Pump station intake geometry 1

[JohnWeal]

Good morning,

We are designing a river intake facility for 210 MLD for a water treatment plant.
We are using 3 duty pumps at 1.05 m3/s each. Pump type is submersible axial flow in steel cans.

I am following the hydraulic institute guidelines and there is a statement saying that diverging inlets are not part of the standards design (words to that affect)

Now the inlet screen channel to the intake facility is 4.8m wide. This is to ensure that at the lowest river levels, the velocity through the screen does not exceed 0.5m/s. It would be silly to increase the 4.8m width to match that of the pumping station chamber width. The flow is straight without turning 90 degrees. (Hopefully you can picture this)

Once the flow has passed through the screens, the stop logs and sluice gate slots, the width of the chamber is increased to accommodate three pumps based on the hydraulic institute guidelines to a dimension of 7.8m. The pump bell intake is 1.021m diameter.

This is quite a divergence required which could extend the distance from the sluice gate to the pumps somewhat.

I would be grateful if anyone can give me any advice on this. I'm sure that hydraulic modelling is the best solution but based on the flow rate per pump, it is not above the 40,000 gpm that the HI recommend for model testing. However, non-standard designs are!!

The pumps operate fixed speed.

Regards
John

 

[Artisi]

always a very tricky call designing intake / forebay / pump installation of this size. I for one wouldn't try redesigning the wheel without a full and careful review of what is being done / has been done previously. Your best bet is to contact major power station owners, of similar users of pumps of this size and similar installations and ask to view their installations and discuss any operating problems encountered and remedial work undertaken if there has been problems.
Even careful modelling of forebay deigns have at times failed with strange flow patterns evident resulting in intake problems on fully developed real flow conditions, but at least with a model test you have an opportunity to revisit the testing and possibly pinpoint where any problem/s may be emanating.

Maybe a bit negative but been there and done that with 2 large (same design etc) pump forebay installations, the problems were model tested at great expense with major design and installation changes suggested to the second of the two installations built at a later date -- even then the outcome was good but not 100%.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[bimr]

Recommend that you review Garr Jones book Pumping Station Design.

 

[piffer]

Use of formed suction inlets can reduce construction cost (footprint size) as well as optimize inlet conditions to the pumps. Check with Flygt-Xylem.

 

[JohnWeal]

Many thanks for your valued responses.

I agree Artisi, modelling is not always 100% but I guess always better than constructing and finding out later.

And yes, have done lots of searches on google images for power station intakes, raw water intakes, river intakes, I have run out of key words!!

 

[LittleInch]

Just a couple of things - please use consistent units and spell abbreviations out first. Makes it less confusing.

210 MLD (million litres per day?) = 8750 m3/hr
1.05 m3/sec = 3780 m3/hr
3 pumps = 11340 m3/hr > 210 MLD.

So which is it?

Do you have a daily total and then a higher maximum hourly flow?

Are all three operating together ( I assume)
Is there a spare?
What is your minimum submergence depth for the bell like intakes (axial pump?)

Everything I've seen tells me the issue is about getting a basin big enough that you don't get individual streams, vortexes or flows occurring which can screw up your design. Easier said than done, but that's the easiest way. My best guess as a start point is you need three diameters of your intake which doesn't hit a wall, another pumps 3D circle and also has at least 1D between bottom and entry into the bell chamber. Maybe too big for your location, but there's no substitute for space.

Ability to add a few plates or anti vortexing devices would be good to adjust things as it goes.

By modeling I assume you mean physical ones? Sounds like a good plant to me if you're trying to economise on space.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.