Cyclone Sighting

Summer 2016

Can One Machine Clean Up Washington, DC?

by David R. Smith and Don Rissmeyer, PE, Associate, A. Morton Thomas & Associates


Cyclone Sighting

Since the 1990s, municipalities and private property owners constructed millions of square feet of pervious concrete (PC), porous asphalt (PA) and permeable interlocking concrete pavements (PICP) in parking lots, alleys and streets. The number one question is about maintenance. The next questions typically are how often should the surface be cleaned and with what equipment?

All permeable pavements require regular surface cleaning to remove embedded sediment and to maintain surface infiltration. Regenerative air machines used for routine cleaning are effective in removing loose sediment and debris. Low surface infiltration into highly clogged pavements with tracked-in or settled sediments can be raised with a true vacuum machine. Equipment availability, costs, personnel time or outsourcing costs for surface cleaning suggests a need for a single machine that provides routine maintenance cleaning, as well as restoration of clogged surfaces when maintenance is neglected.

The diesel-powered Cyclone CY5500 carries approximately 1,200 liters (300 gal) of water.

The diesel-powered Cyclone CY5500 carries approximately 1,200 liters (300 gal) of water.

A machine that might qualify for this role is the Cyclone CY5500. Originally developed to clean tire rubber from runways, this machine was tried in June 2015 on porous asphalt (PA), pervious concrete (PC), and permeable interlocking concrete pavement (PICP) in a residential neighborhood in Northwest Washington, DC. All of the permeable pavements were installed by the District of Columbia Department of Transportation (DDOT) as part of a combined sewer overflow mitigation program.

The sites included PC in two nearby on-street parking lanes. One was cast-in-place PC and the other was a precast PC panel, among several. These two areas received contributing run-on from the impervious center lane of the street. The PA and PICP were situated in alleys, with some or little run-on from impervious surfaces and instead received sediment from adjacent vegetated areas. All of the permeable pavements were subject to leaves and debris from a mature urban forest canopy. None of the pavements were older than a year in service.


The diesel-powered Cyclone CY5500 is an off-road vehicle smaller than the truck-size equipment. This machine carries approximately 1,200 liters (300 gal) of water, much of which is drawn back into the machine, filtered and re-used.

The Cyclone machine relies on water applied under pressure in a circular motion within a surrounding chamber in close contact with the permeable pavement surface. The water pressure can be varied by the operator from 1,200 psi (8 MPa) to 4,350 psi (30 MPa). Water is blasted against the pavement surface and the speed of the rotating head applying the water provides some suction (hence the cyclone name) to pull most of the water back into the machine for reuse. The machine manufacturer claims cleaning rates as high as approximately 10,000 sf (935 m²) per hour on most permeable pavements.

Prior to conducting cleaning, ASTM C1701 Standard Test Method for Infiltration Rate of In Place Pervious Concrete and C1781 Standard Test Method for Surface Infiltration Rate of Permeable Unit Pavement Systems was applied to each surface. The former test method is applicable to PC (and PA). Both test methods produce comparable results. This is illustrated in Figures 2 through 5.

The pre-wetting initial infiltration test measurement was conducted to determine the extent of clogging. All of the pavements were clogged with little or no infiltration within the ring. Then, the four areas were cleaned in the following order: cast-in-place PC, PICP, PA, then the precast PC panel. The Cyclone machine passed twice over the same area of permeable pavement almost immediately after the initial infiltration testing (called pre-wetting). Figure 6 shows the Cyclone machine making a typical pass of about 30 ft (10 m) in length on an alley.

Figure 6

Figure 6: The Cyclone machine cleans a PICP alley.

After the second pass, the ASTM test ring was applied to the pavement surface and an additional (approximate) 8 lbs (5 kg) of water was applied (the surface infiltration rate was calculated per the ASTM standards). Both ASTM standards use the same surface infiltration calculation. If the surface infiltration rate was under 250 mm/hr (100 in./hour) the Cyclone machine made an additional two passes, the ring reapplied in the same location and an additional approximate 8 lbs (5 kg) of water applied into the ring. The table provides a summary of the surface infiltration test results.

The table indicates increased infiltration rates after the first two passes on the PC, PICP and PA. Infiltration rates doubled for the PC and PA but could be considered low. There was little if any change in the infiltration rate of the precast PC panel after the first two passes. The second two passes yielded better results with PC infiltration rate doubling again and the PA almost reaching the same level. The precast panel saw a substantial increase as well, from <20 in./hr (<508 mm/hr) to 130 in./hr (3,302 mm/hr) after the second round of two passes.

Figure 7

Figure 7: PICP after two passes of the Cyclone machine. Note aggregate and sediment removed resulting in open joints and a permeable surface.

The most notable observation is that the PICP only required two passes of the Cyclone machine rather than four to increase the infiltration rate from 20 in./hr (<508 mm/hr) to 327 in./hr (8,306 mm/hr). The joint widths in the PICP were narrow, approximately ¼ in. (6 mm) wide and many of the small aggregates were pulled out with the sediment after the first two passes. Some of the stones were left on the surface of the pavers after the second pass and these could be swept back into the joints. See Figure 7. Additional aggregate should be supplied given these results.

In this experiment, the Cyclone machine was assigned to clean highly clogged pavements. It can be set on a lower pressure setting to clean a less clogged condition, i.e., remove loose material from the pavement surface. In the case of this brief demonstration however, the PICP surrendering sediment with the jointing aggregate to the Cyclone machine explains the resulting high infiltration rate after two passes rather than four passes, as conducted on the other surfaces. This demonstrates the ability of clogged PICP to experience restored infiltration rates as compared to monolithic surfaces, even when heavily clogged.