22 replies to this topic

[sunny143]

Problem:

 

- Two Pumps are installed in field. 01 Duty and 01 Spare

- During pump change-over, cavitation problem occurs.

- It is proposed from client that install equalization line from pump casing vent and return line back to suction vessel as per sketch in blue color.

- Unfortunately in field, no casing vent connection is available.

- Pump is designed for -0.5 Barg suction pressure.

 

QUESTIONS

- How this equalization line works?

- After pump start-up, equalization line will be Closed or normally open?

- As no casing vent is available, so where this line shall be installed? i.e. on Suction side or Discharge side.

- What would be the RO sizing criteria? Flow and Pressure drop across RO.

Attached Files

•  Equalization.GIF   45.42KB   51 downloads

[fallah]

sunny,

 

Having whole system configuration to be able to answer your questions, please submit the system with two pumps along with the interconnecting piping and valves...

[Napo]

Sunny,

 

Excuse me, I have a question: your diesel is in a tank at 143ºC or 143ºF?.

 

Napo.

[sunny143]

Sunny,

 

Excuse me, I have a question: your diesel is in a tank at 143ºC or 143ºF?.

 

Napo.

 

This is a Pressure vessel not a tank. Temperature is 143ºC.

[sunny143]

Sunny,

(Quote/)

- How this equalization line works?

- After pump start-up, equalization line will be Closed or normally open?

- As no casing vent is available, so where this line shall be installed? i.e. on Suction side or Discharge side.

- What would be the RO sizing criteria? Flow and Pressure drop across RO.

(/Unquote) 

 

You have to provide each equalization line from pump casing to suction tank as suggested by the client.:

-The line can vent residual air from the casing to the tank, and makes the pump full with the liquid under vacuum condition.

-After start-up the block valve can either be closed or opened, Even if the valve is fully open, the restriction orifice will keep itself under limited flow.  If the pump is exposed under frequent start-stop situation, then let the valve be opened normally.  

-On the highest point of the casing or discharge line shall the line be connected, but on the upstream of the discharge check valve.

-Orifice size is not critical. just only it has to equalize the vacuum before the pump is started. Larger orifices do nothing but drop efficiency of the pump.

 

~Stefano

 

Thanks for your comments. Nice suggestions.

[sgkim]

Sorry, Sunny,  I myself deleted my previous posting by mistake, so I have to refer to your quote...  

 

Stefano

[sunny143]

 

Stefano,

 

Yesterday, i did some initial design for the equalization line requirement and sent to client for review.

 

I adopted below design philosophy;

 

1.       As there is no casing vent available on pump (041-G-0012 A/B). So, to ensure that pump casing is free from vapors equalization line is taken from pump discharge side.

2.       Equalization line will be utilized for pump startup only and all valves on the line will be normally closed.

3.       Restriction orifice (RO) will be sized to avoid the full vacuum in Pump (041-G-0012 A/B).

[sgkim]

Sunny, 

Item 1 O.K.

Item 2 Conditionally O.K. if it is not so frequent to start and stop the pump.

Item 3 The objective of RO is "not to avoid full vacuum" but to "equalize the pressure of the pump to the suction tank" regardless of the vacuum degree.  Then the size of the orifice can be so determined as to be quickly equalized in pressure during the operator sets up the start-up condition.  Too small orifice may be clogged easily then "1/4 inch diameter", say, would be O.K.    

 

~Stefano (sgkim)

[sunny143]

sgkim:

 

 Item 2: Yes, its not done on frequently basis. Pump changeover is done every 6 months. So, normally closed line is Okay.

 

 

Item 3: Keeping in mind, the line is NC and it will be operational during pump changeover. Pump suction vessel is designed at F.V condition but pump is not designed for F.V. Say, when equalization line valves are opened during pump changeover and if there is any condition occur which may cause vacuum to appear in pump casing due to equalizing line.

So, to avoid the pump damage from vacuum, RO will be sized accordingly.   

[sgkim]

Posted Today, 02:58 AM

sunny143, on 15 Nov 2016 - 6:41 PM, said:

 

Item 3: Keeping in mind, the line is NC and it will be operational during pump changeover. Pump suction vessel is designed at F.V condition but pump is not designed for F.V. Say, when equalization line valves are opened during pump changeover and if there is any condition occur which may cause vacuum to appear in pump casing due to equalizing line.

So, to avoid the pump damage from vacuum, RO will be sized accordingly.   

 

Sunny,  

I can hardly agree with the idea that the pump casing is not  designed for (either partial or full) vacuum services, and that the RO can help to avoid exposing to vacuum condition.   If an operator opens the block valve of the equalization line during pump changeover, the pump shall soon be reached ('equalized') to the same vacuum condition as the suction tank.  Neither block valve nor RO can prohibit the dynamic situation.  

 

The pump, presently, is used as a sort of 'vacuum discharge pump", and the pump has surely been selected from those operable under partial or full vacuum.  Consult the pump vendor as to what should be done on the matter if you encounter pumping problem even with the equalization line. 

 

~Stefano

[sunny143]

 

Posted Today, 02:58 AM

sunny143, on 15 Nov 2016 - 6:41 PM, said:

 

Item 3: Keeping in mind, the line is NC and it will be operational during pump changeover. Pump suction vessel is designed at F.V condition but pump is not designed for F.V. Say, when equalization line valves are opened during pump changeover and if there is any condition occur which may cause vacuum to appear in pump casing due to equalizing line.

So, to avoid the pump damage from vacuum, RO will be sized accordingly.   

 

Sunny,  

I can hardly agree with the idea that the pump casing is not  designed for (either partial or full) vacuum services, and that the RO can help to avoid exposing to vacuum condition.   If an operator opens the block valve of the equalization line during pump changeover, the pump shall soon be reached ('equalized') to the same vacuum condition as the suction tank.  Neither block valve nor RO can prohibit the dynamic situation.  

 

The pump, presently, is used as a sort of 'vacuum discharge pump", and the pump has surely been selected from those operable under partial or full vacuum.  Consult the pump vendor as to what should be done on the matter if you encounter pumping problem even with the equalization line. 

 

~Stefano

 

 

Yes, you are right. Pump is designed for Vacuum. I checked with the DS.

[fallah]

Problem:

 

- Two Pumps are installed in field. 01 Duty and 01 Spare

- During pump change-over, cavitation problem occurs.

- It is proposed from client that install equalization line from pump casing vent and return line back to suction vessel as per sketch in blue color.

- Unfortunately in field, no casing vent connection is available.

- Pump is designed for -0.5 Barg suction pressure.

 

QUESTIONS

- How this equalization line works?

- After pump start-up, equalization line will be Closed or normally open?

- As no casing vent is available, so where this line shall be installed? i.e. on Suction side or Discharge side.

- What would be the RO sizing criteria? Flow and Pressure drop across RO.

 

Sunny,

 

The situation you described for pump cavitation, at start up condition i.e. operating too far right of the pump curve along with a low suction pressure i.e. low NPSHA, indicates you probably are faced to suction cavitation rather than discharge cavitation. Hence if a balance line can be considered as a solution and there is no connection nozzle on the pump casing, it might be better to take the equalization line from suction side rather than discharge side...
 

[sunny143]

 

Problem:

 

- Two Pumps are installed in field. 01 Duty and 01 Spare

- During pump change-over, cavitation problem occurs.

- It is proposed from client that install equalization line from pump casing vent and return line back to suction vessel as per sketch in blue color.

- Unfortunately in field, no casing vent connection is available.

- Pump is designed for -0.5 Barg suction pressure.

 

QUESTIONS

- How this equalization line works?

- After pump start-up, equalization line will be Closed or normally open?

- As no casing vent is available, so where this line shall be installed? i.e. on Suction side or Discharge side.

- What would be the RO sizing criteria? Flow and Pressure drop across RO.

 

Sunny,

 

The situation you described for pump cavitation, at start up condition i.e. operating too far right of the pump curve along with a low suction pressure i.e. low NPSHA, indicates you probably are faced to suction cavitation rather than discharge cavitation. Hence if a balance line can be considered as a solution and there is no connection nozzle on the pump casing, it might be better to take the equalization line from suction side rather than discharge side...
 

 

 What if vapors are trapped in pump casing?

[fallah]

 

 What if vapors are trapped in pump casing?

 

 

Sunny,

 

Please note that the main function of the equalization line is equalizing the pressure between the pump and the suction vessel not to be a venting line...nevertheless it can be supposed with considering such equalizing line in suction side the vapor being generated (and trapped in pump casing) due to equalized suction pressure at the moment of the pump start up not to be so much such that initiates a cavitation event...

 

Indeed, considering there is no vent connection on the pump casing it can be supposed the pump casing isn't a place for vapor trap in relevant vendor standpoint and any vapor can be released through the discharge line in which there would be no vapor pocket till the next equipment...  
 

[saeed.s]

Thanks falalh, napo & sgkim for their valuable comment.

Dear sunny, I’d like to review your question again and please specify which one is your case.

 

During pump change-over, cavitation problem occurs:  I assume cavitation occur for both pumps when both pumps are working , then as fallah said its suction cavitation.

I saw such problem before and it occurred because of wrong design. The height of tower wasn’t enough and because of deep vacuum, there wasn’t enough flow to pump suction. With installing an equalization line on pump suction it was solved.

Then when two pumps are in service the suction flow is not enough for both pumps and you hear the cavitation noise (but in fact, is suction loss noise).

 

During pump change-over, cavitation problem occurs:  Cavitation occurs in spare pump only.

1-after 30 sec or 1 minute, problems disappear: it’s because air leakage in start-up and after pump working, it will be disappear.

2- Problem is continuing with pump working: I think, it’s because of seal damage and continuous air leakage to pump.

 

What if vapour is trapped in pump casing?

For release vapour from pump casing, it’s not necessary to connect it to vessel. You can release vapour to atmosphere because there is enough pressure in pump discharge.

[fallah]

behnam,

Please note that for many fluids relieving the trapped vapor in pump casing to atmosphere is not possible and should be released to safe location,flare or suction vessel...

[Saml]

Perhaps you have already done it, but if you have a chance, talk to the operators that actually do the changeover in person. If you can, avoid being with the contracting engineer. Go for a permit for "surveying" and talk to them while in the field. Ask if this is a problem that has always been this way, if it started after a modification, if they find fouling in the lines and pump when they open, etc, etc.

 

There may be causes you are not seeing right now. A seal type replacement, or it appeared after they increased the plant throughput or after a change in  the product quality being pumped, or after an amergency shutdown (something got loose inside the vessel and is stuck in the vortex breaker), fouling in the suction line,  or even a change in procedures. 

 

Also, It won't be the first time that someone ask (and pays) for a "solution" to cover a previous mistake.  As an example,say, that maintenance installed a seal that was not correctly repaired, they found that but said nothing.  So when you take the pump for making your connection they will schedule a "preventive maintenance". After that everything will be solved. 

 

May be for you is better to play along even if you find something like that (that is your call). But what cannot happen to you, is that you do what you are asked for (build the compensation line) and after that the problem keeps happening.  Mainly because it seems that your client is asking for a solution and only "suggesting" a line. So if it does not work, it's yours.

[S.AHMAD]

Hu Sunny.1. My view is that equalization line will not be able to solve your problem. reconsider better solution.

2. From the drawing provided there is no flow control valve at the pump discharge. If so then consider installing the valve. This installation can avoid pump cavitation. The centrifugal pump cavitation as we all know that due to insufficient pressure at the suction. When the spare pump started, the flow subsequently increased thus the suction line pressure drop to the extent that the suction pressure is bellow the vapor pressure of the fluid at the operating temperature. Another possible solution is to increase the liquid level during changing to spare pump. However you need to do energy balance calculation to determine the correct tank liquid level. Of course you can solve the problem also by cooling the fluid in the tank.. The current operating temperature is too much higher than the flash point of diesel normally at aboiut 60 C. At flash point, some vapor is being formed.

 

My strong view is that the proposed equalization will not solve your cavitation problem.

 

s.ahmad

Edited by S.AHMAD, 16 November 2016 - 05:39 PM.

[Napo]

Sunny143,

Have you checked the suction filter?, maybe the wire mesh is too little (maybe the start up mesh is in the filter).

In Petroleum Refinery (Topping Planta) we work with high temperatures and low suction pressures and normally we don't have problems in the pumps changes.

Under what standard were design your system pumps?

Napo.

[sgkim]

Fallah, Behnam, Sami and S.Ahmad,

 

Lots of valuable comments on this thread by all of you would be helpful to solve the Sunny's problem.

 

But the real problem and its solution is not so complicated - just and simply connect the "equalization line" from the highest point of the discharge piping to the suction tank.  The line with a block valve and an orifice are already suggested by a Sunny's client. No vent nozzle was found on the casing, so a new connection shall be provided on the upstream of the discharge check valve. 

 

The function of the "equalization line" is to vent the (vacuum-leaked) air from the pump casing to the suction tank which is maintained at sub-atmospheric pressure during running and/or before start-up of the pump.   None of pump-casing vent ("priming") 'to atmosphere', 'to flare system' or 'by the liquid from the other running pump' would be possible if the suction tank and pump is under vacuum state.   

 

The term 'cavitation problem during change-over" initially used by Sunny does not seem to be suitable to describe the practical situation; 'air-lock under vacuum', instead, was exactly the correct one. (Sorry, I should have explained this in my previous replies.).   Presumably, the cavitation phenomenon was not noticed after start-up or during normal operation of the pump connected to the 'hot liquid' vacuum tank. So it surely does not seem to be related to NPSH  problem either.      

        

A client's operator, together with Sunny, must have checked all of the system including the mechanical seals of the pump, and he would have surely found nothing special compared with the other running pump at that time.  Nevertheless, vacuum-leak through the various parts of the pump accumulates inside of the casing during stand-by period of time.  The sub-atmospheric air-lock  in the pump casing cannot be vented without the equalization line.   Only after the vacuum is equalized by opening the block valve, the liquid can flow into the pump suction and fill the pump impeller and casing completely by the static head difference.  Then the pump is ready to start for changed-over.    For frequent or for automatic start-up of stand-by pump, the orifice is recommended. 

 

~Stefano

[S.AHMAD]

I forgot to ask.... both pumps cavitate or only the spare pump? If both pump cavitate the root cause is different from the case of only the spare pump cavitate.

in order to solve the problem we need to identify the root cause. Please tell us  the root cause?

Edited by S.AHMAD, 17 November 2016 - 11:07 PM.

[sgkim]

S.AHMAD,

 

I see the abnormality occurs only before (start-up of stand-by pump for) changing one pump over the other, though.

Without each equalization line, therefore, each pump will equally suffer exactly the same trouble -- cavitation (Sunny referred so.), air-lock under vacuum or something of that fashion.

 

~Stefano

[sunny143]

Fallah, Behnam, Sami and S.Ahmad,

 

Lots of valuable comments on this thread by all of you would be helpful to solve the Sunny's problem.

 

But the real problem and its solution is not so complicated - just and simply connect the "equalization line" from the highest point of the discharge piping to the suction tank.  The line with a block valve and an orifice are already suggested by a Sunny's client. No vent nozzle was found on the casing, so a new connection shall be provided on the upstream of the discharge check valve. 

 

The function of the "equalization line" is to vent the (vacuum-leaked) air from the pump casing to the suction tank which is maintained at sub-atmospheric pressure during running and/or before start-up of the pump.   None of pump-casing vent ("priming") 'to atmosphere', 'to flare system' or 'by the liquid from the other running pump' would be possible if the suction tank and pump is under vacuum state.   

 

The term 'cavitation problem during change-over" initially used by Sunny does not seem to be suitable to describe the practical situation; 'air-lock under vacuum', instead, was exactly the correct one. (Sorry, I should have explained this in my previous replies.).   Presumably, the cavitation phenomenon was not noticed after start-up or during normal operation of the pump connected to the 'hot liquid' vacuum tank. So it surely does not seem to be related to NPSH  problem either.      

        

A client's operator, together with Sunny, must have checked all of the system including the mechanical seals of the pump, and he would have surely found nothing special compared with the other running pump at that time.  Nevertheless, vacuum-leak through the various parts of the pump accumulates inside of the casing during stand-by period of time.  The sub-atmospheric air-lock  in the pump casing cannot be vented without the equalization line.   Only after the vacuum is equalized by opening the block valve, the liquid can flow into the pump suction and fill the pump impeller and casing completely by the static head difference.  Then the pump is ready to start for changed-over.    For frequent or for automatic start-up of stand-by pump, the orifice is recommended. 

 

~Stefano

 

 

Dear Sgkim,

 

The objective of equalization line is to prime the pump during startup. As the system is under vacuum so it can not be vented out to the ATM or Flare system. Upon my little research about the topic, this equalization line requirement is also endorsed by the API 686 which states that;

"For pumps taking suction from vacuum towers or columns, an equalizing line from the pump back to the vapor space in the tower or column shall be provided to vent the pump at startup."

 

I hope the API 686 statement and very detailed explanation from Mr. Sgkim settled the dust.

 

Thanks everyone for their valuable feedback. 

 

Regrds,