Siphon flow over a dam 3

[Javaun]

We are trying to remove hypolimnetic water from a reservoir and discharge the water downstream using a siphon. The problem is the deep spot in the reservoir is 800 feet away. The maximum head downstream is ~20 feet. We were planning on using 4" HDPE pipe but from my calculations, first using Bernoulli's (frictionless flow) to get the flow rate for various values of Head and pipe diameters. I Then used the Hazen-Williams formulas for flow rate and enter various length of pipe to find the friction loss. The friction loss is just too high with that length/dia. of pipe. I can get it to work if I use an 8" pipe, but it can only be 50 feet long. Am I correct? The friction loss head has to be less than the actual head below the dam in order for the siphon to work?

 

[dcasto]

you need a valve on the outlet to limit the flow rate

 

[Javaun]

We are planning on using a valve on the outlet as well as at the crest of the dam to facilitate filling the pipe to initiate the siphon.

 

[cvg]

see the attached chart which may help

 

[SteveWag]

You can use any size (diameter) pipe, the question is how fast, or at what rate, you want to remove water. Using 1000 feet of pipe and 15 feet of available head gives 1.5 ft/100 of allowable loss. 4-inch pipe delivers 150 to 170 GPM, 6-inch about 400-450, and 8-inch about 900-950. 1-inch gives about 4 GPM. Siphons can be tricky, the downstream end needs to be submerged, a 55-gallon drum works well for the pipe sizes you are working with. At the high spot you need some way to remove air, both initially to prime and full time to remove collected air. For a 6-8 inch siphon I usually use a laboratory style vacuum pump (1/3 HP or less) or a peristaltic pump. Peristaltic pumps are nice because they pass water as well as air and are subject to less damage. I usually inspect the siphon every other day and remove air (gas) unless experience tells me otherwise.
Steve

 

[Javaun]

We are now looking at running a 10" pipe to the deep spot and reducing the pipe diameter to 4" where it goes over the dam. This seems to work in keeping the velocities/friction loss low in the large pipe and keeping the head required reasonably low.

 

[btrueblood]

"The friction loss head has to be less than the actual head below the dam in order for the siphon to work? "

Yes, with the added phrase "at the specified flow rate".

In your spreadsheet, at the friction loss calculation, you would simply start reducing the assumed flow rate to find the discharge flow that would satisfy: friction losses < available head. Not clear if the equation you are using accounts for inlet loss at the suction.

 

[ione]

Maybe this could be helpful (at page 5 a generalized equation is given for calculation purposes)

 

[ione]

Here a papaer which shows the theoretical background and a practical sizing example.

 

[cvg]

another tip - make sure you have some sort of trash rack on the upstream end. If the siphon gets blocked by trash, your flow rate will decrease or even stop completely. It will be a major effort to clear the plugged siphon. Also, make sure you include any headloss in the trash rack as well as the appropriate inlet loss before you finalize your pipe sizes.

 

[itsmoked]

Any concern about pipe collapse if the suction side gets plugged?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[rconner]

If you want the siphon to deliver the minimum amount of water you need, and particularly over long distances, it may be useful to do some research so as to plug in values for friction coefficients of piping materials that are backed by real world testing of piping systems in similar applications (it appears some manufacturers have in general sometimes promoted values for pipe that in effect, at least for smaller pipe sizes, are smoother than the ideal/smooth pipe curve).

 

[sodengineer]

sorry for the thread creep, but when the OP stated hypolimnetic water, does he mean just the stuff below the thermocline, or will the inlet of the pipe just be at toe of the wet side of the dam? In other words is this a way of draining an old reservoir (as a safety valve), or will the siphon be used regularly?

 

[Javaun]

Hello, and thank you all for the responses. We are attempting to remove the low-quality water from below the thermocline and the siphon would be operating continuously, or until the reservoir level drops ~6 inches. Our problem is the deep spot is 800 feet from the dam and friction loss in the pipe is excessive in anything smaller than 8" pipe.

 

[MikeHalloran]

How many acres does this reservoir cover?


Mike Halloran
Pembroke Pines, FL, USA

 

[MikeHalloran]

You can't just plug the computed flow from the Bernoulli equation into the Hazen-Williams equation. It represents a maximum, only.

Instead, you can, e.g., plug the pipe length into the Hazen-Williams equation, and an arbitrary flow, then goal seek, varying the flow to make the friction loss equal the available head. (Modern spreadsheets do Newton's Method pretty well).

Doing that for 800 feet of 4" id pipe and a head loss of 20 ft gives a flow of 0.5 cfs (cell C37).

Knowing that and the area of the reservoir would tell you how patient you have to be while it's draining down.

Expanding the inlet end, e.g. a bellmouth, is probably a good idea to minimize induced circulation in the bottom.

Don't orient the inlet vertically, or it will capture fish. (They will sense and flee from a horizontal current).



Mike Halloran
Pembroke Pines, FL, USA