Download this article in .PDF format
This file type includes high resolution graphics and schematics when applicapable.

Four plating processes were investigated, all electroless—i.e., chemical reduction. These were: Enthone, Shipley, Dynachem, and MacDermid. Only two were evaluated—Dynachem and Shipley. Considerable work with the Enthone process has been performed and reported by Harry Diamond Laboratories; thus, no further work was done on this.

All plating of test pieces was done using either Dynachem or Shipley processes and in accordance with instructions pertaining to each. In the final copper plating, a 140°F Pyrophosphate bath was used with a current density of 10-15 A/ft2. A leveling agent of Unichrome Additive Agent PC-1 was added to the bath in the proportions of one quart per 100 gallons of plating solution. This additive produces a very fine surface deposit.

The required copper deposit is dependent upon conduction current, or skin depth (δ), and is determined by:

where,

λo = free space wavelength;

ρ = the conductor resistivity; and

μ = the conductor permeability.

No difficulties occurred with copper-plating adhesion to the substrate. It was concluded that good adhesion is mainly due to the mechanical locking to the pores of the machined substrate. Adhesion of molded substrates would seem to depend on interface adhesion.

A most important factor in successful plating of the raw substrate is rack design. Racking is necessary to overcome the buoyancy of the lightweight substrate. Two rack types are required. One is for the cleansing and intermediate electroless process. The other is for the final copper plating process. Both types should be heavy enough to overcome the buoyancy of the substrate and provide good clamping. In addition, the rack for the final plating process should provide good electrical contact with the substrate through relatively large contact area.

The racks should be so designed that when fitted to the substrate, forces imposed will not cause the substrate to bow, warp, or twist.

Preparing the raw substrate

Two methods for preparing the raw substrate prior to plating are machining and casting. The machining method was chosen as most convenient in this study.

To determine surface finishes obtainable, samples were prepared using three separate tools—grinding wheel, end mill, and fly cutter. The surface finishes obtained are compared in Table 1.

Download this article in .PDF format
This file type includes high resolution graphics and schematics when applicapable.

Grinding: In the grinding operation the substrate surface was obtained with a Norton Grinding Wheel, type 32A60K5VBE. The wheel was tapered in 5°, to give a sharp cutting edge. Such a wheel may be considered to have a large number of cutting edges, thus tending to burnish as well as cut. Water was used as a coolant. Optimum conditions for grinding were obtained as follows:

  • Wheel grain size 60, grain spacing 5. Finer surface finishes are not obtainable by increasing the grain size; if grain sizes from 100 to 600 were used, the wheel would “clog” between grains causing burring of the substrate.
  • Optimum grinding speed 2700 rpm; feed rate 3 in./min.

End mill. For end milling, surface finish was obtained with a twelve cutting-edge end-mill cutter, which tended to burnish as well as cut. However, this process, though faster than gridning, does not produce as fine a surface finish. No cooling is required, but vacuum pickup of waste particles is necessary to keep them from impregnating the substrate.

The method of sharpening the end mill is important; i.e., the base should be sharpened before the ribs. This eliminates any burrs on the cutting edges that might score the substrate.

Fly cutter. A fly cutter has only one cutting edge, so there is no burnishing effect. Consequently a relatively coarse surface finish results. The method is quick and requires no coolant. However, the resulting surface finish proved too rough for waveguide applications.