FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1973
1973 - 0139.PDF
FLIGHT International, 18 January 1973 Aircraft Production 87 19 Proving the tape: an IMC problem By WILFRED E. GOFF THE CONTINUOUS-PATH, NUMERICALLY CONTROLLED MACHIN ING of large integral components constitutes a considerable proportion, and the most difficult appli cation, of the NC techniques employed in the production of the modern airframe. This type of machining is essen tially a complicated combination of profile-milling and area-clearance involving the sinking-in and lifting-out of the cutters between the individual pockets or recesses in the component. These operations, which often have the effect of creating something like sheet metal from solid billet, are necessary in order to reduce weight and produce a distribution of metal to meet the stresses that the finished component will be required to resist. Programming the machining of such components is an extended, though theoretically straightforward, process of transcribing the information that define^, the shape of the finished part, into the particular programming language that has been adopted. In practice, however, the pro gramming job is a very complex procedure which, because of the great quantity of detailed data required to estab lish the paths of the cutters that produce the required form, is very susceptible to human error. Programming is a creative process and there is nothing in the initial stages to "jog the elbow" of the programmer and draw his attention to the fact that an error has been made, unless that error is one of logic. The computer into which the initially programmed information is fed will reject a logical error, but will still accept a logical pro gramme that may contain numerical errors. The fact that programming is prone to error leads to what is probably the most difficult, yet essential preliminary to actual production in NC machining—proving the accuracy of the control-tape. It is doubtful whether any organisation has yet succeeded in evolving an entirely satisfactory procedure to verify the information that is placed on the controlling tape. There is, possibly, no final solution to the problem of producing an accurate programme except to maintain particular care and vigilance at all stages in the program ming process. But even given extreme concentration and continual re-checking, it can hardly be expected that all errors will be avoided, or detected and eliminated. More over, there is a limit to the amount (and the value) of re-checking that can be done by the programmer on his own work. The old-established drawing-office practice of indepen dent checking by a second person—a time-consuming but, at least, practicable procedure—is hardly applicable in this type of operation. Although ostensibly a mathematical process, programming is in fact, idiosyncratic, in that no two programmers will process the same part in the same manner. An independent check would entail a detailed follow-through of the entire machining sequence—in effect, a second programming—which could lead to protracted disputes on machining procedure and would certainly give no assurance that errors had been eliminated. Another source of programming problems that can lead to delay and error is the peculiarly aviation practice of frequently introducing design changes. Though such modifications may appear trifling in their extent, they can often necessitate a complete replanning of the machining procedure in the area affected and, in extreme cases, could require the substantial re-programming of the part. So methods of checking the accuracy of the control-tape, mathematical, visual or both, are needed before the stage of actual machining is reached, as well as an actual cutting try-out before the final step is taken of machining the component itself. Various methods of checking are in use and others of greater refinement are being investigated and developed. One method, which can be applied immediately follow ing the computer stage of processing the programmed information, is the use of either a magnetic, or a punched- paper tape to produce a pen trace (on paper) of the path travelled by the cutter-centre. Depending upon the size of the part that has been programmed, such a trace may range in scale from TB to full-size and will show the movement of the cutter in all three axes. The larger the scale of the trace, the easier it is to analyse. Even a small-scale drawing will, however, convey a considerable amount of information to a pro grammer experienced in the interpretation of such diagrams. Some aid is needed in the clarification of these pen- traces, which may be of considerable complexity when completed. Inks of different colours can be used for differ ent cutter-path sequences, the same colour being used for a given path in the x, y and in the z traces. This practice has been adopted by the programming department of the British Aircraft Corporation at Weybridge on their Gerber full-scale drafting machine. By combining such checks with special care in the pro gramming procedure, much can be done to eliminate errors more or less at source, or at least before the tape Concorde rear-fuselage panel, top picture, being machined on a Max-E-Trace numerically controlled milling machine equipped with Ferranti tape control. Above, close-up of the profiling operation; rear- futelage frame for this aircraft and fin attachment points
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
