Published February 2008

Optimum Cleaning

By Doug Wright, John Wolgamott and Gerald Zink (page 78)

A waterblasting equipment maker looks at how standoff distance, traverse speed, rotation speed and other factors affect surface preparation.

High-pressure water is a powerful tool in preparing surfaces for painting or finishing. But what determines the most efficient way to strip off old material and make way for new?

Our team looked at how variables such as standoff distance, traverse speed, surface speed, rotation speed, and head design affect surface preparation when using waterblasting tools at 20,000 to 40,000 psi.

A range of equipment

The most common materials removed in surface preparation are coatings, oxidation and scale. A wide variety of equipment is available for waterjet surface preparation, but most such equipment now includes rotating nozzle heads. Handheld rotating guns are the most basic tools, but also common are self-propelled machines that clean much wider paths at controlled feed rates.

Previous research has focused on massive material removal at lower pressures and higher flow rates, but questions have arisen about whether these results apply at higher pressures and to the removal of thin coatings.

We performed our tests using an air-powered rotating assembly with various heads. This was attached to a traversing mechanism with adjustable feed rates (Figure 1). We tested samples placed underneath at an angle to vary the standoff distance from 0.37 to 3.5 inches.

For test samples, we used commercial-grade, coated-steel-siding trim sections. We rated effectiveness by visually estimating the percentage removal of the top coating and the primer. All tests were conducted at 35,000 psi and 3 to 6 gpm. Jet path diameters of 3.6, 8.4, 12, and 14.4 inches were used, with rotation speeds from 500 to 3,200 rpm.

Evaluating performance

Besides using different path diameters, we compared two general head types: a one-piece bar (Figure 2) and a multiple piece with individual bent jet arms (Figure 3). The bar heads had each jet port spaced 0.37 inches apart. The diameters given above are the largest of the jet paths. The majority of the tests used four 0.02-inch or four 0.015-inch-diameter sapphire orifices. Jet angles exiting the bar head and the quantity of jets were also compared.

Standoff distance. The majority of the tests were conducted with the test sample placed at an angle to produce a varying standoff distance. The main observation was the ineffectiveness of coating removal when the standoff distance was too small — every test showed this to some degree. This region of ineffectiveness is attributed to the jet still being coherent, and not yet having broken into droplets.

The other measurable effect of standoff distance was related to rotation speed. Figure 4 shows the effect of standoff distance relative to multiples of orifice diameter at two rotation speeds. The maximum “too close” range varied from 0.5 to 0.63 inches with the bar head but increased to as much as 1 inch with the bent-arm head.

In terms of multiples of nozzle diameters, this range varied from 18 to 42 times the orifice diameter for the bar heads and up to 67 times the orifice diameter for the bent arms. The most effective removal with the bar head occurred beyond 65 to 95 times the orifice diameter; in the fastest rotation speed tests, the jet effectiveness showed rapid deterioration beyond 150 to 160 times the orifice diameter.

Slower rotation speeds allowed effective removal out to 230 times the orifice diameter, which was the greatest standoff distance tested. No deterioration at the maximum distance was observed using the bent arms at the fastest rotation speed tested.

Rotation speed. The purpose of these tests was to determine if there is a rotation speed where jets begin to lose power. The diameter of the jet path affects the velocity at which the nozzle tip is moving.

For example, a jet path diameter of 3.6 inches rotating at 3,000 rpm results in a velocity of 47 ft/sec, while a jet path diameter of 14.4 inches rotating at 1,800 rpm results in a velocity of 113 ft/sec. One would therefore expect that as the jet path diameter increases, rotation speed should be slowed to maintain an effective velocity.

Four parameters varied the effect of rotation speed on performance: standoff distance, orifice diameter, feed rate, and head design. Increasing rotation speed with increasing standoff distance narrowed the effective standoff distance range.

Jet performance with increasing rotation speed deteriorated faster with a smaller orifice diameter. The effects of rotation speed made a slight difference as the linear feed rate increased.

Rotation speed was tested at three jet path diameters. The 14.4-inch bar head was tested at 500, 1,000, and 1,800 rpm; the 8.4-inch head at 1,000, 1,500, and 2,000 rpm, and the 3.6-inch head at 2,000 and 3,200 rpm. The feed rates were adjusted to produce the same rate of coverage. For the head diameters and speeds tested, the results fell on approximately the same curve, and deterioration in jet power began at a velocity greater than 66 ft/sec.

Feed rate. These tests showed the feed rate to have the greatest effect of all the parameters on percentage of coating removal. The 14.4-inch bar head was tested with rotation speeds of 500, 1,000 and 1,800 rpm at feed rates of 20, 30, and 40 in/min. The optimum efficiency occurred at 1,000 rpm and 30 in/min.

The 8.4-inch bar head was tested at 1,000, 1,500, and 2,000 rpm at feed rates of 40 and 60 in/min. The best performance was achieved at 2,000 rpm and 60 in/min.

The 3.6-inch bar head was tested at 2,000 and 3,000 rpm at 80 and 120 in/min, and the best performance occurred at 2,000 rpm and 80 in/min. The efficiency relative to feed rate is shown in Figure 5; the curve for the larger bar head is much more sensitive than the smallest diameter head.

Jet path diameter. The efficiency of jet path diameter based on the bar heads tested is shown in Figure 6. For the three to be equal in efficiency, the 3.6-inch-diameter head had to have an effective feed rate three times that of the 14.4-inch head. The greatest efficiency appeared with the 8.4-inch head; it produced a cleaner pass than either of the two other heads at a feed rate twice that of the 14.4-inch head.

Referring back to the curves in Figure 5, the sensitivity of the larger head diameter curve may be another contributing factor in head diameter selection. The efficiencies of these heads are not too far apart, but if the trend continues beyond the diameters tested, one would expect a further loss of efficiency.

Jet angle. A 14.4-inch bar head with 5-degree outward angled ports was compared to a similar head with straight downward facing ports. The angled ports removed an estimated 15 to 20 percent more coating than the straight ports.

Bent arm head design. The bent-arm head was tested and compared to the bar head design. The greatest effect occurred with standoff distance. The bent-arm head showed less deterioration due to rotation speed, and the efficiency of coating removal compared to the bar head improved by 25 to 30 percent.

Nearly the same removal was achieved with the 0.015-inch orifice size in the bent arms as that achieved by the 0.020-inch orifice size in the bar head. Further testing would be required to determine if this improved efficiency of the bent-arm design could translate directly into increased production.

Conclusions

Overall, the parameter having the greatest effect on performance was the feed rate, which also directly affects efficiency. The next strongest parameter was the head design; the bent-arm head performance was 25 percent better than the bar head design, and jet angle improved performance by 15 percent.

Jet path diameter appeared to reach an optimum around the 8.3-inch size range, although this was not a strong influence. Rotation speed affected performance in several ways, but was not shown to be very influential.

These tests showed that increasing rotation speed is not necessarily a direct path to a faster feed rate. It should be kept within a range to produce a velocity between 33 and 82 ft/sec for optimum performance.

Doug Wright is an engineer, John Wolgamott is president, and Gerald Zink is vice president and chief engineer with StoneAge Inc. in Durango, Colo. This article is based on a paper they presented at the 2007 WaterJet Technology Association (WJTA) American Waterjet Conference in Houston, Texas. They can be reached at 970/259-2869.