I have a scenario where I have water flowing through a gravel filled pipe, and I need to find the velocity. The pipe is gravity fed (no pumps). My approach is to find the seepage velocity ($V_s$) which is the Velocity divided by the porosity. My understanding is that the seepage velocity is defined as the apparent velocity through the bulk of the porous medium. So that is what I need to find.

I found the velocity using the Darcy Weisbach equation and porosity of gravel from online research, however I am getting a very high value for the seepage velocity (about 56 m/s) even when using the highest porosity value.

Another method that was suggested to me would be to use $Q = KIA$ where $k$ = hydraulic conductivity of gravel, $I$ = hydraulic gradient and $A$ = cross sectional area. However, this condition is only for laminar flow conditions (very slow velocities - such as groundwater under an aquifer. So I don't think I can use this equation. This method gives me a very low velocity which also seems incorrect.

Both methods seem to give me odd values so could anyone assist me with this problem?

  • 3
    $\begingroup$ I suggest you show your complete working, in detail - clearly, for both situations including all your assumptions then we may be able to help. $\endgroup$
    – Solar Mike
    Apr 19, 2018 at 11:51
  • 1
    $\begingroup$ "Gravity fed" doesn't mean anything. You need to know the pressure head involved. Next you need to justify your claim that the answers you're getting are wrong. What's your baseline reference? $\endgroup$ Apr 19, 2018 at 16:56
  • $\begingroup$ Sounds like flow through a porous medium. You may be able to apply poiseuille’s law. $\endgroup$
    – Paul
    Apr 21, 2018 at 13:03
  • $\begingroup$ For porosity you could fill a 1 gallon bucket with gravel then measure the amount water to fill it. If it takes 1 quart then the porosity is 1/4. $\endgroup$
    – paparazzo
    Apr 23, 2018 at 14:31
  • $\begingroup$ Are you sure you mean the Darcy-Weisbach equation, as opposed to Darcy's law? The two are very different, and I would expect the latter to be more appropriate in a gravel-filled channel. $\endgroup$ Aug 4, 2021 at 11:37

1 Answer 1


Theoretically, assuming your pipe is completely full of gravel and fluid this http://www.deq.idaho.gov/media/60177882/rpa-lesson-plan-1.pdf should provide the material you need. If it is not completely full of fluid, and is a sewer/channel with gravel in it, use the manning’s equation.

Below are some answers a professional may give depending on the application:

  • Calibrate – You seem to know the answer you want, reverse engineer to find the friction co-efficient of the material you are using.

  • For drainage – Replace, augment or clean the pipe;

  • For filtering – Can’t change filter material, gravel is about as course as it gets, increase pipe diameter or apply an increased pressure – either through a pump, static head, or increased slope.

Practically, depending on your situation, do one of the following:

  • Seek professional advice; the implications of your answer can have real world consequences such as flooding either upstream or downstream. Such risk requires monetary reward, if only to pay insurance. A “perfect” (always correct) engineer still faces a high risk of their design being incorrectly applied, and to prove you’re correct in a court is expensive.

  • If you are a professional either tell the client you are not qualified, or seek the advice of a principal engineer in your company.

  • $\begingroup$ I remember reading about this in a textbook. The way was to measure the volume of the empty pipe and then fill it with stones to capacity and add water until filled to capacity. Next, drain water and measure the volume drained divided by length of pipe to find sectional surface area. From that result you can determine a theoretical flow rate according to volumetric flow. The wet gravel water content accounts for frictional losses. $\endgroup$
    – Rhodie
    Nov 20, 2018 at 20:10
  • $\begingroup$ The lined page was not found (maybe removed). $\endgroup$
    – r13
    Apr 5, 2021 at 20:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.