Recommendations for discharge piping ( centrifugal pumps) 2

[PedroBC]

Hello,

I am trying to find recommendations for discharge piping fittings installation ... Can you guys give me any help ?


For pump suction i found the L > 10D (L = Length , D = diameter) the length required to install any kind of fittings ( curves, valves etc...).

Any recommendation for pump discharge piping ??

 

[pennpiper]

The Discharge piping of a pump does not require any straight run piping like the alleged 10D claimed by some people for the Pump Suction piping. That 10D issue is still in the hands of the jury and the jury has been out for a long time.
Also, there are so many different types and styles of Pumps it is impossible to make a simple statement about Pump piping and have it apply as Law for all pumps.

Now to the question about Pump Discharge piping. The important issues for the arrangement of all pump piping is Installation, Operation and Maintenance.
All piping should be arranged (routed) so:
- there is flexibility to ensure low stresses on the nozzles
- the Suction and Discharge (S/D) valves can be operated
- the Strainer Basket can be removed and cleaned
- the piping between the Block Valves and the Pump nozzle can be removed when required
- the Pump Case can be removed with a minimal amount of effort or dismantlement of other piping or equipment
- there is proper access room around the pump for people to move and work

I suggest, if you can, take a trip out to any operating plant and spend some time looking at pumps and the pump piping. Take pictures (with permission of course) take notes and then, back in the office you can study these along with others.

prognosis: Lead or Lag

 

[C2it]

Discharge piping ... with due regard to flexibility, avoid locating heavy items like valves such that they cause eccentric loads (moments) on the nozzle. Remember that if you use software to size a spring above your pump, the liquid column above the nozzle will act on the pump, not the spring.

Regarding suction piping ... you will practically never see 10D up-stream of a pump suction offshore ... space is just too expensive. Most pumps have a vertical supply from an overhead manifold, with an elbow adjacent to the pump mating flange.

 

[StressGuy]

"Most pumps have a vertical supply from an overhead manifold, with an elbow adjacent to the pump mating flange. "

I don't think I've ever in my life seen such an arrangement in any of the refineries or chemical plants I've been in. I've certainly never seen an elbow directly on a pump suction except for one special case and that elbow had turning vanes installed in it.

API RP-686 3.1.2.6 calls for 5 pump diameters of straight run into the suction before the first valve/fitting/etc. That's what we typically go by for our pump system layouts. The main point of controversy we run into is does that include the reducer. Tradionally the reducer is up againsts the pump and many of our clients maintain that preference, even though API includes it in the list of items that should be outside the 5D window.

I've only run into one case where the pump vendor called for 10D and that was a million dollar multi-stage very high delta-p pump.

Edward L. Klein
Pipe Stress Engineer
Houston, Texas

"All the world is a Spring"

All opinions expressed here are my own and not my company's.

 

[rmw]

StressGuy,

Come on out to Channelview and I'll let you view a bunch of pump skids with vertical reducing ells on the pump suctions. When C2it suggested that, I assumed that he meant for horizontal split case double suction type pumps. I design our units that way just for the reason C2it stated, we don't have the real estate to have the luxury of having the required straight run piping required for horizontal entry. I think I got the idea out of Goulds GPM Manual back when it was at edition 2 or 3. GPM is at editon 9 now fwiw.

rmw

 

[Artisi]

HSC pumps are Ok with bends at the inlet as the pumps have the advantage of a straightening effect from the pump inlet flange upto the impeller eyes. For an end suction pump you don't have this luxury and problems can arise with bends fitted to the pump inlets.

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.)

 

[StressGuy]

Fair enough, I can buy that one. I don't think I've personally worked with double suction style pumps. I also see that I missed the "offshore" in C2it's post.

Edward L. Klein
Pipe Stress Engineer
Houston, Texas

"All the world is a Spring"

All opinions expressed here are my own and not my company's.

 

[davefitz]

The title of the post is related to discharge piping , not suction piping.

Typical pump outlet pipes are "undersized" with high velocities- the piping connected to the outlet nozzle should have at least 1% chrome to minimize the erosion/corrosion due to high velocities for 10D downstream lengths unless there is added a pump outlet recirc fitting in that intervening length, and the reducer to the next larger pipe size ( to reduce average velocity below 15 fps) to be installed within those downstream 10D. If you skip this step, then you will need to annually UT the wall thickness of the piping to confirm adequate wall thickness remains for safe operation.

Suction piping has other issues, mostly realted to NPSH problems assocaited with excessive suction pipe loss, downcomer vortices , and if saturated liquid is being pumped, then issues of downcomer steaming during sudden pressure transients also becomes an issue. Meeting pump nozzle allowable loads is another pipe flexibility issue. Finally , many pump failures are related to the use of economy couplings between the pump + motor, or incorrect foundation grouting.

 

[Artisi]

Why are typical outlet pipes undersized, please explain.

Any engineer worth his pay packet would size discharge pipe work to minimise friction losses, therefore undersizing defeats the purpose.

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.)

 

[pennpiper]

Davefitz & Artisi,

You are both getting away from the original poster's question and you both seem to be inventing your own terminology for some things.

The term "outlet piping" is not what the piping engineering and design community and the Pump manufacturers use. The correct term is pump "Discharge piping".

I think you will find that the pump discharge piping is properly sized in virtually all cases. The Process Engineer for the Unit is responsible for the line sizing and I am sure they do it correctly.


prognosis: Lead or Lag

 

[Artisi]

pennpiper,

Yes we may have got away from the subject, but don't most ET posts.
After 30+ years in the pump industry I well know that discharge pipe/ piping is what it's called.
I was replying in kind and asking for clarification to what seems a very strange statement.

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.)

 

[DubMac]

Pump discharge pipe is sized to match the pump's discharge flange; or larger in some cases. The pump's discharge flange size is a function of hydraulic design; almost always designed into the casting and can't be changed. Smaller discharge pipe would act as an orifice and waste energy.

I guess the answer to this post is: Just make discharge piping the same size as the pump's discharge flange and you'll be fine.

 

[davefitz]

DubMac,

"Just make discharge piping the same size as the pump's discharge flange and you'll be fine" --Not.

There have been several fatalities at power plants in the last 20 yrs related to failures at pump discharge pipes , related to unacceptable erosion/corrosion at the downstream elbows- part of the problem is that the optimum discharge size for pump dynamic purposes is not appropriate, and would be undersized , for piping erosion considerations and also for "economic pipe size" considerations.

Perhaps if you are referring to a household heating boiler recirc pump, then matching piep sizse might be OK. But in many large industrial applications, it is not.

 

[DubMac]

Yes, agreed it must be larger in some cases, maybe one or two sizes up at most; definitely needs to be left up to the pipers. My point is that it should never be smaller than the pump discharge.

 

[Artisi]

Discharge pipe diameter is based solely on friction loss and cost considerations, it has nothing to do with flange sizes.

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.)