Water transfer system not properly designed 2

[rutherford703]

Please see attached drawings for information.
The system was designed to run 2, 3 or 4 pumps based on pump suction pond level and water user requirement. The system must keep certain pressure at the branch to send water to user 1.

The problem with figure 1 is that the section of the piping (sloped) downstream of the vacuum breaker is equivalent to a pump. Throttling pump suction will cause pump cavitation. Here will cause the vacuum breaker to open whenever the pressure control valve open reduces to let the air in which is not desirable because increased O2 level in water will result in piping O2 corrosion.

I think that the system should be designed as Figure 2. In such an arrangement, the vacuum breaker will not open during normal operation.

What's your thought? I appreciate your input.

 

[BigInch]

Go to correct design and set PIC to hold 186 kPag, closes as pressure drops below, opens if pressure goes above. That setting will hold at least 110 + any frictional pressure loss between user 1 and PIC, at user 1's branch takeoff.

Vacuum breaker is only needed to prevent vacuum conditions from forming at high point when system is shut down and PIC fails open, or system drains by backflow through stopped pumps, or if PIC has a large flow capacity and fails open. When pumps shut down, PIC would normally close as it tries to keep pressure at the setting of 186. Rig PIC to fail closed.

What would you be doing, if you knew that you could not fail?

 

[rutherford703]

Thanks BigInch. I just identified that the configuration is a problem. I am not sure if they are willing to correct it because the piping size is 48" and the new location will require instrumentation connection.

 

[BigInch]

In the current configuration,

Whenever downstream users take more than the maximum steady state flowrate than the down-sloped pipe segment can deliver, or user 1 takes more flow such that PIC begins to close too much to try to hold and flow into the down-sloped segment begins to drop below critical flow, the lowest flow that still keeps the down-sloped segment flowing full, the vacuum breaker will open.

The current configuration may have another problem as well.

Partial flow in the downslope is "cascade" flow, as the water is kind of cascading over the high point as it flows into that segment. At lower flows, not only will corrosion begin, but it may also start concentrating the erosive effects of any particulate matter in the flow stream too. As flow in the downslope decreases, velocities may tend to increase as it tries to maintain a steady state flow while the partial flow condition results in flow only running along on the bottom of the pipe. That might concentrate the erosive action enough to begin to cut through the bottom part of the pipe wall.

What would you be doing, if you knew that you could not fail?

 

[bimr]

How about posting Fig 2 in pdf format.

 

[rutherford703]

Thanks all for your response. Figure 1 is the current operation. Figure 2 is what I think is right. Figure 3 is currently work scope to address the problem.

The scope of work is to reduce water O2 level out of the piping. Two sources contribute to O2 level in the water. One is the air entrance to the piping by vacuum breaker and the other is the turbulent at the exit of the pipe

To address the vacuum breaker openning, two orifices to be installed - hoping to back up the water flow in the sloped piping from partial flow to full pipe flow. My concern is that installing the orifices may not solve the vacuum breaker opening problem and by the contrary restrict the flow through the piping for high flowrates (flow to user1 is about 10,000 gpm, downstream of the user1 takeoff, flowrates can be 30,000, 50,000, 70,000 gpm depending how many pumps running).

To solve the exit turbulent problem, a diffuser will be installed. The diffuser will be an Y shape with each branch about 60m long (48" HDPE DR17) perforated pipe with holes on pipe both sides. I once proposed using slots instead of holes to increase exit area and reduce the length of the diffuser.

Another problem related to the diffuser design is how to ensure uniform flow through each hole.

 

[bimr]

What is the approximate length of these pipeline segments?

What is the pipe material?

What is this application? Does the water have solids?

 

[rutherford703]

bimr to your questions:

upstream of the Pressure control valve 48" CS, 750m. Downstream of PV, 52" HDPE, DR17, 3000m.

Fluid: water with about 5 wt% solids

 

[bimr]

Regarding "increased O2 level in water will result in piping O2 corrosion"

The HDPE pipe downstream of the PV is essentially inert to corrosion. So why are you concerned about the O2 level?

Also note that removing the O2 may not stop corrosion in any case. It is possible to have MIC of steel piping from anaerobic bacteria.

Regarding "To solve the exit turbulent problem"

Why are you concerned about O2 from this turbulence? It appears that you are discharging into an open pond where oxygen from the air can enter the pond.

However, there should be some type of spillway at the discharge point to prevent erosion, possibly a diffuser will help, but a diffuser may be subject to silting problems.

Regarding "Vacuum breaker is only needed to prevent vacuum conditions from forming at high point".

A vacuum breaker is probably not necessary to protect the HDPE, DR17 pipe. I would think that the purpose of the vacuum breaker is to prevent pulling suction through the control valve.

You should check with the pipe supplier and determine whether the vacuum breaker is necessary and if so, if the vacuum breaker is sized correctly. You may be able to use a check valve instead of a vacuum breaker.

I would think either Fig 1 or 2 would work. You do not mention the climatic conditions. Fig 2 with a full pipe may freeze if not buried. Relocating the valve as per Fig 2 will be quite expensive. The vacuum breaker is definitely not needed with the scheme in Fig 2.

In Fig 3, you have added another vacuum breaker. It is also not necessary.

 

[BigInch]

It is possible that he needs the vacuum breaker, if the large D HDPE pipe is thin wall. The need for the vacuum breaker can be negated, if there is a pressure control valve at the end of the pipeline which will maintain full flow across the pipe cross-section at all times. That will also help stop any ingress of air backwards into the pipe from entering at the outlet during the partial flow conditions that he appareltly gets now without any control valve.

Problem with trying to substitute orifices for control valves at the outlet is that they may not have the most desireable effects when subject to variable flow conditions. They will have to be carefully designed with both flow and pressure control in mind. Not easy when flow is variable.

I also do not understand the concern about air with HDPE, and air is probably also coming from entrainment or disolution in the supply water anyway.

What would you be doing, if you knew that you could not fail?

 

[rutherford703]

OK. The corrosion concern is not for the HDPE piping. The water will be discharged to another pond. From that pond the water is pumped through a serious heater exchangers to about 65°C and then the water is sent to another plant at several KMs through a piping line which has been experiencing serious corrosion problem.

Thanks for your response.

 

[bimr]

A large number of parameters affect pipe corrosion, including water quality and composition, flow conditions, biological activity, and corrosion inhibitors. Water quality is usually a larger factor than oxygen content.

I would suspect that you may have some sort of microbial induced corrosion. Periodic disinfection with chlorine would probably be helpful.

If the surface of the pond is open to the atmosphere, you will be absorbing substantial amounts of oxygen into the pond. It is possible to document the source of the oxygen using a DO meter.

 

[rutherford703]

Investigation of piping corrosion has been judged to be O2 corrosion.

Someone also did DO analysis for the water system. They found that the DO level in the pump suction is 0 (ppm based on my memory). The DO near the piping discharge is 9 ppm. At same time, we had another source of water from different pond. The DO level in another source of water was about 3 ppm.

We have to solve the corrosion problem because piping leaking causes environmental issue.

Thanks for your thought.

 

[bimr]

The anecdotal evidence that you have presented is not convincing. The DO in the pond probably fluctuates diurnally, so the samples may be just points in time.

I recently installed a step aeration device at a treatment plant and the DO was only raised to 6 mg/l. So, the 9 mg/l seems to be a little high.

The water quality is probably a bigger factor in the corrosion.

And what about lining the carbon steel pipe?

Anyway, back to the point of keep the oxygen out. Fig 2 is probably the best option. Keeping the pipeline fully pressurized will stop the introduction of air at the vacuum breaker.

As far as the discharge of the pipeline, you should have the discharge point to be underwater to minimize turbulence at the surface. You have not presented enough information on your process to suggest a design, but Cormix has diffuser examples.

http://www.cormix.info/picgal/index.php

http://www.tvrl.lth.se/fileadmin/tvrl/files/vvr176/lecture7.pdf

 

[rutherford703]

Impressive!