Plotting cutter paths in 3D

I worked as a Coordinate Measuring Machine ( CMM ) operator in an automobile sheet metal fabricating plant ( 1989 - 1992 ), so this information is " State of the Ark ". There were 3 gantry type DEA ( Digital Electronic Automation ) machines in the temperature controlled room where I worked. One was a Lambda ( 5 w X 4 h x 8 meters long ) and two were Delta ( 4 w X 3 h X 5 meters long ) machines. These sizes are from memory, so may not be absolutely accurate. You could inspect a full size automobile under the Lambda, with room to spare !
These were outfitted with continuous scanning heads as well as the point to point inspection capability they were originally designed for. This area was referred to as the " Scanner Room ". Corporate strategy was that this was to be an interim process that was necessary only until the CAD-CAM software was able to function properly. It's primary purpose was to gather surface contour data from die construction " aids " , and post process this data to drive several different machine controllers. The aids were physical models poured of Hydrocal ? tooling plaster from the wood models ( masters ) of the sheet metal forms. Probes that had the same ( or slightly + ) diameter and radius were fitted to the spindle of the tracer, so the data was pre-compensated. The operator would determine what cutters to use for the areas that he would define on the aid that was to be scanned. A typical quarter panel ( rear fender ) might have more than a dozen polygon shaped areas marked out. Each would become a separate data file. The DEA machine controllers were PDP11-23's with RT-11 for the operating system. The Delta's had Digital Equipment Corp. PDP11-24 auxiliary computers running RSX-11/M and were used for post processing ( posting ) the data. The Lambda used a PDP11-85 ( also running RSX11/M ) for posting data. DEA supplied several different software utility programs for plotting the posted data on HP-7475 pen plotters ( HPGL ). These could handle 11 X 17 inch paper. We couldn't get 3D paper at that time, so the software was used to rotate the views that we plotted to verify the cutter paths that were output to a file by the post processor. This eliminated the need for the 3D paper. One program was very similar to the new EMC Backplots - you could specify any angle for any axis. This was rarely used but was handy to have when a feature was not readily defined by a standard view. The reason for the rare usage was that the boss wanted standard views so everyone would stand a chance of knowing what they were viewing. One of the other plotting programs would give the choice of X, Y, Z, or Isometric views. The plots could be scaled and windowed. You had the choice of plotting all the rapid ( G0 ) moves or only the first 5 of them. This allowed the operators to review all the moves and look for accidental reversals of the cutting path or unanticipated gouges into the surface. It was very important to avoid any potential crash since the machine downtime was probably more costly to the company than the outrageous expense of the repair to the large mills. You would probably be surprised to see how easy it was to pick out the bad data just from viewing the plot. A text editor ( EDT or TECO ) was then used by the operator to straighten out the bad areas of the G code file before it was sent to the machines.

BTW, the wire cut Elox machines came in with an arm and weighted pen holder and a surface to place the paper on. This is similar to what Jon is describing. Of course this was offset from the wire guide and splash guards, but would plot the exact replica of the G code program.

IMO, EMC and Backplots ( thanks Ray,Paul, and all ) can provide a capability that the major auto companies paid tens of thousands ( possibly much more ) of dollars for a decade or so ago.

Thanks for the bandwidth,

Dave


[Non-text portions of this message have been removed]