UPDATES: December 2011

UPDATES is a monthly email newsletter on stormwater management, assessment (including monitoring), and maintenance research at St. Anthony Falls Laboratory and the University of Minnesota.

Click HERE to subscribe to UPDATES Newsletter

Or Text SWUPDATES to 22828 to Join. Message and data rates may apply.

For Email Marketing you can trust


Cleaning Stormwater with the SAFL Baffle

December 2011 (volume 6 - issue 11)

Contributed by Kurtis D. McIntire, Graduate Research Assistant, St. Anthony Falls Laboratory, University of Minnesota (Advisors: Omid Mohseni and John S. Gulliver)

Funded by the Minnesota Department of Transportation

Sump manholes are common structures used by municipalities as access points to storm sewers. But as anyone who has opened a manhole lid to a sump knows, sediment and debris collects in these structures. Engineers at the Minnesota Department of Transportation asked, "How effective are standard sump manholes as stormwater pre-treatment devices, and can we make them more effective?” To answer this question, SAFL researchers showed that a sump equipped with a device called the SAFL Baffle can be an effective stormwater best management practice (BMP).

In this edition of UPDATES, new SAFL research shows how a sump equipped with a SAFL Baffle will perform when:

  1. Clogged with debris like trash and vegetation
  2. Installed in a sump with an outlet pipe 90 degrees to the inlet pipe
  3. Installed in a sump with water entering through an inlet grate and inlet pipe

Figure 1: A SAFL Baffle installation in St. Paul

At St. Anthony Falls Laboratory, we performed tests on a 6-ft diameter by 6-ft deep sump, a 6-ft diameter by 3-ft deep sump, a 1-ft diameter by 1-ft deep scaled model sump, and a 1-ft diameter by 0.5-ft deep scale model sump. All of the sumps were equipped with a SAFL Baffle, and we were interested in two performance criteria:

  1. How well the system captures sediment at low flow rates (Removal Efficiency Testing)
  2. How well the system holds onto that captured sediment at high flow rates (Washout Testing)

To test the SAFL Baffle with debris, we loaded the system with combinations of leaves, plastic bags, plastic bottles, and plastic leaves, and conducted our Removal Efficiency and Washout tests. For these tests, we used two variations of the SAFL Baffle - one with 3 inch diameter holes, and one with 5 inch diameter holes.

Figure 2: A schematic of two standard sump manholes retrofitted with SAFL Baffles. Left - A 6-ft diameter and 3-ft deep "shallow" sump; Right - A 6-ft diameter and 6-ft deep "deep" sump.

We learned that a clogged SAFL Baffle in a “deep” sump whose depth, including sediment, is equal to its diameter will capture sediment just as well as an unclogged SAFL Baffle. Additionally, a clogged SAFL Baffle in a deep sump will not experience significant washout at high flow rates. This means that the SAFL Baffle performs well under clogging conditions in a deep sump.

For a clogged SAFL Baffle in a shallow sump whose depth is less than one half its diameter, we noticed that water hit the clogged SAFL Baffle and traveled underneath of it. When the water traveled underneath of the SAFL Baffle, it caused washout at high flow rates. Conversely when a 5 inch hole diameter SAFL Baffle was installed in a shallow sump and exposed to the same debris loading and high flow rates, clogging and washout was sharply decreased. A SAFL Baffle in a shallow sump will perform well if it has 5-inch diameter holes, but may have washout at high flow rates with 3-inch holes.


Figure 3: Top - Debris testing on a shallow sump scale model equipped with a 3 inch hole diameter SAFL Baffle. Bottom - Debris testing on a shallow sump scale model equipped with a 5 inch hole diameter SAFL Baffle. Notice the difference in sediment washout.

Figure  4: The 5 inch SAFL Baffle helps mitigate washout in clogged baffles. This 5 inch variation is only necessary if debris loading in your watersheds is common and if you are installing in a shallow sump.

The next set of tests were completed on a sump with an outlet pipe 90 degrees to the inlet pipe. We learned that the amount of washout at high flow rates strongly depends on the installation angle of the SAFL Baffle relative to the inlet pipe. When the SAFL Baffle was installed at a 90 degree angle to the inlet pipe, washout was negligible. But as the SAFL Baffle installation angle was rotated to 180 degrees, washout increased. We found that washout remained negligible when the SAFL Baffle was installed at angles between 90 and 120 degrees to the inlet pipe, giving installers flexibility.

Table 1: 1:6 scale model of a 6-ft diameter, 6-ft deep sump with an outlet 90 degrees to the inlet pipe. All tests were conducted at flow rates around 3 L/s (11 cfs when scaled to a 6-ft diameter, 6-ft deep device).

Test SAFL Baffle Angle Effluent Concentration (mg/L)
1 No SAFL Baffle 1035
2 90 18
3 112.5 19
4 135 46
5 157.5 257
6 180 664

Figure  5: A SAFL Baffle installed at a 113 degree angle with respect to the inlet pipe.

For the final testing series, a platform was built over the laboratory 6-ft diameter by 6-ft deep sump manhole. This platform held a simulated road surface, curb, gutter, and inlet grate. The inlet grate was positioned such that it was in line with the SAFL Baffle. 50% of the inlet grate was upstream of the SAFL Baffle and 50% was positioned downstream. Water could be sent into the sump through both this inlet grate and the inlet pipe.

Figure  6: Stormwater entering a sump manhole through an inlet pipe and an inlet grate

Figure 7: The laboratory testing stand used to send water into the sump through an inlet pipe and an inlet grate.

During Washout Testing, we sent the maximum 0.7 cfs through the inlet grate and varied the flow rate through the inlet pipe. We measured significant washout rates when the flow rate through the inlet pipe was zero. This means that washout was greatest when 100% of the water entering the system was through the inlet grate. When the pipe inlet flow equaled that of the grate at 0.7 cfs, washout was small to negligible (within our measurement uncertainty). The SAFL Baffle was found to function well when there is flow through the inlet pipe equal to or greater than that coming through the grate.

Figure 8: Effluent concentration vs. inlet pipe flow rate. The measurement error for these tests is roughly +/- 20 mg/L, explaining the negative values.

Design Recommendations

  1. When designing SAFL Baffles for use in shallow sump manholes that will receive debris loads, it is important to choose a SAFL Baffle with 5 inch hole diameters. The 5 inch variation is not required if you install into a deep sump. But, if you allow your deep manhole to fill with sediment, it will behave like a shallow sump.
  2. When working with 90 degree outlet manhole sumps, the SAFL Baffle should be installed at an angle of 90-120 degrees with respect to the direction of the flow in the inlet pipe.
  3. Finally, SAFL Baffles should be installed in sump manholes where the drainage basin of the sump inlet pipe is at least three times greater than the drainage basin of the inlet grate.

Supplementary Materials

  1. A link to Adam Howard’s November 2009 UPDATES Newsletter
  2. A link to Kurt McIntire's December 2010 UPDATES Newsletter
  3. A link to www.youtube.com/kurtmcintire for videos
  4. For information about purchasing SAFL Baffles, visit Upstream Technologies at www.revolutionarybaffle.com


We want to hear from you!!!

Let us know your thoughts, experiences, and questions by posting a comment. To get you thinking, here are a few questions: 

  • There are so many other types of devices on the market, how do I know if using a SAFL Baffle makes sense?