Self Cleaning velocities tailored to the type of sediment?

[Shannon K]

I have found a fair amount of literature quoting a number of different self-cleaning velocities for various situations, all of them slightly different depending on the country and/or state which is supplying the figure. However I am wondering if there is some way to select a self-cleaning velocity that is tailored to the particular situation. Otherwise it seems somewhat arbitrary.

I am designing a closed pipe drain to empty a saturation zone in rural Indonesia. Sedimentation is a huge problem here and maintenance is practically non-existent, therefore I want to make sure whatever grade I have specified will definitely result in at least very low rates of sedimentation deposition, especially since I will be incorporating an inverted siphon in my design, which increases the likelihood that sediments will settle out of suspension.

It seems to me like there must be some way to calculate an ideal self cleaning velocity based on the properties of the sediment it will have to be carrying. There are three main considerations

1. Should be fast enough/high enough energy to carry the sediment with minimal deposition (normal operation)

2. Velocity should be high enough that shear stresses will carry the coarser particles along the pipe floor (i.e. force of the water should exceed shear frictional stresses between soil particle and pipe wall)

3. Ideally there would be another velocity where it would be sufficient to exceed shear stresses between the soil particles themselves, thus allowing the sediment deposit to be eroded during a large flow event (this can be less frequent, e.g. a massive flood event)

I have soil samples and can determine the soil composition. The pipes will be RCP (round concrete pipes). There must be some way to relate the shear stress exerted by the water to the shear stresses that are required to be overcome, in order to work backwards and calculate the self cleaning velocity?

-Engineering 3rd year student, this is my first time designing a drain system.

 

[beej67]

Nobody does this in practice for simple culvert design, but the mathematics has absolutely already been worked out. It's a headache:

Terry W Sturm - Open Channel Hydraulics
Chapter 10: Flow in Alluvial Channels

You end up blending material properties, Stokes Law for falling spheres, turbulent properties, and all sorts of other fun and complicated stuff into an expression of 'bed load' and 'suspended load.' It's pretty complicated, and there might be a better treatment in a different book. Sturm can be a bit wordy at times.

I've only ever seen this done in "reverse," to try and create stable natural alluvial channels, instead of keeping pipes clean, but there's no reason you shouldn't be able to turn the treatment backwards and solve for pipe cleaning velocities.

Another engtips user might have a simpler approach.

Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com