Need a Plunger Pump primer (sorry!) 4

[Guest102023]

I am grappling with the failure by cracking of a cast steel plunger pump casing (see attached photos). The mechanism is obvious enough, but I want to get a better understanding of the basic operating principles of these pumps. Links to internet resources would also be useful; I've done much googling, finding lots of operating manuals and parts dealers but not much theory. I understand these pumps are subject to considerable vibration, so the possible effect of that is of particular interest.

Thanks in advance.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[DubMac]

Not sure what Search Engine you're using......but try Googling " Plunger Pump Theory"; you will find plenty of nuggets. Much of the available theory and pump basics are stowed away on the manufacturer's websites, so it may be a little tougher than cutting and pasting links, but its worth the effort.

Think of the Triplex Reciprocating Piston Pump in two parts; you got your power end, and you got your fluid end. Your power end converts the rotational motion of the motor into the reciprocating motion required by the pistons. Due to the nature of this mechanical process, there is inherent constant acceleration or deceleration of the piston rod; it is never at constant velocity. This not only adds peculiarity to the vibration signature, but also complicates the fluid hydraulics on the suction side of these pumps.

The fluid end is assigned the task of introducing the low energy, incoming fluid, to the high pressure, outgoing fluid. The most common of "recip" pumps, and looking at yours, it is most common....... use energy from the piston's suction and pressure strokes to open and close suction and discharge valves. The fact that there are (3) pistons in a triplex pump, and that they are all 120 degrees off from each other in phase, produces pressure pulsations = vibration.

My wife just called me to dinner, I gotta go. There are plenty of knowledgeable folks that can finish off what I've started. Good luck to you.

 

[Guest102023]

Thanks for one little word - 'theory', it made all the difference in googling. (Seems this type of workhorse pump wasn't sexy enough to be taught back in Mech Eng.)

I expect that pressure increases at each stage(?) - that certainly would not help the vibration problem.

My question now is (and I have only been shown the part you see in the picture): is the surface that is cracked exposed to process fluid? Which is a mixture of BFW, condensate & stripped water (~50% of mixture) from the sour water stripper; the stripped water has a few ppm each of H2S and NH3.

Anyone have maintenance / failure analysis experience with this refinery equipment?

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[micalbrch]

Your photo shows the fluid end, the cylinder, the wet end - whatever you like to call it. The cracked surface itself is not in direct contact with the fluid. It is the back side of the cylinder, the side which faces towards the gear. The holes you can see and the studs are for the stuffing boxes. The stuffing boxes carry the packings which seal the plungers.

The pressure does not increase at each stage. There are no stages. One Rotation of the crankshaft means one suction and one discharge stroke of each plunger. The plungers run with a 120° crank offset.

Vibrations can occur for several reasons: Worn or damaged suction or discharge valve (in this case there will be backflow and non-uniform output from the pump), broken pulsation dampener bladder (if there is any installed), too high suction pressure that opens the suction valves too early. Can you tell us a bit more about the vibrations?

Triplex plunger pumps create pulsations, much less than duplex or simplex piston/plunger pumps but still there are pulsations. A pulsation dampener can reduce these pulsations down to whatever is required. It depends upon stroke length, plunger diameter and pressure whether or not a dampener is required at all but your photo seems to be from a rather big pump. So, I would definitely recommend a dampener.

 

[ccfowler]

It may be wise to pay some attention to the system pressure characteristics with respect to the rated operating pressures for this pump. Nasty things can happen when there are times of unexpectedly high pressures.

Also, there may be some issues introduced by the actual geometry of the pump's crank linkage. The plungers do not move according to simple harmonic motion. Somewhat surprisingly high accelerations may be found at some points in the cycle. These are generally very rugged machines, but all do have their limits for endurance.

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[Guest102023]

Only one picture was uploaded, here is the other. Hard for me to understand that the cracked surface is not exposed to fluid; it is part of the chamber above the lower valve. I am trying to determine where the pitting I see around the crack origin are cavitation or just corrosion.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[Guest102023]

Seems I am only permitted one pic at a time. Another -

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[micalbrch]

Your origin photo shows a different view or I completely misunderstand it. Anyway, I do not think that this is a corrosion problem. "The chamber above the lower valve" is the area where cavitation in a plunger pump has the biggest effect. I cannot see the pittings which might be a problem of my computer but corrosion should occur everywhere in the wet end.

 

[Guest102023]

Here's a pic near the crack origin. Looks as much like pitting corrosion but then my experience with cavitation is more with hydro turbines than pumps.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[Compositepro]

It is possible that the cause was a defect in the original casting.

 

[DubMac]

Absolutely the crack is in communication with process fluid. From what little I can see, I would agree with Compositepro; casting defect very likely,porosity etc. Could also be that the pump exceeded its Max Working pressure, or surfaces weren't machined true before mating up. A crank going out of line could do it as well. Would be easier to guess if we knew how it crashed.

Cavitation damage typically appears more like erosion, uneven pitting like moon craters or acne.

 

[Guest102023]

Already ruled out. Have a look at the attached, which I believe shows cavitation above a band where I think the valve seat was.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[DubMac]

Maybe its just my lyin eyes, but I'm not seeing anything that looks like cavitation damage. Looks like two machined surfaces and some raw unfinished casting surface. I wish I could make better sense out of the picture.

Your cracks are not caused by cavitation.

 

[ccfowler]

My guess is a material flaw or a weak point where failure initiated during a high pressure excursion. As I noted before, these are generally very rugged machines, but that is because they need to be very rugged.

Based on a failure analysis of a plunger pump many years ago, I suspect that your pump likely experienced an unfortunate momentary pressure peak. While this failure appears to show a material flaw, it may actually merely represent a normal material weak point, and the underlying problem is external to the pump. I would want to check the connected system for something that could have caused a pressure excursion at the pump to avoid a repeat of this problem in the future. Plunger pumps are very much subject to the general reality that apparent pump problems are likely to be merely the symptom of a problem originating in the connected system.

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[LittleInch]

If this is an isolated failure I think you may just have been unfortunate. Hair line cracks like that in those types of locations are not fatigue or cavitaiton, but would appear to be an un found cast defect.

a slice through the crack may reveal somehting, but i suspect this a metalurgists expertise not a piping / pipeline engineers realm so you way want to link a question on one of those forums to this one and see what they think.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[Guest102023]

Its not a fluke or one-off, and there were no casting defects found. I'm a metallurgist and not a piping or rotating equipment engineer so that's why I posted here.

This was fatigue, with lots of corrosion on the fracture surfaces. Pitting at the origin acted as stress raisers. Anyway, I think I've sorted this one out, so thanks to everyone.

p.s., I'm not sure why so many people assume that castings must all have defects.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[LittleInch]

So can you tell us?

Information exchange works in both directions.

I don't assume casts in general have defects, but without all the knowledge you have on your own bit of kit then it's a good place to start. Plunger pumps are like any recicrocating unit going to generate vibrations and pulsing and may then fail by fatigue.

WOuld be good to know your conclusions.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[DubMac]

SO many people assume casting defects because so many castings are now poured in crap foundries with little QC in China, India, & Mexico. Was not the assumption 25 years ago when almost all engineered castings for rotating eqpt. were poured in the good old USA.

 

[desertfox]

Hi

page 25 of this paper looks like the area where the packing glands are on your pump case, it gives a reason for erosion in that area etc so it might be worth while finding out about the operating conditions the pump was under prior to failure.

http://www.tekna.no/ikbViewer/Content/22983/Steve Digby pumpeteori.pdf

 

[Guest102023]

DubMac,

That is true but I've heard this kind of stuff for many years.

I suspect it comes from engineers who were obliged to sit through one materials or manufacturing course and mentally link cast steel with cast iron. As they say, a little knowledge is a dangerous thing.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"