Sand filter sand

[swmguy]

Just curious what the stormwater design/maintenance community has to say about the Austin sand filter design spec for media- they call for C33 sand, but limit the grain size range to .02"-.04". I've been in the construction materials business for over 20 years, and in stormwater design and maintenance for an additional 8-10 years or so. I have yet to find an aggregate supplier in my area that will produce a sand filter media that meets the .02"-.04" limitation. From what I can tell, EPA used the Austin sand filter design and promoted it, so many jurisdictions have this .02"-.04" spec for sand filter media. Basically, the media must pass a #18 sieve and be retained on a #35 to get a product that is close to .02"-.04".

Just the logistics of producing a product that meets this range seems daunting- we're looking for natural sand, not crushed material. How much sand and gravel do you have to mine to generate any quantity of this narrow band? Is a typical screen setup able to be fine tuned to produce this range? Screens that fine must use a wet wash process, seems like a dry screen would clog constantly. In fact, at the level of production needed in a sand/gravel mine, I would be surprised if any sieves smaller than #10 are viable.

My jurisdiction(I'm a public employee now, it's a bit more stable as I count down toward retiring) is considering dropping the .02"-.04" spec in favor of the generic C33 spec. My opinion, after inspecting and maintaining several hundred above and below ground sand filters over the years, is that a wider range of particle sizes, a "well graded" blend that meets C33, will be less likely to clog. It may not have the pollutant removal efficiency of the narrow spec, but, taking the long view- I'd rather lose a bit of filtering efficiency over the long term than lose it all due to premature clogging. And we all know how thrilled a property owner is to pay for the reconstruction of a failed underground sand filter(especially when the designers only provide access via manhole- ever move a few hundred tons of dirty sand via single manhole on a filter the size of a basketball court? It gets expensive from a labor perspective.)

I hope I'm not operating in a bubble, hence this post. Bosses ain't gonna pay to send any of us to StormCon to discuss this over drinks with our fellow regulators- so here I am.

 

[cvg]

this is what happens when agencies dabble in engineering. leave the sand filter specifications up to the design engineer. C33 sand is generally a good filter material and readily available. I'm not sure what your application is, but you might want to consider specifying:

no lime or other additive to control fines
less than 5% fines
passing Vaughn & Soares test

https://directives.sc.egov.usda.gov/OpenNonWebContent.aspx?content=17751.wba

 

[Twinkie]

C33 is all I ever spec

 

[swmguy]

That jives with my thinking. I'm gonna pull that narrow ratio out of our design manual. Reminds me of the days when I'd get a set of approved plans specifying a geotextile with properties that no existing product met. Too much "unobtanium". Now that I'm in a position to make these changes, hopefully we'll see better performance out of the BMP's. Now I just gotta figure out how to compel owners to maintain what they have without stirring up a political hornets nest. Appreciate the feedback.

 

[TerryScan]

Someone must have copied this spec everywhere. Delaware's spec for sand filters also calls for C33, .02-.04. However, I have never heard of enforcement of that gradation. As you point out, it is not available and the hundreds of installed sand filters seem to work acceptably with just the general C33 spec.

 

[pinkpig]

sand filter/fabric filter or the combination is to retain underlying soil from being washed out and at the same time to allow water release to ground. The filter needs to be designed to meet the grading requirements (FHWA, USACE, or NRCS has such criteria) and sometimes you need multiple layers of filters. I know in practice quite a lot of engineers just throw a filter per whatever agency's std specs and this is dangerous. For you question, I would suggest check the filter gradation against the criteria and use what's right in the design instead of what's is the std specs.