Pumping lake water

[pointsman]

Good morning to everyone from a long time lurker and first time poster.

I’m working on a design to pump water from a lake for industrial use in a mountainous region, spanning a few dozen miles. The distribution system consists of a long main with branches leading to each point of use. The network will have typical distribution system elements such as booster stations, pressure reducing valves, pressure release valves, control valves (such as flow limiting and float valves), air release valves, water tanks, and hydrants. The system will be entirely constructed new.
The source water will be from an impounded lake, which appears to be eutrophic. A series of samples taken one day in summer showed suspended solids in the lake up to around 2 to 50 mg/L, with lower concentrations at the surface. Algal growth is present and I don’t know if blooms occur. Upper strata are oxygenated and lower strata are anoxic.

The project currently only provides for coarse screening of the raw water, since the client plans to treat the water according to their needs at the points of use. Therefore we are currently planning to pump this raw lake water through an extensive pipe network without any pretreatment except for the coarse screening. A quick calculation shows that with our flow rate of 150 l/s, that would be up to 650 kg of solids pumping into the piping network every day, which gives me some pause, starting with wye filters around the network.

Our supplier of the control valves advises wye filters (2 mm mesh) upstream of each control and pressure reducing valve, and I can imagine a scenario in which these filters are frequently blocked by lake solids. I could try to substitute a number of these valves, but some will be necessary due to the high grades present.
I’m also wondering about sedimentation in the lines during periods of downtime, and subsequent regurgitation of sediment particles if a flow disruption (eg. pump trip, valve slam) occurs.

Biological growth (biofilm, mussels) is also on my mind. Long tracts of the pipe line (HDPE) will be above ground, exposed to ambient heat (about 40F to 100F).
I’m considering providing finer screening of the raw water, such as a pipeline strainer with a 2 mm mesh, to limit blocking of the downstream wye filters for the control valves. However I want to be sure that I’m not missing the bigger picture, in the case that more extensive treatment to clean up the source water is needed.

This kind of system is new to me -- my experience is moreso with clean water systems. The closest experience I have this would be an effluent water supply system from a wastewater treatment plant, however that system was not nearly as extensive.

I would appreciate any guidance regarding the design of this type of system.

 

[1503-44]

Since you do not provide a system sketch and topographic details, or minimum and maximum flow rates, you limit any serious discussion to the aspects of pumping clean water verses dirty water.

Assuming that there is a high danger of sediment plugging during down time, you must minimize down time. You can do that by designing for the a constant minimum flow rate. Supposedly you would provide tanks near your points of high rate use to provide temporary maximum flow capacity from the tanks. Perhaps settling sediments in the tanks could be part of your cleaning methodology.

Designing for a constant minimum flow rate, your system must have some excess capacity above minimum demands, such that the system flow rate will be able to also fill tanks at a rate high enough to have your tanks full when maximum demand is needed. You should design pipe for the optimum flow rate between minimum and whatever you need to keep tanks full enough to supply max demands at points of use.

It may also be advantageous to consider having spare pumps to avoid down time due to pump maintenance, such spares might also be run during high demand periods to meet your system's maximum flow fates.

You know that you will treat the water at some point, so you must decide if your water treatment program will be best served by having one treatment package, probably at the source lake, or having a number of treatment packages at, or near user locations. Typically it will be more cost effective to have one large package, rather than many small units. The maintenance of these facilities is very much easier if there is only one. Two facilities can require 2 to 4 times the effort and costs, keeping in mind that different sized spare parts may be needed, if some standardisation cannot be adopted. That also applies to booster pump stations design.

There is probably a huge advantage in maintenance and operation costs in having the cleanup package at the source. Once the dirt is in the system, it is far more difficult to control and get it out. If you have bios such as mussels, they usually have very bad consequences in pipelines. Much more than a little sediment does. I would not risk getting them into the system. Screens won't help. Biocides are nasty all around. Calcium is hard and abrasive. Pipe roughness makes them expensive to tolerate. And their reproduction rate is astronomical. Not cleaning up the water at source risks having very high operating and maintenance costs.


Can't really say too much more without some idea of system configuration options, topographic details and min/max flow rates.



--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[pierreick]

Hi,
Consider using sand filter(s) to remove the suspended solid in your raw water. I've used some of them after pumping river water as a make water on cooling system.
Like Mr 44, mitigate the risk of having accumulation of solid in your network.
Good luck.
Pierre

 

[pointsman]

Hi 1503-44, thank you for your response.

Due to the nature of the industrial use, downtime of the system is inevitable. There will be stagnant periods.

Using the tanks in the system for sedimentation is an interesting idea. In any case sediments that do fall out in the tanks will need to be managed.

Is it your opinion that the system is risk for sediment plugging, and that a filtration system is not going to mitigate the problem? What information would you want to see in order to be more certain?

Playing devils advocate, sewerage systems successfully convey waters heavy in solids without major problems. We’re going to be using pumps capable of passing solids, and will maintain service velocities between 3 and 4 feet per second. The control valves will need filters yes, but really is this system at much more risk than a sewer pressure main?

In your experience is a chlorine feed enough to resolve the biological issues? I’ve seen this done for desal plants.

Unfortunately the client is set on a decentralized treatment approach (and we are well into the design phase by now). I agree that centralizing the treatment is makes sense, but changing treatment direction at this stage would be a significant departure from the currnet project direciton. I’d need to be quite sure of my concerns and recommendations before proposing that this solution be put to the client.

 

[pointsman]

Thank you for your reply Pierreick. I've seen sand filters used for solids control in pipelines before, and this is certainly something to consider.

In your opinion what sorts of issues might we see if no pretreatment (like filtration) is provided?

 

[1503-44]

Anytime you have sediments, there is a risk of plugging. You must always have enough velocity in the pipe to keep these sediments in motion and washing along through the pipe. That still does not guarantee freedom from local accumulations that might occur and say "slide back downhill" and form a more solid plug as soon as flow stops. In full slurry flow pipelines that will form hard plugs, you never, never, never want to stop. You always want to maintain at least the minimum scour flow velocity. Of course sand may not form tight plugs, but alge, mussels and shells will tend to lock up or glue stuff in place when the opportunity exists. Risk also depends on configuration of the system. If you have a long uniform downslope, that's less risk, but you probably do not, since you are mentioning booster pumps. Low spots could be problems.

Alternately you might consider having sufficient clean water reserves to fill the system with it before shutting down, or maybe some, or all of, locating cleanouts and holding tanks at low spots, or regular use of cleaning pigs, but all that would cost nearly as much as a pretreatment facility at source, not considering potentially heavy maintenance activities and maybe guaranteeing nothing in the end. Every intake line I've ever heard of with mussels in the system seems to be a operation and maintenance nightmare. They are not usually found in sewerage systems.

Some fine sand does not particularly scare me, as long as it won't lock up, although you may need some extra velocity for cleaning purposes from time to time. If you get it little by little, not too much problem, but if it comes in large quantities when lake level gets low, or when demand is low, that could be a problem. A lot of the risk depends on the details of your design, operating philosophy, maintenance skill and budget.

I'd want to see a system configuration and profile before I would make any attempt to evaluate plugging risk, especially if there is any anticipated down time.

Evaluate your treatment problem with a cost-benefit approach with appropriate operation, staffing and maintenance costs. To me I have no doubt that treatment at source must be better. Actually I have never seen anything else... ever. Have you? Petroleum flow lines take whatever comes out of the well, but we are prepared to handle that with source seperators when needed and pig launchers and we do not shut down very often, certainly not before running a batch of cleaning pigs whenever we notice pressure increases. We do not have mussels. Got no direct experience there, but I would tend to think fine filters in a pretreatment system, rather than screens. Screens will not stop eggs. From my readings, they are extremely prolific. If you're not worried about water quality, then maybe not a problem, but I sure wouldn't want traces of dead mussels showing up in my drinking water.

Clients always want stuff they shouldn't have. Part of the job is talking them out of doing dumb things.


--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[EdStainless]

You need to do biological control at the intake.
I would suggest using a bar screen for big stuff and then a fine wedge-wire screen for finer trash.
Often biocide is injected between the two screen systems.
Sand separators will save a lot of pain and maintenace costs down the road.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Compositepro]

The great things about a sand filter are that it will work well in removing fines as well as coarse debris, and backwashing is pretty much fully automated.

It may be beneficial to recirculate water through the filter back into the lake so that the lake water does not become too dirty.

 

[1503-44]

Clients always want stuff they shouldn't have. Part of the job is talking them out of doing dumb things.


--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[pierreick]

Hi,
Very obvious, risks identified are Fouling, Erosion, Corrosion. Stagnant pipes are prone to develop bio (bacteria, algae,etc)
BTW you don't want to have your hydrants plugged the day you need to use them. Regular tests to be implemented.
Note: Talk to the insurance broker and Firefighting department and may be EPA about your project to ensure you will get your operating license.

Pierre

 

[LittleInch]

Deliberately putting dirty water into this system is simply a recipe for continuous filter plugging and things getting worse and worse over time.

I've done seawater extraction systems and they all had either back wash filters or cyclones to remove as much dirt as possible at the inlet. I would strongly recommend you do the same or at the least allow the space and connections at the inlet so that you can add them when the client sees sense.

I would also recommend you remove the interior weld bead on the PE pipe. The contractor will hate you but again the long term benefits are huge.

Also watch out for thermal expansion of your above ground pipe and sagging over time as PE expands a lot and also slowly creeps under constant load or weight.

Then dose the water with something to stop all the algae and animals growing inside it.

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

 

[bimr]

It will be less expensive to have a central treatment plant instead of multiple treatment plants. The central treatment plant will also keep the transmission piping clean as well.

The water supply system should consist of a well screen type intake in the impoundment, either an ultrafilter or sand filter to remove the solids, and disinfection.