Pump Station at the bottom of a hill 1

[ChEMatt]

Working on a project to install two pipelines, freshwater and produced water, going 10 miles from one atmospheric pit to another. They don't mix if course. It has been determined that a booster pump station must be installed at a midpoint which also happens to be at the bottom of a hill/draw/whatever you'd call it. This is due to access. There's about 100' of height before the booster pumps, then about, oh, 90' of height after the pumps. I did my best to convince folks that we should put the station at the top of the other hill; seemed to make sense to me, but employee access (there's no easily accessible road to the other location) was more important.

My concern, not having done any of these in this type of location, is the velocity and momentum of the water coming entering the pumps. Could the water somehow rush ahead, possibly vaporize, cause cavitation, etc. issues at the pumps? If the pipe does, in fact, stay liquid full, does the speed of the pump have a factor at all with reference to boosting the pressure? Is this no different from any other pump install, with the exception of having more NPSH available? I feel like I'm overthinking this but I just believe, common sense (I hope), that having a gravity fed pump like this has many more considerations than just the discharge pressure to get the flow back up the hill and to it's destination. I'm not confident that the engineering company that did the hydraulic assessment did much more than just determine the pressure required at the bottom of that hill to get the water to its destination.

I am aware that I would need to preform a surge study to determine the pressure surge at the pump station should things get shut down suddenly, or blocked in.

The pipeline is about 10 miles long. Flow rate is around 35,000 bbld through a 10" poly line, SDR9. We'd like to keep the pressure below 200 psig; it's a comfort level thing for the guys I suppose. I need to refresh myself on the required flow so that might not be the right number. Just getting back into the project...

I'd appreciate y'all's advice on considerations, things to look out for and ask, etc. The project was started by me at which time I confirmed the design basis and routing of the pipelines, another engineer took it over and nudged it along a bit and had the hydraulic study, and now I'm taking it back over. I'm going to re-engage the engineering company but also get some pump skid fabricators and the pump vendors themselves involved to get their thoughts on how this seems to be coming together.

Thanks for your help and advice!

 

[LittleInch]

Can you post a profile so we can understand this. I'm confused as to where this 100 ' and 90' and the elevation of the start and end points are.

Also can you describe a bit more how you plan to start and stop this system.

Booster pumps like this can have some specific issues, but at the end of the day it's still just a pumped system of some sort that needs to be designed. Depending what your static pressure is the pump inlet may need to be designed / specified a bit differently, but not too much of a bother.



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

 

[ChEMatt]

Something like that. Even more fun is the system needs to be bi-directional, so we'll be pumping either way through either line.

Starting and stopping the system would be simply shutting down the pumps at the source and destination. My people insist on these things being manned, so someone would be at the middle pump station to start and stop it, line up the valves, etc. If things are already lined up, I imagine they'll just hit the "go" switch from the Ops Center and get confirmation that the pumps are running. (Staggered a bit, of course). We're filling open pits, so think in terms of one million barrels. We've got plenty of time to get out to operate the system. And ESD of some kind would simply shut the pumps down. Definitely concerned about surge at the inlets/outlets of the pumps so PSV sizing will be important.

 

[LittleInch]

Errr, what is going to make the water flew 4 miles from either pit to the midpoint pump?

You mention a pump at the source. Is this pump good enough to establish flow from one end to the other before you start the booster pump?

Do you have a head loss vs profile graph?

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

 

[ChEMatt]

There's pumps at either end. This pipeline is going to make the connection between the two sets of pumps.

We could flow from one end to the other with no booster. We can't make the *rate* we need, however. We'd be much, much lower than desired.

No graphs at the moment, sorry.

Just wondering about what needs to be done other than PSVs at the entrance to the pumps in the middle at the bottom of the valley. Any special design considerations with the gravity assist?

-m

 

[miningman]

I'd suggest you start educating yourself as to the flow characteristics of pumps in series......... I think you'll be disappointed

 

[Artisi]

Seems to me it's time to employ a qualified and experienced engineer. As LittleInch has pointed out, it's just another pumping system - - certainly not rocket science.

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.)

 

[ChEMatt]

miningman, telling me to educate myself... In a thread I started, asking questions... I'm wondering what you thought I was trying to do? Maybe you missed the first post. Scroll up perhaps? If you have some knowledge on booster pump stations, this is your chance to shine!

Artisi, not sure where you see me making up theories without data. Perhaps you'd care to elaborate? Or is that just your signature?

The pumping system is simple. Feeding one centrifugal pump into another isn't rocket science as you all have noted. My questions were around the dynamics of water rushing down a hill. Perhaps there's nothing more to be said about that and it's just water in a pipe, perform a surge study for safety, and be done with it.

-m

 

[LittleInch]

OK,

There are a few issues here which you need to address.

The first is the hydraulic profile of your line, initially in operation with one pump and then with the booster pump.

Being what appears to be a valley on a flat plain? you have an issue with pressure. Your 100 ft valley equates to about 45psi out of your 200 available. So you might end up being limited by this pressure.

Then when you're working out the hydraulic head line in booster pump mode you can't run your hydraulic line direct to the pump inlet, but to the top of the hill on the inlet side(s).

Series pumping with the pump in the middle requires the pumps to be essentially equal and even then you need to be able to control pressure and flow from each one so that the system doesn't become unbalanced.

If you maintain a fixed back pressure at the booster pump equal to the depth of the valley then you won't have any issues with water rushing or breaking away.

The key is in the initial design and hydraulic assessment. Your flow rate given seems to be with the booster pumps installed from a quick assessment I've done and accounts for the pump location in the valley, but it's very tight and needs confirmation.

The other issue you can get is booster pump start creating an initial over pressure and then tripping the pump due to high pressure (assuming you have a high pressure trip). To start the pump you might need to either run it as a soft start / VFD, control the pressure with a valve for a short period or accept the short term over pressure.

Oh and by the way it's best not to "react" to comments. If you don't like what is said or the way it is said far better to ignore it. Just saying.

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

 

[ChEMatt]

LittleInch,

Excellent points, and a star for you of course. We've got one other install with a similar setup, boosters in the middle of a pipeline from point A to point B. However there isn't nearly as drastic a difference in elevations. This company uses very little in the way of automatic controls but VFDs were a part of the install; that's how they matched up the curves so to speak. I think you've nailed what I need to cover, and to those still reading I am enlisting the assistance of a 3rd party engineering company. I'm definitely not doing this in a vacuum, but at the end of the day I'm advising my company on the direction, scope, and schedule going forward.

It isn't a flat plain, but this is the most drastic drop, and as much as I wanted to put the pump station on the side or top of the 2nd hill (going left to right) we can't access that location easily, roads and all. I've modeled the system in both VMG (now called Symmetry) and PIPESIM so I'm confident about the pressures required to meet the flows, and where a boost is needed.

And as far as responding to someone saying something patently obvious... Telling someone to educate themselves when said person is trying to educate themselves... I think they need to get called out just a little bit. I ended my response with an encouraging note tho! Now's their time to shine!

Thank you sir, I do appreciate the helpful information and advice.

-m