issn = {{0043-1648}},
keywords = {{Hard-turning; Machine learning; PCBN; Tool condition monitoring; Tool utilization; Tool wear}},
language = {{eng}},
month = {{07}},
publisher = {{Elsevier}},
series = {{Wear}},
title = {{Improvement of tool utilization when hard turning with cBN tools at varying process parameters}},
url = {{http://dx.doi.org/10.1016/j.wear.2021.203900}},
doi = {{10.1016/j.wear.2021.203900}},
volume = {{477}},
year = {{2021}},
}