Worn pump shaft 7

[replica]

Hi all,
I am new to this forum and would like to ask whether anyone had experience on worn nitronic 50 pump shaft used in VS6 pump. The length of the shaft is 4.35m and pump is 19 stage pump used to transfer crude oil offshore. It was observed that all 19 locations where the bowl bush bearings were placed on the shaft shown the severe wear on the shaft surface and not on the bronze bush. The length and locations of the worn shaft region has similar length as the length of the bush. The clearance between bush and the shaft was self lubricated with the transferred crude oil. The worn shaft at the first and second stage shows severe wear with smooth and shiny surface (mirror finish) where else the worn shaft in contact with bowl bearing at other stages have rough surface. The shaft which is in contact with the line shaft bearing also showed a rough surface and some are not concentric (uneven wear). The diameter of the shaft is 1.5 inches. The bush bearings are made of C836 bronze and the shaft is made of Nitronic 50. The hardness of the shaft is 44HRC and the hardness of the bush is 90HRB. All the bush bearings observed have either no sign of wear or minimum wear and tear. This problem happened about a year ago and kept on reoccurring even though the shaft and bush bearing were replaced during maintenance using the same new spare parts. The longest the pump in operation after maintenance was for two months and the shortest was two weeks before the same failure happened. The shaft and bush bearings can normally last for a year. The main issue is why the shaft worn out and not the bush bearing. Normally from what I understand, the bronze bush should wear out and not the shaft. Any suggestions for possible causes of the worn shaft are highly appreciated. Is it possible for softer bush to abrade the harder shaft and under what condition?
Attached herewith are the photographs of a portion of 4.35m pump shaft showing worn shaft region and the unaffected bush bearing.

http://files.engineering.com/getfil...edd-b0e2-753650ac6f94&file=wornshaftforum.pdf

Regards,

 

[MikeHalloran]

I might speculate that the bushes are entrapping sand particles from the crude oil, and lapping the hell out of the shaft.

Is there some way to feed the bushes with less dirty oil?

Does the pump manufacturer offer softer, e.g., plastic, bushes?





Mike Halloran
Pembroke Pines, FL, USA

 

[EdStainless]

I agree, it looks like trapped solids.
Perhaps some relief so that there is more flow through them.
What is the clearance on new ones?

This is why we rarely ran bushings directly on shafts.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[replica]

Thank you very much for the feedbacks,

MikeHalloran

Entrapped sand particles, that is what i am thinking as well. Anyway how is it possible that the particle only polished the harder shaft and not the soft bush? Is it because the shaft is rotating and the bush is stationary that made this possible? When I looked at the bearing inner diameter surface under scanning electron microscope, I could not find any embedded particle on the surface.

It is almost impossible to lubricate the bush with other means but we can find out and probably filter the particle.

Yes we can try to use softer bush but how this will help to solve the issue? from what I understand the plastic bush will swell under H2S, hydrocarbon and heat the may cause the shaft to wear out, but we can try.

EdStainless
The clearance on the new one is 0.008 inch for 1.5 in shaft diameter. When we increased the clearance we found that the pump life improved, to date the pump had been operated for almost two months without failure. We will monitor the progress of the pump. The issue is can we increase the clearance beyond the maximum recommended value?

This is why we rarely ran bushings directly on shafts.
EdStainless ..do you have another method of running the bush not directly on the shaft?

 

[Artisi]

Can you change to a fluted, cutless rubber bearing in a compound compatible with the pumped product, this could be your best option.

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.)

 

[replica]

Artisi

I am not familiar with the rubber bush as you stated but we had tried vesconite but not at the same well and found out that the bush broke in brittle manner (may be due to reaction with the crude oil) but with less wear on the shaft. However this is not really representative as the crude oil properties at that particular well probably had different properties ie less particles, etc.

Thank you for the suggestion and will look into that option. Anyway I really appreciate if you can provide more information, links etc regarding fluted, cutless rubber bearing as suggested.

For a time being we are monitoring at the pump with increased clearance. Recommend clearance is 0.004 to 0.006 inches and we used 0.008 inches which already beyond the recommended value.

Thank you

 

[MikeHalloran]

The traditional method of lapping is to embed/crush ("charge") an abrasive into a softish material like brass, and rub the now abrasive face of the brass on whatever hardened steel you are trying to polish. I'm thinking that maybe a filled plastic, e.g. Rulon, would allow the particles that you can't avoid to embed themselves a litle deeper than does the bronze, and maybe eventually even bury themselves so they stop wearing the shaft.

The 'cutless' bearing that Artisi mentioned is often found running against long propeller shafts, totally lubricated by whatever water the boat is in. They typically have a polygonal bore and a round OD, so as to maintain a pretty good flow of water to lubricate the steel/rubber interface. That might be a problem for an interstage bearing, that presumably runs with a pretty decent pressure difference across it; it would certainly reduce the pump flow capacity somewhat, with all the reverse flow.

If you have radial access to the bushes, maybe you could filter the pump output and pump cleanish high pressure fluid radially into each bush, so they'd be flushed with cleanish fluid for at least part of their length. You might have to change the filter once in a while...



Mike Halloran
Pembroke Pines, FL, USA

 

[Artisi]

replica : try

http://www.exalto.co.uk

for idea's and tech. data

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.)

 

[replica]

MikeHalloran

I like the idea of allowing the particles to embed themselves a little deeper by using the softer bearing materials, at least softer than bronze. However as I mentioned earlier I could not find embedded particles such as sand or oxide debris on the surface of the bronze bush in contact with the shaft. So the assumption that I made is based on the presence of a lot of sand in the pump during maintenance and the nature of smooth and mirror finish worn surface. I will try to find the embedded particles on the other bushes which may throw some light to the issue. It is also possible that the particle has been washed away during cleaning. I will look at the wet and unwashed bush for the evidence. I am desperate to find the evidence that failure is due to the trapped hard particles. Is it possible that corrosion attack or erosion corrosion play rules? Just a wild guess.


Artisi

Thank you very much for the link, but after looking at the link and as mention by Mike Halloran I am afraid that the pump might loose some flow capacity due to reverse flow. But anyway at this stage anything worth trying.

 

[MikeHalloran]

Corrosion? Erosion? Sure, both are possible; you're pumping crude, right?


Mike Halloran
Pembroke Pines, FL, USA

 

[Artisi]

replica, the loss through the interstage's might well be insignificant, it will depend of the pump/ impeller design which at the moment to us is an unknown.

Have you contacted and discussed the problem with the pump manufacturer?

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.)

 

[replica]

MikeHalloran,

Yes we are pumping sour crude oil, so corrosion and erosion corrosion is possible. But as stated in NACE MR0715/ISO 15156, Table A3, nitronic 50/UNS S20910 used as a pump shaft is suitable for any combination of temperature, H2S partial pressure, chloride concentration and in situ pH occurring in production environment is acceptable. However as for solid sulfur there is no data available. However it is very difficult for me to find the evidence of corrosion attack on the the polished shaft surface and trace of corrosion attack if any ( unless I know the type of inhibitor as well as drag reduction chemical used, which is not in my knowledge). Should the trace be on the surface which is very difficult to analyze by EDS, in that case we must used XPS for surface analysis which is very tedious. It is also possible that the corrosion create the passive layer of sulphite which then removed by embedded particles and further corrosion will continue. Anyway this is very difficult to prove.

What we are doing now is trying to study the topography on the surface of worn shaft and bush by Scanning Electron Microscope and try to do some EDS analysis on the embedded particle if any and establish a possible cause of failure. What bother me is that some other area where the shaft worn out ( at other bowl bush ) did not show smooth and shining surface (rather roughly abraded surface) and as in the first and second stage area. Or is it also possible that some region has more embedded particle and polish more than the other. However my believe is more on particle entrapment rather than corrosion.

Artisi

No I did not contact the pump manufacturer yet because I would like to establish the cause of the failure first.

 

[Artisi]

Bear in mind, as any abrasive particles are pumped and eroding "whatever" they are also becoming less abrasive as the sharp edges are being abraded off so it is possible to see reduced wear in some areas,also any particle trapped within the "bearing" the is circulating round and round, something which doesn't usually happen with fluted cutless type bearings as they are usually passed thru'.

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.)

 

[JJPellin]

I work in an oil refinery that runs extremely sour, heavy Canadian crude. Our crude primary pumps are typically two or three stage vertical turbine pumps. I checked on the shaft and bushing materials that we use. About half of our primary pumps use Graphalloy bushings. These would be much softer than the bronze that you are using. The other half of our pumps use Bronze bushings called out as "660 BRZ" or "SAE 64 BRZ" on the bills of material. All of our pumps use 416 SS shafts. We do not use any outside flush to lubricate our bushings. We have not seen the sort of damage that you are describing. Our pumps typically run for several years between overhauls. I did not research the exact configuration of our bushings for features such as the axial groove shown in the picture. If we were to use rubber bushings, the only compound that would likely hold up in our service would be something like Kalrez.

Johnny Pellin

 

[replica]

Artisi

Thank you for the info on cutlass type bearing.

JJPellin

Thank you very much for sharing your valuable experience. It seems that I have many options for bearing materials. Considering your experience on sour crude using 416SS shaft, I may exclude the corrosion issue here since I am using Nitronic 50 (I think suppose to have better corrosion resistance under H2S environment). My main concern is the entrapped sand particles. What is your bearing clearance? Does your crude contain sand? It will be very nice to have pump that can last years between overhaul...lol

 

[JJPellin]

The pumps I looked at also have 1.5" diameter shafts. We work on these pumps so rarely, that I was unable to find the clearances for the bushings in our repair records. Our standard table of clearances for bushings in a pump of this configuration would call for 0.006" to 0.009" under a bronze bushing and 0.005" to 0.007" under a Graphalloy bushing. We probably make these bushings in our machine shop and probably cut them to these clearances. They are probably straight, smooth bore bushings without any axial grooves. Our primary pumps all run at 1800 rpm.

We process heavy, sour Canadian crude in addition to diluted tar-sands bitumen and synthetic crude cracked from tar sands bitumen. I could make the argument that our crude is among the nastiest on the planet. I have never seen it myself, but I am told that when we open up our crude tanks for maintenance, we find sand bars as much as 8 feet deep. I believe we do pass some sand through our primary crude pumps. But, I am not sure of the concentration or particle characteristics.

Johnny Pellin

 

[MikeHalloran]

Johnny brought up another good point. The axial vrooves feed fresh abrasive with every revolution. Did the bushes always have them?

Mike Halloran
Pembroke Pines, FL, USA

 

[replica]

JJPellin

Thank you very much for sharing and bringing a good point regarding bush design, for your information, our pump runs at 3650rpm. I know that Graphalloy bearing has graphite in it which is very good lubricator and graphite is always use in self lubricating bearing (without lubrication fluid) It seems that your bronze bearings do not have grooves on them . I am not sure about the benefit of having groove other than for lubrication purpose. Which is better with groove or without groove? I think it probably depends on the type and design of the pump. I have to find out weather it is permissible to have bearing without groove or even with more grooves.

MikeHalloran

All our bearings have 3 axial grooves on them. Is it possible by having more grooves will provide better flow and prevent the entrapment of solid particles? or the opposite? Any suggestions? Johnny experience had shown better performance without groove.

 

[romke]

one point of interest that struck me in this discussion was the fact that using a larger clearance resulted in longer shaft/bush life. can it be that lack of lubrication plays a part in all this, in the sense that insufficient "crude oil lubricant" enters into the bushing? "crude oil" of course being a "lubricant" with a lot of shortcomings, since it usually contains dirt, can be highly corrosive and most likely has a viscosity that will not help in getting sufficient fluid into a bush that needs to be lubricated.

 

[JJPellin]

If those bushings have three grooves, they are probably there for the purpose of shaft stability. A vertical shaft will tend to be unstable in a smooth bore cylindrical bearing. It would tend to be prone to whirl/whip. A tri-lobe design would be a classic solution to this problem. We had a set of liquid sulfur pumps that were converted from 1800 to 3600 rpm and had vibration problems. We solved these with a tri-lobe bearing (bushing) configuration. Knowing that your pumps are running at 3600 rpm, I would recommend that you be very cautious about removing those grooves.

Is it possible that the grooves are contributing to the problem by feeding a fresh stream of sand into the bushings? It is certainly possible. I would consider shortening those grooves so that they don’t break through both ends of the bushings. This would keep the stabilizing influence of the tri-lobe design, but block the entry path for more solids.


Johnny Pellin