World Sweeper Logo

Street Sweeper Pick-up Performance


& ASSOCIATES, INC. (KAI) -- July 1995

For more information, contact:
Roger Sutherland (formerly of Kurahashi and Associates, Inc. (KAI), and now president of Pacific Water Resources.)
Pacific Water Resources
4905 SW Griffith Drive #200
Beaverton, OR 97005
Wk: 503-671-9709

As part 1 of KAI's work for the Port of Seattle on the stormwater quality analysis of the Sea-Tac International Airport, we conducted an "All Known and Reasonable Technologies" (AKART) search into stormwater quality management practices and passive stormwater treatment devices. The overall results of that search will be discussed in a separate memorandum to you on that more general topic. The purpose of this memorandum is to specifically address street sweeping as a stormwater quality management practice.

Previous Research

The Nationwide Urban Runoff Program (NURP) studies of street sweeping effects on stormwater quality published in 1983 (Reference 1) concluded that street sweeping proved to be largely ineffective in its ability to reduce the event mean concentration of pollutants found in urban runoff. This conclusion is largely based on the fact that the street sweepers used and tested were not able to effectively pick-up very fine accumulated sediments that have been found to be highly contaminated with most of the pollutants observed in urban runoff. The reason? Broom sweepers of this era were effective at picking up litter and large dirt particles, but harmful contaminants are concentrated primarily in the fines-the particles less than 63 microns. Not only were these fine particles left behind in the pavement after broom sweeping, but once the heavy covering of sediment was gone, the fines and their contaminants were even more likely to wash into storm drains during the next rain.

Therefore, the focus of this memorandum is to document any improvements in the newer street sweeping equipments ability to effectively pick-up accumulated sediments including fine sediments. It is also important that we compare the performance of any new equipment or operations to that of the NURP era sweepers to determine just how much change has actually occurred.

Promising Sweeping Technologies

Our search has resulted in three promising technologies that appear to provide significant improvements over the performance of the NURP era sweepers. The first technology is the use of a tandem sweeping operation. A tandem operation involves the combined use of a mechanical (i.e. broom and conveyor belt) sweeper followed immediately by a vacuum-assisted sweeper. The pick-up performance of a tandem operation using a Mobil broom sweeper followed by a TYMCO vacuum sweeper was monitored for over a year in a mediumdensity residential area located in Southeast Portland, Oregon. (Editor's Note: the term 'vacuum' used by the author is a generic one. Although the TYMCO sweeper used a vacuum-assisted sweeping process, their machine is actually a 'regenerative air-type' sweeper, not a 'vacuum-type' sweeper. Vacuum-type sweepers exhaust their debris-laden air into the atmosphere, rather than recirculating it like a regenerative-based sweeper.) The detailed description of this study and its results can be found in Reference Number 2. A brief summary of this monitoring effort and its results are provided in Reference Number 3 . The pick-up performance data obtained from the Portland tandem sweeping operation forms the basis of the comparison of this technology to others presented later in this memorandum.

The second technology is the stand alone use of a regenerative air sweeper. Regenerative air sweepers use air blown on to the pavement and immediately vacuumed into the machine to entrain and remove accumulated sediments. Regenerative air machines were in their infancy during the NURP era and to the author's knowledge were not extensively tested in any of the NURP sites. Regenerative air sweepers are generally considered to do a good job of removing fine sediment providing the accumulated loading are not too great. KAI measured the pick-up performance of the Port of Seattle's Elgin Crosswind regenerative air sweeper at Sea-Tac on April 21, 1995 . The results of these tests forms the basis of the comparison of this technology to others presented later in this memorandum.

The third technology is the stand alone use of a new highly effective vacuum-assisted sweeper called the Enviro Whirl I developed and manufactured by Enviro Whirl Technologies Inc., located in Centralia, Illinois. The Enviro Whirl I was developed from an earlier technology designed to vacuum and contain spilled coal dust along railroad tracks. As a result, the Enviro Whirl I appears to be extremely effective in picking up fine sediments and containing those sediments by filtering air emissions down to four microns which represents significant air quality benefits also. The rotating sweeper brooms located in the powerful vacuum head appear to have combined the benefits of a tandem sweeping operation into a single machine. In fact, as a direct result of the publication of the American Sweeper article (Reference Number 3) the author was contacted by both the manufactures and local distributor of the Enviro Whirl I. On April 24, 1995, at no expense to the Port of Seattle, KAI staff traveled to Las Vegas, Nevada (site of an air quality conference) to independently measure the pick-up performance of the Enviro Whirl I. The results of these tests forms the basis of the comparison of this technology to others presented later in this memorandum.

It should be noted that as a result of the pervious Portland study mentioned earlier, we were able to document the pic-up performance of a newer mechanical (i.e. broom) sweeper which was a 1988 Mobil. These results are also shown for comparison purposes.

The pick-up performance for the NURP era sweepers was based on the author's previous analysis (Reference Number 4) of the Bellevue, Washington NURP data (Reference Number 5). The author was a consultant to the City of Bellevue during the NURP study and has directed access to the street sweeper pick-up performance data collected as part of that historic study. The standard mechanical street sweeper tested in Bellevue was a Mobil probably manufactured around 1978.

Analysis Procedure

The street sweeper's ability to significantly interact with the accumulation and washoff of contaminated sediments readily available on directly connected impervious surfaces like streets actually determines the overall effectiveness of a street sweeping operation evaluated over a designated period of time. The Simplified Particulate Transport Model (SIMPTM) can accurately simulate this complicated interaction of accumulation, washoff, and street sweeper pick-up over a period of time (Reference Number 6). In fact, SIMPTM is being used to characterize the annual storm water quality loadings from Sea-Tac International Airport and the pollutant reduction effectiveness of using the Elgin Crosswind regenerative air sweeper or the Enviro Whirl I sweeper (Reference Number 7). However, what is of interest in this memorandum is: 1 ) how SIMPTM model's street sweeper pick-up performance; 2) how that model compares to real pick-up performance data; 3) how the calibrated model parameters vary for the various technologies described in this memorandum, and; 4) the estimated pick-up effectiveness of each of these technologies for several hypothetical initial loading conditions. The last item will form the most useful basis of comparison between NURP era sweepers, the newer mechanical sweeper, and the three promising sweeping technologies described

Pick-Up Performance Model

The street sweeping component of the SIMPTM model was based on the results of Pitt's street sweeping study conducted in San Jose, California and published by USEPA in 1979 (Reference Number 8). This model was confirmed in additional studies conducted by Pitt and Shawley (Reference Number 9) and Pitt and Sutherland (ReferenceNumber 10).

Figure I illustrates the street cleaning component and equations used by SIMPTM. For each size group, the amount removed (Prem) is proportional to the accumulation (P) in excess of the base residual (SSmin) by a sweeping efficiency (SSefi):

Prl, = SS, ( P- SSrl.r.) for P > SSrr

The above-mentioned studies found that street sweeping removes little, if any, material below a certain base residual which was found to vary by particle size. Above that base residual, the street sweepers removal effectiveness is described as a straight line percentage which was also found to vary by particle size. Therefore, to describe a unique street sweeping operation one simply needs to know the operations SSmin and SSeff values for each of the eight particle size ranges simulated by SIMPTM.

Figure 2 shows an example of how the simple model component actually compares to real pick-up performance data. The plotted points are the data obtained from the monitoring of the tandem street sweeping operation on Portland's Sellwood drainage basin. Please note that the colTelation coefficient squared for the eight particle size fits ranged from 94.3% to 99.9% which means the model is doing an excellent job of reproducing the actual observations. These high explained variations were typical of all of the model fits to the pick-up data from the various sweeping technologies.

Tables I and 2 present the calibrated model parameters SSmin and SSeff, respectively, for each of the five sweeping technologies presented in this memorandum. In Table 1, note the dramatic improvements in reducing residual loadings for all the newer technologies when compared to the NURP. Both tandem sweeping and the Elgin Crosswind regenerative air are very impressive, but the across-the-board zero residual loadings for the Enviro Whirl I is hard to believe because it is perfect.

Table 2 also shows some impressive removal efficiencies above the residential loadings. Once again, please note the dramatic changes from the NURP era sweepers performance. The effectiveness of the Elgin Crosswind (regenerative air) and the Enviro Whirl I for the finer particle size groups may not look that impressive in the table. However, remember that these two machines are operating on all initial loadings for Group Number 1 and 2, and the Enviro Whirl is operating on all loadings for all groups. In fact, if the SSeff values for the Enviro Whirl were 1.0, the street sweeper performance would be perfect or 100% of everything available would have been picked up.

Pick-Up Performance Comparison

Working with the average particle size distribution observed in the fifteen street dirt accumulation samples collected at sites throughout the Sea-Tac International Airport from September 30, 1994 through April 21, 1995 (i.e. see Table 3) and assumed initial loadings, the projected street sweeper pick-up efficiencies are presented in Table 4. The assumed initial loadings represent the entire range of conditions that could be reasonably observed at Sea-Tac or throughout the Seattle area. The average accumulation value observed at Sea-Tac was 200 Ibs/paved acre with an observed range of 8 to 1,130 Ibs/paved acre. The maximum accumulation observed during the Bellevue NURP was approximately 500 Ibs/paved acre. The average Bellevue accumulation was approximately 250 Ibs/paved acre. The 1,000 Ibs/paved acre would generally represent a site heavily influenced by erosion from construction, an area of poor pavement condition, or an area adjacent to a source of erodible sediment.


Table 4 clearly shows that all of the newer street sweeping technologies are significantly more effective than the NURP era sweepers. So the findings of the NURP in regards to street sweeping lack of effectiveness may not be valid today. Also, note that in lower loading conditions (i.e. 10 to 100 Ibs/paved acre) which are common at Sea-Tac and throughout the Seattle and Portland metropolitan areas, the Enviro Whirl I sweeper pick-up is extremely effective especially in less than 63 microns . The tandem sweeping operation becomes competitive with and appears to surpass the effectiveness of the Enviro Whirl I at higher initial loadings. Note that the regenerative air machine is also quite effective in total removal effectiveness at these higher loadings but it's effectiveness in removing the fine sediment lags behind the other two promising technologies.

Two important items should be noted The first is that this comparison in Table 4 is based on an assumed particle size distribution which is not very fine but somewhat coarse. The Enviro Whirl I would be much more effective at higher initial loadings if the initial particle size distribution were finer. The second is that the Enviro Whirl manufacturer informed us following our Las Vegas testing that the sweeper was not operating at maximum effciency because there was too much air in the tires and the vacuum was losing an inch of suction all around the head. In addition, a portion of one of the rotating broom's bristles were found later to be missing. As a result the Enviro Whirl sweeper will be visiting the Northwest in September of 1995 when further performance tests will be conducted. A demonstration of this impressive machine's sweeping abilities is being scheduled with the Port of Seattle and other Seattle governmental agencies.


1) U.S. Environmental Protection Agency, Water Planning Division, Results of the Nationwide Urban RunoffProgram, Volume I - Final Report, December, 1983.

2) HDR, Inc., Combined Sewer Overflow SFO Compliance Interim Control Measures Studv. Final Report, prepared for City of Portland, Bureau of Environmental Services, May 1993.

3) Alter W., "The Changing Emphasis of Municipal Sweeping...May Be Tandem," American Sweeper, Volume 4, Number 1, February, 1995.

4) Sutherland R.C., Water Oualitv Related Benefits to the Citv's Current Street Cleaning Program - Phase 2 Results, letter to Ms. Lori Faha, City of Portland Bureau of Environmental Services, February 1,1990.

5) Pitt, R.E., Characterization. Sources. and Control of Urban Runoffbv Street and Sewerage Cleaning, Contract Number R-80597012, U.S. Environmental Protection Agency, Of fices of Research and Development, Cincinnati, Ohio, 1985.

6) Sutherland R.C., and S.L. Jelen, Simplified Particulate Transportation Model - Users Manual, Version 3.1, December, 1993.

7) Kurahashi and Associates Inc., Sea-Tac International Airport Stormwater Oualitv Characterization, memorandum to HDR Engineering Inc., August 1995.

8) Pitt, R.E., Demonstration of Nonpoint Pollution Abatement Through Improved Street Cleaning Practices, EPA 600/2-79-161, August 1979.

9) Pitt, R.E., and G. Shawley, A Demonstration of Nonpoint Pollution Management on Castro Vallev Creek, Alameda County Flood Control and Water Conservation District, Hayward, California, 1982.

10) Pitt, R.E., and R.C. Sutherland, Washoe Countv Urban Stormwater Management Program - Volume II Street Particulate Data Collection and Analvsis, Prepared by CH2M Hill for Washoe Council of Governments, Reno, Nevada, November, 1982.

World Sweeper Logo

© 2005 - 2019 World Sweeper
All rights reserved.

Street Contents

Street/Studies Contents

Site Map / Table of Contents