Manual Grinding and Polishing

Grinding. Rough grinding of the mount must produce a planar surface for subsequent grinding and polishing. The preferred procedure involves using a water-cooled and lubricated wheel with 240, 400, and 600 mesh SiC papers. The paper is held to a rotating disk that makes use of the vacuum created by a thin layer of water under the sheet of abrasive paper. Alternatively, the paper can be held in place adhesively or by special flat magnets that grip a thin ferrous backing on the grinding paper. The same grinding may be carried out on wet papers placed on top of a sheet of glass. When using grinding wheels, the specimen is held in a fixed position on the wheel so that all scratches are in one direction, which requires even and moderate pressure. When changing papers, the specimen is rotated 90° to allow one to note the disappearance of the previous scratches. The use of one single sheet of SiC paper for more than two or three specimen mounts is not recommended because it leads to lack of flatness of the specimen surface. Grinding using 240 to 600 grit abrasives and moderate pressure at 125 to 250 rpm on 200 mm (8 in.) laps requires approximately 30 s for each paper. Using Al2O3 for edge retention (Fig. 3) gives a hard mount surface that requires longer grinding and polishing times.

It is important to grind a 30° bevel around the mount periphery. This allows the mount to pass smoothly over the subsequent polishing cloths and prevents the plowing aside of abrasives by a sharp edge. Failure to use a bevel slows polishing on long-nap cloths.

Polishing. Following grinding, the specimen is flat to the edges, and the pores are almost completely smeared (Fig. 4). Subsequent polishing generally rounds the specimen edges, because the mounting resin is much softer than the metal specimen. This rounding can be prevented by the use of embedded ceramic materials for edge retention, or such conventional techniques as plating of the specimen surface before mounting. Polishing must open all the pores, show true area fraction of porosity, remove scratches and disturbed metal, and minimize edge rounding. The presence of epoxy resin or wax in the pores facilitates opening the smeared pores, but does not eliminate the problem.

Fig. 4 Pressed and sintered Fe-0.8C alloy (6.8 g/cm3), as-ground on 600 grit silicon carbide. Micrograph shows the closure of pores and flatness of specimen (the surface is shown at left). Arrows indicate closed pore edges. 95x

During polishing, the abrasives first open the pores closest to the specimen edges (Ref 5, 6), giving the impression that surfaces are less dense than interiors (Fig. 5, 6, 7). For ferrous materials, the fastest way to reveal the pores results in slight edge rounding but is adequate for routine work. These steps should be followed:

1. Etch 2 min in 2% nital by immersion. For materials other than low-alloy steels, use the customary agents. Etching before polishing initiates pore opening but does not exaggerate pore size.

2. Rough polish 2 min using 1 /'m AI2O3 moderate hand pressure on a long-nap felt cloth (Ref 7). Use 250 rpm on a 200 mm (8 in.) diam wheel rotating the specimen counter to the wheel to prevent comet tails. The long-nap cloth and the fairly coarse Al2O3 rapidly open the pores (see Ref 8 for an example of another technique that requires 300 min).

3. Repeat steps 1 and 2 once or twice. This procedure generally opens all the pores. To the unaided eye, the surface of the specimen should exhibit a uniform orange-peel appearance with no shiny specular (mirrorlike) regions. If necessary, repeat steps 1 and 2 until the surface is uniformly roughened. Even P/M forgings and metal-injection-molded parts at 98 to 99% of full density display porosity to the unaided eye.

4. This aggressive rough polishing exaggerates the pore area fraction by eroding and rounding pore edges. That is, the specimen will appear erroneously low in density. Final polishing must restore the true area fraction of porosity.

5. Polish 2 min using 1 /'m diamond on a short-nap cloth at 250 rpm with moderate hand pressure. This sharpens pore edges and restores the pores to their true area fraction and eliminates most scratches, but

leaves the edges of the specimen slightly rounded. A 19 mm (4 in.) long bead of diamond paste, weighing approximately 0.06 g (0.002 oz), is recommended for each 2 min of polishing. Use an alcohol-base solvent or thinner for the diamond paste so that it will wash off in water. Oily thinners penetrate the residual pores and bleed out of the specimen.

6. Final polish 30 s using a long-nap microcloth and 0.05 f'm deagglomerated AI2O3. Use light hand pressure or an automatic polisher with 100 g weight (1 N, or 4.5 lbf) on the specimen at 125 rpm. This removes the fine scratches on most ferrous materials. The true area fraction of porosity of the surface is now restored. Reference 6 includes information on this method, and it also demonstrates that it is possible to open pores and show the correct porosity area fraction using a method that requires approximately 12 min and does not use diamond. It consists of 10 min of hand polishing using 1 /'m ai2o3 on a synthetic suede, short-nap cloth at 250 rpm on a 200 mm (8 in.) lap and 2 min of light hand polishing using 0.05 /Jm ai2o3 at 125 rpm on the same type of cloth.

Fig. 5 Effect of polishing on pore opening in a pressed and sintered Fe-0.8C alloy: deliberately under-polished specimen. This region, which is adjacent to the specimen edge, shows all the pores open. Compare with Fig. 6 (center of specimen). 180x

Fig. 6 Effect of polishing on pore opening in a pressed and sintered Fe-0.8C alloy: interior of same specimen as in Fig. 5. After 2 min of polishing, there are numerous smeared pores. Compare the amount of porosity with Fig. 5. This micrograph shows how the inner part of a specimen polishes more slowly than the edge. 180x

Fig. 7 Effect of polishing on pore opening in a pressed and sintered Fe-0.8C alloy: repolished version of Fig. 6 showing more pores in the center of the part (some remain smeared over). The density appears higher than the true density of 6.8 g/cm3. 180x

To produce a surface with no edge rounding, it is necessary to eliminate the 1 /'m AhO, and long-nap cloth polishing. Instead, after the 2 min etch in 2% nital, step 5 should be repeated several times. More than five repetitions may be required to open all the pores on a large specimen, particularly if it is soft and undersintered. The wheel should be recharged with diamond at each repetition. However, the 2 min etching should not be repeated, because the 1 /'m diamond paste does not rapidly remove etching effects.

Newly developed P/M materials require modified polishing procedures to show the correct area fraction of porosity. Such methods could be developed using standards of known density. Other modern methods of automatic polishing may open all pores, but should first be tested by preparing a specimen of known density, then measuring the area fraction of pores. The measured area fraction should agree with calculations made from the known density.

A vibratory machine in which the specimens circulate around a bowl with the abrasive slurry may also be used. The use of a short-nap chemotextile cloth (Texmet) and 0.3 or 1 /'m AhO;, will yield a specimen that is virtually free of edge rounding. However, for specimens that have been ground through 600 grit, this procedure requires approximately 3 h because of the slow material removal rate and the need to open all the pores.

The rate of material removal may be measured using a Knoop indenter mark as a reference (Fig. 8). First, a mark approximately 100 /'m long is made in a known location on the specimen. A simple reference point in the interior of the specimen can be made using a Rockwell superficial indenter with the 15 kgf load. The Knoop mark is then placed approximately 0.4 mm (0.015 in.) away from the superficial indenter mark and at a known orientation to it. The Knoop mark is measured from a photograph or with the measuring stage of the microhardness tester. After polishing for a fixed time, such as 1 to 2 min, the Knoop mark is relocated and remeasured. The material removed normal to the specimen surface is the change in length of the Knoop diagonal divided by 30.51 (for a standard indenter). For a 25 by 25 mm (1 by 1 in.) specimen, polishing using a 250 rpm, 200 mm (8 in.) diam lap, 1 /'m AhO;, on a synthetic suede, short-nap cloth, and moderate hand pressure will remove 0.4 /'m/min. A smaller specimen, such as 12 by 6 mm (0.5 by 0.25 in.) will polish at 1.45 /'m/min. Additional information on material removal rates can be found in Ref 5.

Fig. 8 Knoop indenter mark (100 gf) used as a reference to note the rate of material removal from the surface by measuring the change in length and depth of the indentation. Surrounding black pores in this unetched, pressed and sintered Fe-0.8C alloy (6.8 g/cm3) are also revealed. 295x

Soft material, such as pure iron or copper, may still exhibit some fine scratches after the 0.05 /'m AhO;, polishing described above in step 6. One solution is to use a new long-nap cloth (microcloth) with adhesive backing attached to a flat glass plate or to a flat bench top. With the deagglomerated 0.05 /'m AhO;,. charged onto the cloth at a ratio of 1 part (by volume) powder to 4 parts distilled water, the specimen should be polished in the abrasive slurry using approximately 50 light hand strokes straight back and forth. This will eliminate the fine scratches from the prior polishing; remaining scratches will be aligned

parallel to the direction of polishing, and their source identified. This light final polishing does not cause comet tails or pore beveling. Colloidal silica, for example, Mastermet (Buehler, Ltd., Lake Bluff, IL), is excellent for a final polish.

Titanium alloys require a 4 min rough polishing using 1 /'m AfO;, on felt cloth at 250 rpm on a 200 mm (8 in.) diam lap with moderate hand pressure. This opens and slightly enlarges the pores. The autopolisher with 3 /'m diamond on DAC cloth for 10 min at 30 N can be used. Final polish is colloidal silica on microcloth, at 1 N (4.5 lbf), 125 rpm, and for 1 min.

0 0

Post a comment