centrifugal pump

Reduce the external diameter of impller by 2mm to 3mm on lathe machine and Check the trueness of shaft also and then refit with tight fittting of Key & keyway and assemble the pump with new bearing and Mechanical seal.It will solve your problem.

you may have cavitation issues, this will eat up the impellers

Need some clarification. Is the bearing failing? When the pump binds up to the case is the bind on the OD of the impeller or on the sides? Whats the clearance between the OD of the impeller and the case? What kind of axle play is there? The impeller condition badly worn fins or broken off? Bearing housing is the bore worn out?

i m handling two types of centrifugal pump in which one type is getting damage many times and which is the heart of our plant.

Dear Lokhande

1-First of all you must have a history record for your pumps.

2-the causes of the hole of the left impeller may be one of these reasons:

a-the alignment between the pump and the driver is not correct.(check)

b-the casting of the impeller is not an investment one.( the point is defective)

c-Did you change from packing to mechanical seal? (that causes shaft deflection)

d-is the fluid that you pump has high temp (that causes thermal expansion of the shaft,so the impeller could be stuck to the casing)

e- Has your fluid solid particles?

f- close this hole by welding and notice the place of the new hole.

g-I noticed that the left hand side (of the left impeller)looks brighter than the right side one,that means the friction on this side is greater than the other because of:

the bearing may be defect or the shaft deflection.

Notice please: the left impeller is closed one ,this the is most common deigens with ISO standard pump,while the right impeller is semi-open one this is common with ANSI

standards.is that your case?

Hoping that helps you to solve this problem.

Aly

dear aly

you are really great i can not expect such a perfect answers but let me clarify you for some points so that i will get more clear information from you

1) bearing casing worn out several time but i was not there how they machined it is it through with each other i don't know

2)those three holes are symmetrical so that i think its the part of design of the impeller

3)right side impeller if u observe it has one bright curve on left side which shows that it balanced but left side one don't have anything like this to balance may be it balanced or may be not

4)even right side impeller has engraved "fail" what is that means? note it - right side not left side impeller

5)how can i understand that impeller is investment casting? please let me know how can i check?

6)from beginning it has mechanical seal we not changed for gland packing

7)yes fluid has high temperature and pressure which is causing material (soap) leak from mechanical seal and even it enters in to the oil reservoir of the pump. please suggest me right type of oil seal so that i can come over on this problem

8) both are open type impeller only the design is different back side of the impellers i shown u in first picture let me show u the front side please let me know more about the standards as u mention ISO and ANSI. pump manufacturer is the same pampetravini why they r using different standards.

These two impellers are are of different designs.

One on the right is comparatively for low pressure service. The vane like ribs are called as repletion/expulsion vanes. These ribs/vanes push the liquid that try to gain access to stuffing box past back side of the impeller. There by keeps the stuffing box area at low pressure, which means lesser burden for stuffed glad or mechanical seal. But in these types of impellers the the pressure exerted on either sides if the impeller shrouds may not be the same. So the impeller would be subjected to differential pressure and undue thrust would be acted up on the impeller, intern on the shaft and thrust bearing. This won't cause much of a problem when the application is for low pressure service (Discharge pressure - suction pressure).

The on on the left is typically for medium pressure service. The circular ring is called 'impeller wear ring'. A mating ring called 'casing wear ring' (in most of the designs) would be fixed on the casing. The proper clearance is normally advised by the manufacture. Or can be refereed from API 610 or ASME. There are 'thump-rules' as well.

The purpose of these wear rings are to reduce/minimize the amount of discharge fluid getting in to stuffing box. However perfect the clearance between these rings be, the liquid slowly, eventually rise the stuffing box pressure near to discharge pressure. To take care this pressure raising, three (or more) holes are availed as to equalise this pressure to suction pressure. With this arrangement the stuffing box is ensured of suction pressure (not discharge pressure), which means lesser load for sealing arrangement.

In this case the evidences for the failure are inadequate.

What is the nature of failure:

- Bearings? Which bearing 'thrust or radial'?

- Wear rings? Uniform enlargement of clearance or a eccentric wear?

- Is the casing: bottom supported or center line mounted?

- How long the pump runs after each overhaul? What are the symptoms of failure: Vibration, temperature, cavitation? and where: in bearing bracket, inboard or out board or casing, around stuffing box area or suction area?

- It is told that the application is 'high temperature service', what is the pumpage temperature? What is the temperature difference between discharge vs suction?

At least these data are required for troubleshooting.

Looking at the second set of pictures, the impeller at the left (the subject impeller), at the tip of the vanes found to be heavy rubbed off. This mechanical contact suggest following questions:

- The corresponding wear at case: Is it uniform or concentrated at one area?

- If uniform: Wrong assembly or undue/un-accommodated thermal expansion. Many pump technicians do not know which is thrust bearing and which is radial bearing in an 'Over-Hung' pump. Improper/inadequate locking of thrust bearing from axial play would lead to internal rubbing. Not allowing radial bearing axially to float (arresting it) would bend the shaft. This again would lead to internal rubbing.

- If un-uniform: Uneven/non concentric tightening of casing and bracket. Many occasions, in the field, the pipe line load or thermal expansion, distorts the joint between the casing and bracket. This leads to leakage at this joint even when the pump is wormed up. The uneven tightening in panic leads to warped case vs bracket joint which intern lead to internal rubbing.

Why mechanical seal leaks? Soap is a sober and lubricative liquid. These are good options in selecting an appropriate seal for every application. However good a seal is, it could be (first) victim of vibration due to failure of other components or even vibration due to misalignment.

Hope more feedback could bring out right solution to these undue failures.

thanks for your advices. please check some more details for clarification.

COMPARISION BETWEEN TWO PUMPS SPECIFICATIONS

please look for suction pressure of left impeller is 4 bar and another one is negligible.bearings we are using in both pump is 6309 and 5309 or 3309.

please suggest me which bearing will be suitable 5309 0r 3309. please see the picture of my left impeller pump which is heart of my plant.

after overhaul it work fine for 20-25 days. another observation i realise that when we stop the plant for any other reason, this pump breakdown.

You said, "this pump breakdown when the plant is stopped for any reason".

Please elaborate this statement. Are the breakdowns happen while stopping the plant or restarting the plant?

Most likely the former could be the case.

Looking at the picture again:

Is the product congealing/solidifying in nature on cooling down?

Does the pipelines have 'heat (steam or electrical) tracing lines' to prevent this solidification?

If your answer is yes for these questions, is there any heating arrangement for the pump case (jacket or heat-tracing)? I could not see any elaborate arrangement for warming/heating the case or the stuffing box.

When pump is not running (during the plant is down), the product gets solidified. On your restart, the impeller experiences a huge breaking effect might lead to breakdown.

The worst failure could be: shaft sheers at the weakest section. If not this failure, this situation could lead to accelerated wear and tear on seal and bearing due to high vibration and heat (due to viscous-shear).

Is this scenario sounds possible? The man at the spot is the best judge. If you feel the same way, we will discuss of the possible interim and remedial actions.

yes breakdowns happen while stopping the plant and restarting the plant.

yes that one is heat steam tracing line.

and product is becoming solidified at room temperature.may be after stoping the plant soap which enter in stuffing box get solidified and then it causes the viscous-shear.

because of that only i ask for good type of oil seal which can help us to overcome on this problem. please let me know for the bearing either 5309 or 3309. which one will be better. please advice me.

Both 5309 and 3309 are 'double row angular contact ball bearings' of same overall dimensions. Except marginal load bearing capacity there is no big difference between them. Since the speed is only 1500 rpm and the axial/radial loads are minimal both are suitable for this application.

Importantly, ensure:

- This 'thrust bearing' is properly locked from axial float. Take care not over clamping the outer race while such locking.

- Ensure the radial bearing is allowed to float axially on thermal expansion while running and contraction while at standby.

- Ensure the proper running clearance at 'wearing rings'.

- Ensure proper axial clearance between the impeller vanes and suction case. Too close will lead to seizure when heated up on running, more clearance will lead to loss of through-put due to internal circulation . This could be achieved by adjusting the relative position of the thrust bearing with respect to bearing bracket or by adjusting the gasket thichness between bracket and casing. Note: This activity is required since heavy scoring is noticed in the picture.

You have mentioned:

- Bearing bore was rebuilt, why? Normally this will not be called for, even when bearing failure very badly. This will happen if only the bearing rotates as one unit, along with outer race! Quite possible in your case, since product (solidifying in nature) gets in to bearing bracket, contaminates the lubricant and makes the bearing as an en-mass.

The possible failure continuum in this case could be:

Product solidifies, leads to > Mech-seal leaks > Contaminates the lubricant > Bearing failure…

Avoid solidification: by establishing heating/warming arrangement around the pump casing and stuffing box. This should be in action even when (more so when) the pump is not running.

Review the design of Mechanical Seal: Mech seals work satisfactorily even for hash, corrosive, congealing liquids like: asphalt, molten Sulphur etc. But for the high temperature, the soap is a docile, non-corrosive (even slight alkaline should be alright), lubricate liquid. As long as the entrapped product in stuffing box remains in thin liquid form, there is no dander, seal will be active. So heating/warming around seal chamber is very essential.

- Hope the seal is a 'single coil design'. If not consider using this type, since 'single coil' is more suitable for congealing liquids.

- Consider 'external flushing' with a compactable, preferably a non-solidifying liquid. May be hot water/condensate or like. Note: This flushing would contaminate product.

- Consider quenching with 'steam', to prevent solidifying at gland area.

- Double seal arrangement is costlier but more ensuring the best result. Need more studies to justify this option.

Consider 'Bearing Isolator' for inboard bearing cover. This prevents ingress reaching and contaminating the lubricant.

Hope these help...

hi yesyen

you are realy the king of pump. i want to make u my guru for the pump. please help me to get more knowhow of the pump.

please explain me some of the terms like

heavy scoring what is that means

makes thebearing as an en-mass

what is the bearing isolator

please explain me so that its easy for me to explain my director

please tell me how can i select one water transfer pump just give me an example

how can i select pump for sodium silicate it is very viscous.

Scoring at the edge of vanes:

These marks indicate the mechanical contact of impeller with suction casing.

Bearing getting galled/seized in such a way it become a single block. Inner race, balls and outer race all rotates together as a single block inside the bearing housing. The continuous rotation of outer race will eat away the bearing housing surface. This only would eventually end up and call for 'rebuilding' the bearing bracket at the bearing bore.

Bearing Isolator is an arrangement like mechanical seal but much simple in design. These are for superior to ordinary 'oil seals' made of rubber. Find more info on the following sites/links:

www.inpro-seal.com/
www.daemar.com/dmr_bearing-isolators_39.htm
www.skf.com/portal/skf_us/home/products?contentId=344216&lang=en
www.johncrane.com/iPortal/upload/banner/S_protech.pdf

dear sir

as i get lot of knowledge from u please clear my this doubt.please see the picture.

this is the line diagram. in which u can see green centrifugal pump. rector column and pressure valve. suction side of pump all the raw material is entering.reactor temperature is 125 degree cel. pressure valve is getting operated above 2.5 bar.

Suction nozzle of the pump is connected to the header of 7 raw materials which are just close to suction as per the picture. This raw materials are get pumped by different multistage centrifugal pumps. Details of discharge pressure are as bellows:

stearine -8 bar, fatty acid- 8bar, coconut fat-8 bar, brine- 7.3bar, process water-7.3 bar and half spent lye -9.2bar , caustic -6bar

If we calculate all this discharge press it will be 53.8bar. Even if we consider half of friction loss it will be 26.9bar. And in pump specification suction pressure is 4bar.

will it affect?

if axial and radial load are minimal for 5309 and 3309 bearing can i introduce another type of bearing if yes which one will be suitable

Hello Sandeep

Looks as if your plant suffers from serious heart problem. (You said this pump is the heart of your plant) So why not try a transplant. This time use a pump that is suitable for the purpose. It looks as if your pump is working in a very highly corrosive environment, and maybe all your troubles are a result of this. Try use SS pumps and talk to a seriuos pump manufacturer/supplier to get a serious advise. There are several things you didn't mention. Based on what you say, one pump has a positive displacement, one has an open impeller, and the other is closed why?. I must come to the conclusion that you are using the wrong pump for the wrong application...

Hope it helps.

Wangito.

lokhande: (Use Copy & Paste or drag text to quote the original text.)

dear aly

you are really great i can not expect such a perfect answers but let me clarify you for some points so that i will get more clear information from you

Dear Sandeep

I really waited to have a concentrated answer from you about THE HISTORY of your pump ( or pumps )this will help us to have the min. proposals and get the max. gain to solve that problem . After this long introduction let us begin:

1) - I see your pump is very sad and is weeping with hot tears, Do you agree with me? I hope we can see another picture with new look for the pump smiling and dancing .

Your problem is concentrated at one of (or altogether ) the following reasons:

a-Excess heat (I see from the picture that there is a lot of dry product covering the pump from outside , also covering the motor where you can not see it's fins , these fins are designed to dissipate heat, so this heat transmits to the pump through the shaft, also the ambient temp ) it is known that increasing the temp with 10 degree C means increasing the rate of corrosion with 50% .

b- vibration (misalignment – harmonic vibration from the surrounding machines –working out of the BEP )

c- Wrong Choice (oversized – unsuitable elastomeers – standard pump, while you need Heavy duty pump- may you need positive displacement pump instead of centrifugal pumps this is due to the viscosity)

d-Aging( if you gave me the correct history of your pumps ,may be you need to buy a new one ,so you do not have to have these long discussions) you mentioned that the problem is repeating after every overhaul for the pump about 20-25 days this means there is something up normal my dear. The normal one is after 20 months.

2- for the bearing you have to follow the instructions of the manufacturing company( please give me the complete information about this company).here I say may be your problem is concentrated here ,because repeating assembling and disassembling the bearings ,sure ,the bearing seat is fully damaged so there are no correct tolerance more(between the outer casing of the bearing and the inner casing of the seat) so you lose the alignment of the shaft and this very difficult to be repaired ,that mean there is no way ,you have to buy a new one.

3-Investment casting ( using modern machines) means more accuracy and possibility to do any difficult curves in your casting body also it means smooth surface comparing with the sand casting.

4-From inspecting the left impeller I see severe notches , may be you have metallic particles coming through the pump (here you need strainer) or some one made that using portable grinding machine.?

5- Is the right impeller locally made? If yes , that should not be done,because this needs an accurate way of balancing.This make very bad vibrations.

6- I spoke about the peripheral hole only ( check my last answer I said hole not holes) the other three are made for balancing pressure with that behind the impeller.

7 - for correct choosing the elastomeers you have to ask the manufacturing company according to your data (temp .press….etc.) .

Hoping that , we come nearer and nearer for the correct solution.

Regards

Aly

dear aly

i dont have history of that pump because i m just newly appointed engg there

here its kenya no maintenance record no system with this i have to impliment some good systems which is very difficult.

temp of pump itself is very high because we are handling soap at 120 degree c.

there is no vibrating machine around

we can not put positive displacement pump because this pump is mixing the rew material

pump suction nozzle is connected to the header where 5 different types of raw material is connected.

please let me know if any information u want regarding this pump.

thank you

Sandeep,

You offer a curious and interesting bit of information, i.e., that your "heart of the plant" pump is actually mixing the five component materials of the soap you manufacture.

First question asks if that is standard practice in soap factories. Are the product pumps also the instrument for mixing, for producing the product?

Second question asks is not one of the five components you are mixing with that pump the highly corrosive alkali potassium hydroxide (lye)?

Without my knowing the answer to the first question, but nonetheless, would it not be better to premix the product in some sort of rotating drum, as for instance in cement manufacture, so that without disrupting the continuous production process, the pump is not subjected to the high temperatures produced in the soap manufacturing process itself due to the lye reactions?

As I remember it, but far from being an expert on the process, is there not also a phase where excess potassium hydroxide is washed out of the product or otherwise neutralized?

I would guess that the economics of setting up that plant were and are a critical consideration and that using the product pump itself as the actual process device, before packaging and/or shipping product, is a pretty clever and asset conserving idea.

Nonetheless, your plant assets are getting eaten alive by the continued and often failures of that critical pump.

Just some thoughts that might be useful in a situation where constraint of capital might have been a factor on process.

j.

<...dont have history of that pump...>

The plant desperately needs maintenance records going forwards, then.

It would help if all plant items were tagged, then the history of the equipment installed there would be easily related to the tag number.

hi

during this discussion i gain lot of knowledge and incouragement. let me add some more information. plesae see the picture again for suction of the pump.

suction nozzle of the pump is connected to the header of 6 raw materials which are just close to suction as per the picture. this raw materials are get pumped by different multistage centrifugal pump.details of discharge pressure are as bellows:

stearine -8 bar, fatty acid- 8bar, coconut fat-8 bar, brine- 7.3bar, process water-7.3 bar and half spent lye -9.2bar(lye is very corrosive).

if we calculate all this discharge press it will be 47.8bar. even if we consider half of friction loss it will be 23.9bar. and in pump specification suction pressure is 4bar.

will it affect?

Sandeep,

Another issue.

Brine, sodium chloride or salt in solution. It too is corrosive although not as greatly as the lye.

When you shut that thing down do the materials left in the pump dry out sufficiently for the brine/salt to return to its crystalline form and thus provide an abrasive to scour the pump on start-up as appears in the impeller pictures?

Instead of allowing the material to solidify in the pump on shut down would the pump not be better served if you added to the header a water line, preferably hot, and a diversion out-take, to flush that pump before shut down?

As a matter of curiosity, is the proper proportioning of those materials in the final product, soap, achieved via the varying input pressures to that pumps header? Otherwise, how do you meter the inputs for proper mix?

Another question. Is the final product a hard soap, like in soap bars, or a liquid soap?

j.

"stearine -8 bar, fatty acid- 8bar, coconut fat-8 bar, brine- 7.3bar, process water-7.3 bar and half spent lye -9.2bar(lye is very corrosive).

if we calculate all this discharge press it will be 47.8bar. even if we consider half of friction loss it will be 23.9bar. and in pump specification suction pressure is 4bar."

Imagine you have a 6" line with 4 bar pressure and a 1/2" line with 20 bar is joined to this. After the juncture the pressure is not going to be 4 + 20 = 24 bar, right?

The best way to understand the pressure at the suction is to install a temporary pressure gauge on the suction line, very near to pump.

On subjecting the pump to a higher suction pressure than specification: Pump case would be subjected to this high pressure, will it stand? Then the seal is subjected to a higher pressure (through balancing 3 holes). If the seal itself is not of 'balanced type' the seal faces will wear off fast.

The impact on the dynamic side: If the pump is capable of raising 'X' head, if the suction pressure is 'Y', the discharge pressure is going to be 'X + Y'.

Hi Sandip!

The ( right hand side impeller ) so called blades on the impeller are called expeller. You will always find them on back side of the impeller. Function of them is just opposite that of impeller. i.e. They reduce the pressure developed. They are provided for reducing load on the gland / gland box.

For right side impeller please check compatibility of the MOC of the impeller with fluid handled.

Best Luck!

Anil

pardon me sir, but the round section you are pointing to is the wearing ring which in additional is a big help in maintaining the concentricity of its operation. The onther one (so called blade) is added to the impeller for the purpose of preventing a build up of solidifying particle at the back of the impeller. Those two parts are very important to the units operation.

If your unit is experiencing a stuck up during the operation, its probably the wearing ring i pressume, then your problem is not on the impeller type but on the medium's availability. When the impeller stuck up as you say, it only shows that part is poor in lubrication or its medium. You're probably experiencing a limited suction flow due to other reasons like the SUCTION LIFT having a very low level (check the pump curve). Another reason could be is your suction line, it probably has too much restriction like pipe ID build up due to corrossion and others.

Bottom line is, your unit is experiencing a deprived medium.

In your case of bearing worn out, you should check your unit for vibration due to installation, allignment, cavitacion and poor bearing lubrication.

you can reach me at technical@qrphil.com.ph or mater_tuh@yahoo.com

What interesting reading, and what a splendid example of the international co-operation that the CR4 forum can generate. Look what a comprehensive discussion has ensued as a result of a clear statement of the problem, and photographs of it.

One question - what is the fluid, please?

From the pictures, it looks like there is a containment problem.

Sandeep,

Some of us have raised questions about the obvious scoring, I would use a word stronger than scoring, that show in the pictures of the impeller.

I specifically raised issues, when you told us just what you were feeding that pump as to the corrosive nature of the sodium hydroxide you were feeding it and as well the brine.

Both of those materials, when you shut the pump down, will produce crystals as things dry out. Those crystals could be the source of the scoring that we see on the vanes when you restart the pump.

Likewise, damage to the bearing and bearing casing, might be the result of the battering that we see being done to the vanes.

Now you have sent us a diagram of the process. Unfortunately, outside of the fact that we now see there is a reactor vessel also involved I certainly cant make out what any of the markings on the lines feeding the suction side of the pump say. Part of the reason is the diagram is cut off on the left and the right. One line seems to say something about soda water.

Your process seems to involve inputting raw materials to the pump which it would appear mixes them and then feeds them to the reactor vessel which seems, from the diagram to feed them back to the pump.

Is that line going right from the reactor vessel the product out take?

I would suggest that the first thing we focus on is the mechanical damage to that pump.

I made a number of suggestions. One is to premix the materials before they enter the pump. I suggested a way to do that.

Another suggestion was that you install water lines so that just prior to shut down you flush the pump rather than just let material set up in the pump.

You have not responded. I would suggest that the best way to collaborate with invited advisors is to respond to questions especially when they have to do with the obvious battering that pump is taking.

Your problem is interesting and I know we all would like to help, but, speaking for myself it is difficult to help if you focus on what you think may be a quick fix (Or that at least is what it looks like), and not respond along other lines.

I say this not to offend you, but because I can't see how to help you as we are going.

Both I and others have raised questions about the mechanical condition of that pump so obvious in the pictures. I certainly don't think you will find a fix without directly addressing that issue.

Again. Not trying to offend. Just trying to help.

j.

THANK YOU TO ALL OF YOU.

ULTIMATLY I SUCCEDED.

There was a problem in pump bearing housing which i replaced from another pump which is not that much important and having different casing and impeller design.

lets have a look of my pump beauty.

Dear Sandeep

Finally the new look of your pump reflects your actual success.

Can you try to have a training course (with agroup of your country) and visit some factories in Egypt,that is through your governmental or privte sector associations?

Please, contact me ,may be we can have success in this direction.

baly2000@rocketmail.com

Best regards to you and all of our friends who shared with us in this amusing fourm

Aly

For those who like myself, outside of knowing lye and fats were involved, were ignorant of the industrial production of soap, this url.

http://www.nzic.org.nz/ChemProcesses/detergents/11A.pdf

Sandeep, how much of the process spoken of there do you folks in Kenya have, if that is not too nosy a question.

I for one did not realize how complex soap production was. They never covered it in Industrial Processes at Brooklyn Tech.

But, wow! With CR4 as a goad and web access I am getting one hell of an education.

j.

from the way i look at it, your pump is used to transfer and mix at the same time, and you medium is undergoing some extreme changes in the process. . . .

may suggest a study in replacing your centrifugal pump (radial flow) into a pump designed in an axial flow, in that manner you will minimize or hopefully eliminate an abrasive (solidifying particles) attack.