Continuous efforts by Rank Pneumo resulted in the development of a "Technology Test Bed". The technology introduced in 1988 is used for understanding the concept of diamond turning.
The machine makes use of several concepts that were not used previously in the diamond turning field, however, most of the technology is careful implementation of previously tested and proven techniques.
The machine is supported on a 48 x 96 inch block of granite with a thickness of 18 inches. The size was not based on the structures to be mounted - they were the dimensions of an old surface plate from the company’s quality assurance department. The large area of the stone enabled the selection of a laser interferometry layout to reduce the number of optics to be used instead of space.
The granite is supported by an active air leveling and vibration isolation system. The equipment used is identical to that provided with every MSG-325, with the exception that external surge tanks were used instead of the steel frame from the 325. The external tanks used were the common 20 lb propane tanks.
The gas tanks are provided in the “air console” assembly which is made of plywood. The assembly is used for mounting different components such as valves, gauges, filters, and regulators used for operating the spray mist coolant, the tool set station, the spindle and the leveling system.
The equipment mounted over an isolated free standing air receiver is used for carrying out air drying and filtration. Unlike the conventional vane type rotary pump, a noise-free aspirator generates vacuum for the chuck.
The slides are supplied with 15 cubic inches of oil in a minute at 250 psig through a small hydraulic pump and tank unit. The system also employs a DC servo motor-driven gear pump, located at the right of the granite.
There are two different types of slides used for the cutting tests. The X-axis slide is a box way design, and the slide top has a width of 15 inches and length of 18 inches. An intermediate slide triggered by a ball screw and nut drives the slide and the X axis via a hydrostatic thrust bearing. The bearing is used to avoid cyclic errors generated by the ball screw.
The Z-axis slide, however, uses a different configuration, and has similar dimensions to those of the X-axis. This slide features dovetail designs. Upon comparing the cost of manufacture of the two configurations, it was revealed that the box way design was more cost-effective than the Z-axis design.
The configuration of machine includes the spindle along the X-axis and tool holder along the Z-axis. The X-axis is placed on fabricated steel risers so as to allow the way covers of the Z axis to pass below the X-axis. The way covers do not influence the motion accuracy of the slides as they do not make contact with any part of the machine structure.
The position feedback of both X-axis and Z-axis slides are explained as below:
Laser interferometers are used to provide feedback of slide position. The cutting tests were carried out using the Hewlett-Packard equipment. During the cutting tests, the X-axis beam path was positioned in the central axis of the spindle, and behind the chuck.
The Z-axis interferometer was located on the centerline of the axis and beneath the way cover before the slide. Gaps have been provided for the addition of a second interferometer on this axis to measure pitch. A fourth interferometer was included in the system as a refractometer following the completion of cutting tests. Test data has been collected while determining the appropriate way to load data in the machine controller.
The cutting tests were carried out using a Hewlett-Packard HP-5507 controller with two 10936A servo-axis boards. The IEEE-488 bus was used to download the tool path from an IBM XT operator interface to the controller. The tool path information was produced using the Rank Pneumo Tool Path Generator software. The HP servo-axis boards were produced using the intermediate interpolation coordinates. The feed rate of the tool can be adjusted by controlling the coordinates and sample clock rate. The clock rate for all cutting tests was maintained at 1 ms.
To some extent, the test bed capacity is determined based on the spindle location on the X-axis. The spindle was placed at the rear side of the slide during testing, and cutting was carried out on the front side of the parts. The toolholder location on the Z-axis determines the axial capacity. The maximum length part of the test setup was taken as 12 inches.
The spindle used for the test cutting was a UP-2000 lathe spindle with a maximum speed of 2400 rpm and a load capacity of over 100 lb. Except for the large flat that was cut at 800 rpm, all cutting procedures were carried out at 100 rpm.
Contour Fine Tooling Ltd supplied the cutting tools for the cutting tests. Some of the key tools used in the cutting process include zero degree rake diamonds with maximum of 20 µm waviness over 100 degrees and 0.030 radius. The tools were placed in an Aloris toolholder mounted on a Rank Pneumo micro height adjuster. The feed rate was set at 0.002 inches per revolution.
Cutting tests were carried out on electroless nickel plated steel and using 6061-T6 aluminum, copper, acrylic, and germanium parts. The size of the workpiece ranged from 0.5 inches diameter proof of center studs to 18 inches diameter flats. Flats, aspheres, spheres and some stepped parts were cut. Both machine axes were energized in all axes. Figure 1 shows the schematic of workpiece.
Figure 1. Schematic of workpiece
The results of the cutting tests are shown in the following illustrations. The data are obtained from RTH, Wyko and Zygo measuring equipment.
Figure 2a. Surface profile of 2 inch diameter aluminum flat
Figure 2b. Histogram of surface heights of 2 inch diameter aluminum flat
Figure 2c. Cross-section of 2 inch diameter aluminum flat
Figure 3a. Surface profile of 2 inch diameter electroless nickel plated flat
Figure 3b. Histogram of surface heights of electroless nickel plated flat
Figure 3c. Cross section of 2 inch diameter aluminum part with 6 µ inch deep grooves
Figure 4. Partial RTH FormTalysurf trace showing 1 µ inch and 5 µ inch steps 0.100 inch apart on a flat part
Figure 5a. Surface profile of 2 inch diameter 10 inch radius aluminum sphere
Figure 5b. Histogram of surface heights of 2 inch diameter 10 inch radius aluminum sphere
Figure 5c. Cross section of 2 inch diameter 10 inch radius aluminum sphere
Figure 6. Schematic of 18 inch` diameter aluminum flat
The surface finishes of the components machined using the test bed were two to four times better than those achieved using other diamond turning equipment. Form accuracy corresponds to the lack of "hash" or noise in the slide traces and straightness of the new slides. Therefore, the accuracy was improved over parts machined on other machines.
Upon proving the operational state of the controller, all the components were cut one after the other without the need for extensive tuning of the equipment. Although efforts to carry out cutting using 0.1 µ inch steps were made, these steps could not be found despite having the evidence that axes respond to commands at that level. The reason for this can be determined by using RTH Talystep to analyze the surface. The undulations of 1 µ inch order can be observed even in the topography of good surfaces. The 0.1 µ inch steps are masked by the general surface texture. The machine error compensation can be applied to diamond turning machines at 0.1 µ inch level without producing detectable marks on the surface or affecting surface finish.
This information has been sourced, reviewed and adapted from materials provided by Precitech.
For more information on this source, please visit Precitech.