sewer pump wet well sizing 1

[entp]

Hi everyone,

We're in the planning stages for a project in the middle east, mixed residential and commercial (~60% of the flow is residential). The downstream sewer flow has to be pumped across a bridge.

My guess is that you would size the pump for the max day flow (approx 2 x average flow), and size the wet well to store the peak flow in excess of the max day flow. When I sketch a simplified diurnal curve (simplified to triangles), it looks like the necessary storage would work out to about 6 hrs x (2 x average flow)/2 = 6 hrs x average flow. (The diurnal curve only applies to the residential portion, I know, but it would be more conservative to treat it all as residential, which has a higher peak factor. Or maybe it should be considered, since they peak at different times, and would therefore reduce the peak hour flow significantly?)

Another option would be to have two pumps that each handle average flow (and would therefore both switch on for max day flow), plus an additional emergency pump, and store the excess peak flow. I think there's also usually an allowance for emergency storage.

How should the pump and wet well be sized? I don't have experience with sewer pumps and wet wells, so would like to know how those of you who do have experience would do it.

Thanks,
entp

 

[SteveWag]

It sounds like you are trying to design an equalization facility rather than a pump station. Pick two pumps that will handle max flow each or three pumps to handle max flow/2 each. Use VFDs and alternate two pumps or rotate three pumps. Look at the newest MultiTrode controller to do this. Place an analog level sensor in the wet well. Install a float backup. Make the wet-well just big enough to install and flood the pumps. Big or over sized wet wells collect grit ‘n grease and may have odor problems. VERY important, install and USE hour meters (Running Time Meters). Too many times I have been call because BOTH pumps failed the very same day, only to find that the first pump failed a long time ago. You may also alarm the lag pump operation, that is the second of two or the third if three.

Steve Wagner

 

[PELS]

I could give you a detailed step by step design guide but I would have to first, ask for a retainer and then charge you for my time.

As a general overview process, you first determine both the average and peak flows (in GPM) and you will use them to calculate pump cycling frequency.

Through an iterative process you try different wet well diameters (usually 4' 6' or 8', etc.) to arrive at a pumping frequency that is commensurate with pump manufacturer specs. The depth of the wet well is a function of your physical constraints, i.e., ground elevation, incoming pipe elevation, downstream conditions, storage depth in the wet well and the dimensions of the pump you are specifying.

Then using Hazen Williams, you analyze the hydraulics, trying different outfall pipe (force main) diameters (4, 6, 8 or 10-inch, etc.) and taking into account length of outfall pipe, TDH, minor friction losses, hydraulic slope and velocity (2-5 FPS). Then you utilize manufacturer's specs and pump performance curves to zero in on a model of pump that satisfies TDH, peak flow capacity (GPM) and pump efficiency.

That's it in a nut shell.

 

[entp]

Great, thanks.

We're not designing it in detail yet, we just want to know what flow to design it for. It sounds like you guys are saying that the pumps should be designed to pump the peak flow (~4 x average flow), and not the max day flow (~2 x average flow) that I had guessed.

So you're always pumping it as fast or faster than it comes in, and are *never* letting it lag. ie. It is never pumping and filling up the wet well at the same time, unless there is an equipment failure. Is that correct?

(I was previously thinking that you would store the excess flow while it peaks beyond the pump capacity).

Thanks,
entp

 

[SteveWag]

If your using variable speed pumps the control system will be designed to maintain a constant level in the wet-well and the pump(s) will shut off only when the flow drops below the low flow capacity of the pump.
Steve.

 

[civilperson]

Design for two pumps to equal maximum calculated peak flow and provide three pumps. The volume of the well is better to be large than small with fifteen minutes capacity as a minimum.

 

[ch81pc]

A larger well will give you more time to react in the event of pump failure before you overtop, but with hot conditions septicity will develop possibly causing you issues at the treatment plant. I would have 3 pumps, each capable of pumping 3x DWF on their own in a duty/assist/standby configuration.

 

[onyeuko]

You had a good start by determining the average flow and calculating the max flow as approx 2 x average flow. Volume of your wet well should be V =(max discharge x T)/4. Do not oversize the well as this will lead to stagnation of liquame, oil etc and lead to putrification in the wet well that could affect the functioning of the treatment plant on the downstream end of the system. Only give a little room for accumulation of solid which you should provide by slantihg the base of the wet well towards the pump(from the far end from the pump). T being the intervals between successive switchings of each pump, if you have n pumps, the useful volume for each pump to pump on a circle is equal to Vn = (3600 x max flow)/(4 x n). Decide on the number of pumps that you are to employ(use 2 pumps-one for pumping and the other as reserve if you have up to a max of 10l/s flow, else increase on the number of pumps but always keep at least one for emergency. Decide on the number of your pump circles say 10 times/hour. you should size the pumps for the max day flow (approx 2 x average flow) for which you have to share between the pumps. Using the location of the pumps in relation to the discharge point, calculate the head diferences, loos on the pipe work etc and with these, determine TDH and using the information on the pumps to be employed, determine the appropriate rating for the pumps.
Teddy