What I see that is unusual is that the cavitation forms behind a vane but stays up on the impeller eye. This says to me that there is insufficient flow velocity through the impeller to carry the bubble down through the flow path. Perhaps this pump is operating too far to the left on its pump curve and is overheating.

What does this sit into?

Got a sectional drawing of the inlet chamber?

Does the flow then turn 90 degrees and exit from the outside of the impellor?

Bit difficult to tell from only one photo tbh.



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

TugboatEng:
Thank you for your reponse. That is a good observation. Thank you!

LittleInch:

The process fluid is jet water and pump is used in decoking process of refinery, where the remains of crude oil from distillation is refined further (coker unit).

Below is the pump sectional highlighted impeller shown in picture.

I can't really see the impellor wear ring surface but it looks heavily scored implying considerable flow perhaps from the outlet side back into the inlet.

Flow into the eye of the impellor looks quite complex.

A lot will depend on how the pump was operated, fluid conditions, temperature, pressure flow etc.

Was it allowed to dead head for long periods?

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

Suction recirculation cavitation damage.
The pump is pobably running away from its BEP.

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.)

The damage along the roots of the vanes looks like cavitation.
The damage to the rim looks like it could be erosion or excessive recirculation (way off of BEP and or end play).
The hole in the mid-vane is a typical casting defect with wear piled on top of it.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

Thank you all for replying.

The application is application. Can you recommend a material for this impeller that has cavitation resistance against this kind of damage and has high corrosion resistant. In other words, what material will resist this kind of damage the most? Edstainless?

Cavitation resistance? The 300 series stainless steels have very good cavitation resistance. If you need more you're going to need to go with rubber. What temperatures are we talking?

Tugboat:
Thanks for quick response. Operating temp is max 110F

Before going down the road of material change you should carefully evaluate the pump performance capabilities against the actual operating conditions. That means a thorough controlled accurate performance test at and around the performance conditions compared to any documented factory testing.

For me, a material change is probably unnecessary as it is likely a hydraulic performance problem.
My suggestion is to contact the pump manufacturer for advice on the application.

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.)

What is the current material?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

From the picture shared, it does not look like cavitation effect. Kindly look for these possibilities for assessment .
1. Water recirculation due to low sizer impeller fitted in casing. Please check cut water length and impeller dia.
2. Check for impeller material. Non-standard impeller can cause this problem too.

Artisi:
The operating point will stay the same. You can call it bad original design impeller or necessity to run the pump off from BEP. I know it will happen again but what material will resist this kind of damage the most and last the longest out of all materials?

EdStainless:
The material of impeller shown in above picture 17-4 PH SST.

Rubber is the most cavitation resistant material. 110°F is acceptable for a rubber coated application.

But ,if you're operating far to the left on the pump curve there may not be sufficient flow for cooling local temperatures may be higher

who is the pump manufacturer, what model is it and what duty was the pump selected to meet and what is the actual operating duty, how long has the pump been in operation, is this a new or ongoing problem?

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.)

So this is the cast version of 17-4PH, what is the heat treatment?
It should have been solution annealed and then aged, what is the aging temperature?
The edges look like erosion, which is a different animal to deal with.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

EdStainless: agree erosion appears in many places and was noted for later discussion.

Establishing actual performance against selection is the first step, once known it will eliminate a lot of questions and point to a likely resolution, however at this point I don't see it solely material problem.

Changing material just for the sake of guess work is like closing the stable gate after the horse has escaped - the question is "why was the gate open" and will go round and round for want of meaningful information.

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.)

Artisi ! Looking at impeller's condition, it can be either material's issue or design issue. Other possibility could be operational fault (discharge valve remains open and causes water to circulate inside casing) which is rare possibility.

Daren Samy, following a life-time in the pump industry involved with sales, application, design, trouble shooting it is my experience that the majority of pump problems result from wrong hydraulic selection or wrong application or duty change without considering the changes on the pump hydraulic capabilities. Material problem for a correctly selected pump is rare, and a design issue from a reputable manufacturer is extremely rare.

At the moment, we have no manufacturer, application data, performance curve versus actual site operating conditions etc - therefore its all crystal ball gazing and spinning wheels.

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.)

Artisi:

Sorry for late response. It was a crazy weekend for me. Here is the response to your questions asked on Oct 12th.

-Pump manufacturer is Pacific Pumps (now called Flowserve).
-It is diffuser type barrel pump with element/bundle inside the barrel (API designation BB5).
-I don’t have actual operating conditions because we don’t monitor suction pressure and flow rate. I can look into providing recommended duty point and pump curve.
-Pump was repaired last time in 2018 and this impeller was replaced.
-This is an ongoing problem.

There is nothing about the material that suggests to me that changing alloys would fix this.
It is either an external issue (flow/pressure wrong for this pump or internal (impeller clearances or position wrong).
There should also be some fairly serious damage at the inlet and/or outlet of this impeller.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

Quote (mech5656)

I don’t have actual operating conditions because we don’t monitor suction pressure and flow rate.

Unbelievable. You have a machine costing many thousands of dollars and you haven't got a measurement of pump inlet/suction pressure or any sort of flow measurement??

No wonder it's an ongoing problem.

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

Heavy recirculation of fluid in the casing that has chopped off the material. This recirculation can be caused by not maintaining the clearances properly/design issue/operational fault.
Mech5656 ! Needs more data especially Pump overhauling data to analyze.

Must agree with LittleInch, it is unbelievable that for the want of and the cost of a couple of quality pressure gauges and a flow meter you are prepared to run the pump knowing there are ongoing operating problems.
I also agree with EdStainless that a change of material will not overcome what appears to be poor operating conditions.
Pacific Pumps is a well established manufacturer so you can assume the manufacture and hydraulic design are valid as is the materials selection provided correct data was given originally for pump / material selection and this data hasn't changed over time.

Again as LittleInch has said "More details = better answers"

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.)

Decoking water pumps typically operate in a batch process. The pump is cycled from Fill (lower head, high flow), Cut (rated flow, high head) and Recirc (low flow, very high head). The damage is probably occurring when the the pump is recirculating to the tank for bit switching. You can either lower the suction specific speed or increase the size of the recirculation orifice to increase the flow during this part of the cycle. I have used both of these solutions for this exact problem on decoking water pumps. Artist nailed it. Suction recirculation cavitation from low flow operation during bit switch.

Johnny Pellin

Artisi, JJpellin, LittleInch, Darn Samy, Estainless & Tugboat,

Thank you all for your comments and feedback. You are all very knowledgeable and experienced.

Regarding process operating conditions, the process is the process and nothing can be done. The right thing would be to replace the whole pump with new design and that will take a lot of money and time. The second best thing is to replace the first stage impeller (shown above) everything 4 years (as happened here). Note: only first stage has this issue due to suction impeller.

A new design is unlikely to cure your problem - fixing the way the pump is applied is the better way to go. Reread what Johnny Pellin said about recirculation line.

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.)

Mech5656 ! Replacing pump would not solve your problem. Instead , dug into root cause of issue. Check the operating conditions, maintained clearances etc before marching for any step

If you have to replace the first stage impeller to repair the damage, install an impeller with a higher NPSHr which results in a lower suction specific speed. I have used this exact solution for this exact problem on this exact service and proven over 30 years that the solution worked.

Johnny Pellin

Quote (mech5656)

the process is the process and nothing can be done.

Sorry, but I just can't believe this. There is always something that CAN be done, the question is if you want it to be done.

If the key issue is low flow then either revamp or introduce some sort of recirculation loop - look at ARV's or similar.

You need to look more closely at the various flow and pressure conditions that this pump sees on a regular basis and amend the process design to accommodate it without trashing your pump on a regular basis.

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

Mech5656, it is also highly likely that operations have changed the conditions over the years.
You know they want more output or efficiency.
And they don't care what it does to things like pumps.
You need to monitor operation conditions carefully.
You will have a lot better chance selling a system modification if you can show that it is because conditions have been changed.
And you need to get more involved in the pump re-build.
I know that we like to leave this 'to the experts' because inspections and audits cost time and money.
But this a case where minor variations in clearances and end play can have a large impact.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed