Pumps in produced water treatment 2

[Alexander R.]

I'm looking for the information about what types of pumps are commonly associated with produced water treatment systems/facilities. I was looking for the information elsewhere, but could not find much. I'm interested to know approximate % of each pump type used for these applications.

Thanks in advance.

 

[Artisi]

End suction process pumps or horizontal split case for higher flow rates. Many manufacturers offering high quality equipment. Are you required to meet any particular specification, search Goulds, Weir, Sulzer and many others on the net.

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.)

 

[LittleInch]

Why?
Most will be centrifugal.

There's nothing particularly special about produced water.

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

 

[mk3223]

Typically for the operation plant, the common types of pumps are either centrifugal or positive displacement pumps. IMO, the number of the centrifugal pumps is more than the positive displacement pump. The actual percentage may be depended on the process needed from one plant to the other.

 

[Alexander R.]

Thank you for the answers. Sorry, if I didnt specify, but I'm looking at produced water treatment in the oilfields. I'm consulting NORSOK standard, and there are couple of places where it instructs to use low shear type pump. But this definition is rather wage, in terms what are the criteria to qualify pump to be low shear. So, from this point, I wanted to hear what pumps operators are choosing for this application.

 

[georgeverghese]

Low shear centrifugals for this application are typically those operating at <1000rpm, or better at 800rpm if you afford the extra CAPEX. Keep pumping differential head as low as possible and go for the higher hydraulic eff models if you can. Theoretically, a PD pump is much better at minimising shearing effects, but a very bad idea in actual practice - these include Moyno pumps (or any other peristaltic), gear pumps etc.

 

[Alexander R.]

isn't a low RPM and low differential head kind of opposite operating conditions for high hydraulic efficiency?
And could you elaborate a bit more why Moyno pumps (and other PD pumps) is "a very bad idea in practice"?

 

[Artisi]

why low shear on water or are we missing something?

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.)

 

[Alexander R.]

As I understand there is usually a rag layer of oil that need to be separated. A high shear pump would shatter droplets into the smaller ones, which would be harder to separate in slop tanks or hydrocyclones.

 

[rotw]

can we translate low shear to be simply low flow (as an over-exageration)?

 

[georgeverghese]

On the contrary, you should be able to find a pump vendor who can offer a pump with a curve that gives you reasonably good eff at the duty point you need.

All PD pumps are a bad idea for produced water service - abrasive solids in the feed will tear up just about any type of liner you can throw at it.

 

[Alexander R.]

Thanks for the answers. We deviated here slightly from the topic, though. I was wondering if there is a place I could look up some statistcs of pump types used in specific branches/applications.

 

[LittleInch]

None that I know of. No one cares enough to spend time and money collecting the data.

That data lies in the head of an experienced engineer familiar with the particular aspect you're looking at.

NORSOK is fairly notorious for being at the high end of what is required - which standard and section?

From my experience you seem to be looking at a particular way of handling produced water which may not be the standard method, but the description of "produced water treatment in the oilfields" is rather vague and wide in application.

Normally produced water is stipped out in a facility to various levels of separation. Then you either process it a bit more so that you can dispose of the water without too much oil in it or you don't care very much and just inject it back into the reservoir. either way, use of a low shear pump can be good in certain places, but it's far from a common requirement in my experience.

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

 

[Alexander R.]

Standard is NORSOK P-002, section 22 - Oily Water treatment. Standard instructs to use low shear pumps and valves upstream the treatment system to avoid shearing and break-up of oil droplets.
Regarding water treatment, you correct in some way about handling it. There are three ways to go:
- clean the oil down to certain ppm (29ppm in North Sea) and dispose
- clean to reinject produced water into the reservoir. In this case, the requirements to water quality is actually even stricter in certain aspects, since some contaminants might react with the reservoir rock, and deteriorate the injectivity, leading to problems.
- clean to dispose into another formation. Again requirements are quite strict, since some contaminants might deteriorate the injectivity.

I know that some research on shearing characteristics of different pump types was done in late 80's, early 90's, specifically by Flanigan. His conclusion was that PCP type is best, and PD generally performed better than centrifugal type. I guess, majority of oil companies use PCP pumps in water treatment systems. But I could not find any infographics on that, whatsoever.

 

[LittleInch]

I get that and I was probably being a little simplistic - injection water often needs to be virtually oil free.

I think the thing I have a little difficulty with is "pumps and valves upstream the treatment system" Normally to pump oily water, which seems to be what you're describing, this needs to have gone through some sort of treatment or process to get to that point and this usually is undertaken using well pressure. Why do you need a pump?

Is this offshore or onshore field?

PCP are great, but quite long, not cheap and have a limited capacity so you end up with a lot if you're not careful.

A slow centrifugal as noted above would probably be cheaper and have a higher flowrate and do much the same thing.

"I guess, majority of oil companies use PCP pumps in water treatment systems" - News to me.

I don't think the information you're looking for is freely available I'm afraid.

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

 

[Alexander R.]

I am mainly looking at offshore applications. I agree that most of processing facilities are utilizing the reservoir pressure to avoid additional pumping. However, there are still some stream that need pumping, such as recycling streams. There is a good overview of treatment schematics for typical North Sea and Gulf of Mexico platforms here

https://doi.org/10.2118/159713-PA

Perhaps I am wrong about my perception of pump type use in this application, I just got that impression.

 

[rotw]

have a look here

http://petrowatertech.com/wp-content/uploads/2015/08/SPE-134505-Walsh-Frankiewicz2010.pdf

it seems Shell company has made some sort of survey and most importantly has summarized selection guidelines; see page 10.
You can take it from there.

I suggest to translate those guidelines in a selection diagram for pump types, you wont get percentage but which pump type most likely satisfies best the guidelines. I am looking to see if I can find such a diagram;

 

[Alexander R.]

Yes, I have read this paper as well. Though would like to mention that similar conclusions were drawn in this paper, which dates back to 1994.

https://doi.org/10.2118/28815-MS

I'm not sure how you suggest I should use these guidelines to the selection of pump types. For me it seems like these guidelines were specifically addressing centrifugal type pumps.

 

[rotw]

I think the article suggests centrifugal as guideline. It states for instance that PD are to be avoided due to their pulsative effects on hydrocyclones. If you want to dig dipper into centrifugals, then my suggestion was to translate the guidelines into a selection chart.

I tried to post a diagram that I fished out of the internet to identify some trends.
Based on the Shell guidelines, it seems that we are looking at pumps with a flowrates in the range of roughly 200 gpm, most likely multistage pumps are expected (e.g. Worthington WD pumps). Clearly there are multifactor entering into play and this should be a case by case problem - but as far as typical trends are concerned this would me my ball park guess.

 

[Alexander R.]

Thanks for this suggestion to look from this angle into my problem.