Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.632546
Title: Analysis of cold rolling with particular reference to roll deformations
Author: Golten, Jack Winston
Awarding Body: Swansea University
Current Institution: Swansea University
Date of Award: 1969
Availability of Full Text:
Access from EThOS:
Abstract:
Several existing theories of cold-rolling were reviewed with particular attention to Orowan's Homogeneous theory. It was decided to use this theory throughout us it was the most well tried and proved of those available. Many of the other theories are still in a speculative state, and often require parameters which are not well understood. Orowan's theory enabled one to examine the effect of varying basic parameters with the least amount of complication. The effect of roll force on roll flattening was discussed and a simple model was used based on the assumption that the elastically deformed arc of contact was circular (Hitchcock's Formula). This was incorporated in a digital computer program for evaluating roll force and torque. Consequently a hybrid computing method for obtaining rapid solutions of the rolling equations was developed and using this technique results were obtained faster than the equivalent purely digital solution. There is a need for measuring stress distributions in the roll-bite to substantiata the theory, so a thin film manganin pressure transducer was proposed for measuring these stresses. The techniques involved in the manufacture and application of such a transducer are dealt with in detail. Results were obtained using the transducer for surfaces in rollil1g contact, but it was not possible to ohtain satisfactory results when rolling strip,so that this experimental technique was abandoned. Suggestions ar.e made fot future work on this type of transducer. -viii- Orowan's theory lVas modifiad to incorporĀ£!.te arbitrary roll profiles since it was felt that the assumption of the cir~ular roll surface was an over-simplification. The method using influence functions to obtain roll deformations due to normal loads was reviewed (Jortner). The influence function for point shear loads was derived using a similar technique. A program was writt.m which included the effects of shear stresses on roil deformations. The program calculated roll force and torque more precisely than hitherto; moreover, a coefficient of friction that could vary throughout the roll bite was incorporated. The effect of shear forces on roll deformation was discussed and in certain cases was shown to affect roll force and torque significantly. The use of complex variables j.n stress analysis was reviewed. A novel method for producing a series solution for point loada using Dirac Functions is presented. Finally closed-form solutions were obtained for point shear and point normal loads, and found to tie up with those obtained using elementary stress distributions. The advalltases of this method are that no a priori assumptions need be made, and no complex integrations are required to obtain surface deformations. This method also explains certain results which are obscured using ~he previous approach. The final chapter applied the preceeding theory to explain the so called "speed-effect". The variables in tha rolling theory were examined and the coefficient of friction emerged as the IDOst likely speed-dependent parameter. Using exper.imental data on the - iY.: - speed eff(!ct obt;'linad by H. Ford, theor.etical resulU. predicted the e.:>rre!.!t roll forCl~ and torqu,: for thick a.Ulealed mater.ial. Using the same coefficient of friction, the results obtained when the strip became thin and hard were much lower than those obtained in practi~e. A stress dependent coefficient of friction was proposed to account for this, but it was shown that using the modified OrO'.Jan theory it was not possible to obtain. sufficie.ntly high roll forces. The speed effect behaviour was produced by choosing a coefficient of friction which decreased with the relativ(:~ velocity between the strip and roUs. The limitations of present theory are discu3sed and suegestions for further work are made.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.632546  DOI: Not available
Share: