Multiple pumps in a pipeline?
Multiple pumps in a pipeline?
(OP)
I am a summer intern at a large oil and gas producer. My project is to design a pipeline and gathering system for saltwater disposal in one of our units in the Bakken. I've got a few general questions about pipelines and pumps. Attached is a crude picture of the pipeline I am designing. Each black square is a central tank battery (will store produced water from different wells, which will be pumped to the disposal site once a day), and the red dot is the disposal site that all the produced water from each battery will flow to. I am trying to figure out what style of pump would be best for this project (a supplier has recommended either a centrifugal or a PD progressive-capacity Moyno.) Here is my first question…say that all the CTBs have the same pump that will discharge 100 gpm. And let's assume that the pump at CTB A on the top branch is pumping. What happens if the pump at CTB B kicks on at the same time? And if the one at CTB C kicks on as well? Do pressure and volume both double? Is this situation different for centrifugal pumps than it is for PD pumps? Basically, I need to know what the worst-case scenario (all pumps running at once) for this pipeline will be, so that I can size it accordingly. Just like the "flush-test" major football stadiums must be designed for. Any help will be greatly appreciated.





RE: Multiple pumps in a pipeline?
Case 1: 100 GPM of water to destination - Calculate pressure drop
Case 2: 200 GPM of water to destination - Calculate pressure drop
You can start at the back end with the highest flow and calculate the pressure required at each pump.
There is a system resistence line for each case.
Then each pump at its location has to be analyzed against the potential pressure at the pump discharge.
To answer your main questions:
The pressure will not double it will increase according to the flow and the pump will move along its curve to match that pressure
For two pumps of 100 GPM in parallel the capacity will be less than 200 GPM due to the increased head of the system.
There are various software tools that can be used for such analysis. You may want to ask if your company has any of this software.
A progressive cavity pump is usually used to pump high solids liquids or viscous liquids. Ask around if this is needed for your application.
Good luck!
RE: Multiple pumps in a pipeline?
I'd likely start with the estimates of how much produced water each site is expected to produce over a day. Using 100 gpm (or whatever pumping rate you want to consider) see how long that pump has to run. That will allow you to determine how many sites should be assumed to be trying to pump into the pipeline system at the same time. If it takes 6 hours for example for a site to pump their produced water into your collection pipeline, you will need to plan for several sites to have to operate at the same time simply because of the number of sites you have. If the time though was for 1 hour, you might decide to design the collection system for 2 sites operating and thus it would be smaller and less expensive. Discuss this design basis with the affected groups for input and once you have agreement, you can start on the details of the pumps and pipeline sizing.
You likely won't be able to go with the same pump at every site because of the differences in pipeline lengths and necessary discharge pressure but if you can standardize on a few sizes, it will be an advantage in terms of spare parts. This likely is a good application to look at VFD drives on the pumps as well to save energy. If you design the system for example for 3 drill sites all pumping at the same time but only one is running, the discharge pressure will be significantly reduce and there is no point just wasting power across a control valve.
Also, at the end of the day give the uncertainity in the produced water estimates and that wells tend to produce more and more water with time, I'd upsize the design basis to give you future capacity. Alternatively, management may decide to just install what is needed today and then install a parallel pipeline in the future if they need more capacity. These type of designs are iterative in design, you won't go in a nice path from A to B to C to D, there will be recycle and reworking previous work.
RE: Multiple pumps in a pipeline?
In the middle of typing this, I just realized that I could just run a safety relief line back to the tank at each CTB, which would direct the water right back into the tanks if pressure in the pipeline reached, say 600 psi, for example.
So I guess my next question is solely about pumps. I have heard that centrifugals have a "sweet spot" so to speak, and don't handle changes in pressure very well. PD pumps, however, don't care what pressure is, and always flow the same amount of fluid. So for this reason, would PD pumps be my best bet? If at all possible, I'd like to make this as "cookie-cutter" as possible. I'm looking for something that we can buy in larger numbers, and that will be able to handle the situation at each node of the line, without being too overkill. However, I also realize that that is not always the case, and that this may be a little wishful thinking!
RE: Multiple pumps in a pipeline?
Weatherford sells a "constant pressure valve" that keeps downstream pressure constant by dumping excess back to the tank. This valve wastes considerable energy (it always dumps some water, sometimes it dumps a lot of water), but for produced water I find them essential. I like them a lot better than VSD on PD pumps.
The PCP (generic name for the technology behind Moyno) is just about the most expensive option, but they handle varying discharge pressure better than anything else and they have a lot less pulsation than a plunger pump.
I would probably be looking at centrifugal and a VSD myself.
David Simpson, PE
MuleShoe Engineering
"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
RE: Multiple pumps in a pipeline?
RE: Multiple pumps in a pipeline?
Speaking of that, IMO the system is way overdesigned. You will wind up operating each pump only 2 hours per day. Considering that the equipment could really run 24 h, this is an extremely high overinvestment factor of 12. Very high.
Considering that extreme under-duty capacity that the system is being sized for, I see no need at all to increase your initial 2400 BBL capacity by 50% to 3600 BBLS. I don't think your boss is doing you any favors by telling you to overdesign the system by such an amount. You are worried about the "sweet spot" of the pumps, but miss the full-capacity optimum design of the system by 12.
I think you should design the optimum system first, then you can see how much material and power all the fat will really cost you. I think that would add a lot of value to your experience this summer.
I would also suggest a star pattern, bringing the farthest wells to a central point amongst them, there combining flows in a larger pipeline running to the disposal well, that in conjunction with the closer 6 wells each having their own pipelines running directly to the disposal well, provided the central area of your system would not be directly under a farmhouse, or something.
The problem with the "U" configuration you have there is that it will be difficult to buy 11 of the same pumps. The heads reqired for each one will be quite different and buying 11 of the same models may be impossible. I think a starburst pattern would help that problem a lot.
If it ain't broke, don't fix it. If it's not safe ... make it that way.
RE: Multiple pumps in a pipeline?
3 x 4" and 2 x 6" doubles the pressure when all pumps are running, but the two farthest pumps could still run with the same 228 discharge pressure of the 2x4" & 3x6" config.
4x4" & 1x6" needs 700 psig with all pumps running, and still needs 300 psig with the farthest 2 pumps running. 3 farthest pumps running needs about 550 psig.
A starburst to the farthest 6 wells would only need 300 psig if you use a 6" trunkline, and you could connect the other wells direct to the disposal well with shorter runs of 3 or 4" pipelines.
Those are using your 3600 BBL, so you might not need more than 2 sticks of 6", if you went back to 2400 BBL and used your unused operating time as your safety factor instead of bumping up your base flowrate soooooo much.
If it ain't broke, don't fix it. If it's not safe ... make it that way.
RE: Multiple pumps in a pipeline?
The philosophy is that a pump will try and pump into a main on a timer basis. if the pressure is above the set pressure the pump trips. Basically the pump is not allowed to start because other pumps are running. When the running pumps have evacuated their sump the other pumps can start.
Controls are all local tot he pump.
Companies such as E One have the control boxes to operate such a system. There is design is fairly rudimentary using a spreadsheet approach only.
“The beautiful thing about learning is that no one can take it away from you.”
---B.B. King
http://waterhammer.hopout.com.au/
RE: Multiple pumps in a pipeline?
If it ain't broke, don't fix it. If it's not safe ... make it that way.