FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1932
1932 - 0764.PDF
SUPPLEMENT TO FLIGHT 54 THE AIRCRAFT ENGINEER JULY 29, 1932 possible, so that some slip is inevitable. But, by making Ri and R, as nearly equal as possible, the amount of slip will be reduced to a minimum. This consideration is less important where the radius of bend is large—as at the curve " D "—because the damage done and power lost in friction when slip occurs is proportional to the work done in bending the metal. On the other hand, the transverse length of surface being worked is generally greater where the radius is large, thus increas ing the power loss due to slip. Consequently, it is desirable to have all bends; sharp or otherwise, as nearly in line as possible. Incidentally, there will be increased slip if the upper rolls are not free to rotate independently, due to the fact that the angular velocity of both rolls will be the same if they are geared together, whereas the diameters cannot possibly be the same. Such a condition results in the slip between the rolls themselves being of a fixed value. If, on the other hand, they are not geared together the slip automatically reduces to a minimum. One other point that needs emphasis is that the line of work should be as nearly as possible midway between the two spindles in order to keep the smallest diameter as large as possible. The reason being that any given linear difference between R, and R,, say, will cause a greater difference in peripheral speed if R, and R2 are themselves small, since, obviously, the difference becomes a greater proportion of the whole. Referring to Fig. 2 (b), it will be seen that, although no very severe work is done in the second pair of rolls, three more sharp corners and one curve are partially formed. Fig. 2 (c) shows the section after passing through the third pair. It will be seen that the corners " B " and " C " are now finished, while " F " and the two curves " D " and " E " are well on the way. " D " could easily have been finished in this or even the second pair, but a little consideration will show that to have done so would have made it impossible to form the corners " B " and " F " satisfactorily and would also have given the section an undesirably great depth. In the fourth pair " D," " E " and " F " are finished and the section is completed. In passing through this last pair it is distorted to overcome the effects of spring- back, but takes its proper shape on leaving the rolls. The subject of spring-back will be dealt with in detail later. Considering now the question of waviness in side view as the metal passes through the mill. Since the section changes at each pair, certain parts of the metal must inevitably wave, but the amplitude of the waves can be restricted considerably by careful design. The diameters of the lower rolls are, of course, the governing factors because the lower spindles are all on the same hori zontal level. When determining these diameters, the middle portion of the strip should be considered chiefly, since the extremities being in the nature of free edges are not so likely to stretch or take permanent sets through waving. With regard to the number of pairs of rolls that are necessary to produce a section satisfactorily, experience is the only true guide. The aim, of course, is to use as few as possible in order to minimise initial cost, main tenance cost—all rolls need trueing up occasionally and some very frequently, cost of setting up for each batch of material and power consumption—since all pairs in excess of the optimum number will only increase the frictional losses, thereby using more power (or slowing down the work if the speed is not controlled). On the other hand, if too few rolls are used the amount of work required of each pair becomes excessive ; with the result that either the mill is unable to cope with the loads imposed or the strip stretches—or both. Such stretching, if it occurs at the edges, will result in a straight centre portion (perhaps) and wavy edges. If the middle of the strip is stretched, then hills and hollows will occur in the main part of the section. In either case the section will probably be bowed length wise. The most serious disadvantage, however, to having too few rolls is the likelihood of cracks developing in the corners of the section owing to the working being too severe. These are often practically invisible, but are, nevertheless, such a potent source of danger, that the presence of one hair crack half an inch long will necessitate the scrapping of possibly 30 ft. of material. Inspection for cracks is a lengthy and, therefore, costly process, and in certain cases where the section is a closed one with re-entrant bends, is not possible at all without destroying the work; so that it pays to have sufficient rolls to put the work beyond suspicion in the first place. One rather serious problem which arises in the actual process of rolling, is the elimination of bowing (longi tudinal curvature of a length of rolled material). This defect is not usually attributable to errors in design of the rolls. It may be due to errors in setting up, but the cause is more likely to be found in the raw material itself. Taking the section in Fig. 2 as an example, if the strip is curved when laid out flat before rolling, the finished section will almost certainly be bowed sideways. This can sometimes be overcome by deliberate uneven tightening of the rolls, but such action is most undesirable, as it may lead to all sorts of trouble—from cracked rolls to cracked work. Generally speaking, if the bowing is too serious to ignore and can be directly traced to the raw material, the stuff should be returned to the makers. When the strip is laid out flat, wavi ness is often apparent. If at all serious, this also will lead to bowing. If the edges are wavy, sections similar to that of Fig. 2 will tend to bow downwards on leaving the rolls. Conversely, bagginess of the middle of the strip will cause upward bowing. Undue loss of thickness towards the edges of the strip will also tend to have the same effect. On the other hand, however, if the material is not faulty, either the setting up or the rolls will be to blame. In which case the following remarks should serve as a guide to correction: — Lateral bowing. Rolls may be unequally tightened—e.g., outer clamping; screws too tight—this will cause the section to bow towards the mill on leaving the last pair. Rolls not in proper alignment sideways. Upward bowing. Centre portions of one or more rolls may be over size. Rolls working on middle portions of strip may be too tightly clamped—thereby squeezing the metal. Downward bowing. The reverse of the causes of upward bowing. It should be noted that the above remarks apply to sections similar in shape to that illustrated and would not necessarily be true for sections of a radically different character. General Design Considerations. The first requirement is an accurate enlarged layout of the finished section, preferably ten times full size. This scale gives ample magnification and minimises the possibility of errors occurring, since it is only neces sary to move the decimal point when translating dimen sions from the actual section to the layout. The section so drawn should be divided up into its elemental straights and curves as shown in Fig. 9, and the lengths of all the pieces calculated and checked graphically. This accuracy is important because a small error at the commencement may lead to unsatisfactory results with out giving any clue as to their cause. Much time may thus be lost in costly experimenting before it is decided to work through the design again from the beginning. Next, the proposed sections at the various rolls should be drawn out—preferably to the same enlarged scale. 710/
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
