Switch from API 53B to API 53C 1

[Meshry]

In oil and gas facility; 4x crude oil shipping pumps are working in series. number of pumps in the series combination will vary according to the required daily production.
each pump is equipped with dual seal pressurized mechanical seal "Flowserve make" API53B plan
seal chamber pressure will vary from 6 barg to 72 barg according to the number of pumps running in series. the barrier fluid is potable water and pressure is 85 barg for all pumps.
problem is that; all mechanical seals are failing frequently showing very high external and internal leakage.
Several Design modifications have been done by seal manufacturer without any improvement.
Now, I am thinking to switch to different API plan other than API53B in order to avoid the high differential pressure across the seal faces.
what about 53C??
what about dual containment seal??
what are the other alternatives for this problem?

 

[ronfrend]

Meshry, I have some questions
1) Can I ask why you are using potable water as the barrier fluid? It is more usual to use a hydrocarbon with better lubrication properties on a crude oil pump.
2) I also assume that when you say the seal chamber pressure varies from 6 to 72 barg that this is not the same across each pump, but that each pump will see a consistent pressure. Is that correct?
3) Could you also advise if you have a balance piston on these pumps? If you do then the seal pressure should be a heck of a lot less than 72 bar.
4) If the pumps are running in series then each pump will see a different seal chamber pressure so why do you have 85 barg as the barrier pressure on all pumps?

Flush plan 53C will track the seal chamber pressure so the barrier fluid is always the same pressure above the seal chamber giving a consistent pressure drop across the inboard seal but remember that the outboard seal will see continual changes so you might just be moving the problem from one seal face to the other.


Ron Frend

http://www.predicon.net

 

[Meshry]

Thanks Ronfrend for your reply.
1) potable water as a barrier fluid was decided by seal manufacturer (FLOWSERVE). do you think this has an impact on seal performance?
2)Seal chamber pressure will be consistent for pump no1 and pump no2 i.e. for pump no 1, the seal chamber pressure will be 6 barg (booster pump discharge), for pump no2 the seal chamber pressure will be 35barg which is the discharge pressure for pump no1.
for pump no3 and no 4 the seal chamber pressure will vary according to how many pumps are running down stream i.e. if we have pump 1 and 3 online then seal chamber pressure for pump no 3 will be 35 barg. if pump no1 and no2 and no3 are online then the seal chamber pressure for pump no 3 will be 72barg.please note that, the operator will decide the number of the running pumps (in series) according to the required daily production. sometimes 2 pumps and other time 3 are running.
3) No, there is no balance piston, it is single stage pump with double suction impeller.BB arrangement.
4) I would say ,85 barg seal barrier pressure was decided considering the scenario of 3 pumps running in series whereby pump no 3 will have seal chamber pressure of 72barg.also I believe the designer preferred to standardize the mechanical seal for all pumps i.e. same part number.

if 53C will not solve the problem, what is the other option? on other facility we have double seal with containment seal and it is working fine!!

 

[georgeverghese]

Have you checked the operation of the barrier fluid pressurisation pump ? Is there a leaking recycle PCV or RV on the dicharge of this pump ? Is there some way of confirming net forward flow of potable water into the double seal ?

 

[ronfrend]

Hi Meshry. Have alook at SEALING SENSE from Oct 2013

http://www.fluidsealing.com/sealingsense/Oct13.pdf

The paragraph on water as a barrier fluid describes the temperature limitations - it freezes at low temperatures but at higher temperatures water is a terrible lubricant and will cause seal failures. When using water it's more usual to use a water/glycol mixture.

To confirm this (or otherwise) I suggest you have a close look a the carbon rings from the failed seals. As you say you have failures on inboard AND outboard seals there seems to be something basically wrong with the installation or design. If there is a compatibility issue between the water and the carbon you will see pitting if it is a chemical incompatibility. If it's a lubrication issue you should see signs of overheating and/or surface distress. As you are having multiple failures you really should be carrying out a root cause failure analysis.

Suggestions:
1) Try putting a pressure transmitter (piezoelectric) into the F connection to make sure the seal chamber pressure is what you think it should be. Look at static and dynamic pressures to see what the real world situation is. I would be looking particularly for pressure transients.
2) Install a thermocouple to check the actual operating temperature of the barrier fluid.
3) Get a lab analysis of the barrier fluid water to make sure there is no unexpected chemistry (including pH and dissolved oxygen).
4) Check alignment and runout of the shaft. Anything more than 3mm/s rms axially is going to cause rapid seal failures.

That should keep you busy for a day or two I really don't think you should be considering moving to a 53C until you have all the data you need. GeorgeVerghese made a good point in checking the operation of the pumping ring. I have seen these installed incorrectly so they ended up pumping against a virtual dead head.

Let us know how you get on.

Ron Frend

http://www.predicon.net

 

[Meshry]

Dear georgeverghese ,
No, there is no PCV or PRV installed in the barrier fluid closed loop. only PRV installed on the barrier fluid make up header i.e. down stream the make up pump. this PRV has nothing to do with barrier fluid circulation.

no other way to observe the barrier fluid indication.
only the barrier fluid temperature can give us an indication about proper circulation of the seal barrier fluid. without this circulation thru the cooler, this temperature will shoot up.

 

[Meshry]

Dear ronfrend ,
Thanks for the great information about water as barrier fluid.
The root cause analysis along with all the advices you have given already done and verified by seal manufacturer, this is why they have changed the design for more than 7 times!!!
the incompatibility was not one of the findings.
I can tell you that; this is mainly related to the higher differential pressure across the seal faces and specially for the outboard seals. the seal faces always found worn out.
the seal manufacture has changed the balance ratio up to 75%!!!!
in our facility we have around 300 53b PLAN scattered in different applications and working fine, however with lower pressure values i.e. seal barrier pressure.

in this case and if 53C is not the salvage. what are the other possible API plans which may reduce the differential pressure across the seal faces? now I am thinking of API 52.

 

[ronfrend]

Hi Meshry, If the seal faces are wearing out and the hydraulic balance is not too high and the barrier fluid is a good lubricant then the carbon is too soft a grade. Go back to Flowserve and ask them for a harder wearing carbon. You then run the risk of the carbon cracking more easily but it should last longer.
DO NOT go to a plan 52 for this application. All the pressure drop would be across the inboard seal and you run the risk of crude oil blowing out the top of the header.
I have seen seals running at 100barg across each seal without problem. I'm still convinced that the problem is due to poor lubrication of the barrier fluid. Try something like a kerosene or diesel. I've had good success with diesel in the past.

Ron Frend

http://www.predicon.net

 

[Meshry]

Thanks for the valuable information.
I will try with different fluid as advised.

Thanks

 

[georgeverghese]

Sorry, my previous response was based on the impression you have plan 53C now.

For plan 53B, you have the seal pot at 85barg for all pumps from your narrative? So, for the lower stage pumps, you have a very large dp between seal supply pressure and seal chamber pressure - what did Flowserve have to say about this ? Typically, we have only 3-4bar dp only ( maybe no more than 7bar dp) between seal fluid supply pressure and seal chamber pressure. Must say I've not seen such a high dp before.

Agreed, either the potable water may not be pure enough, or there may be some other mechanical issues. Else, see if you can come up with some other pumping arrangement that will keep each pump running at a fixed suction pressure that will be matched by a suitable seal supply pressure?

 

[Meshry]

Dear Georgeverghese,
Yes fully agree with you, this is too much DP across the seal faces particularly the first pump in the train.
this is why I started to think about other plan where this DP is not there.

 

[georgeverghese]

Else may we can set up a control scheme that will vary pump seal pot N2 supply control valve set pressure to be 3-4bar higher than the expected suction pressure ? That may be the easier solution if high dp is one the root causes of this problem and can be done relatively quickly.
The quality of the potable water may also need to be checked - else the lower risk approach may be to go for some better quality seal fluid than potable water as suggested.

 

[Chpal1984]

Hopefully you've resolved this problem by now, but here is my 2 cents anyway.

It would appear to me that you've found the most likely fault, high pressure differential. A plan 53C should solve this problem over the inboard seal, however over the outbound seal you have a 85 barg pressure differential which is a high pressure for a seal with a contacting design (not impossible). If your not worried about the barrier fluid contaminating the process, then water would not be my first choice as a barrier fluid (I would choose a light oil).

In theory you could use plan 52 with pressure drops staggered over multiple seals, however I agree with ronfrend that you would need to consider the safety implications of this.

 

[Meshry]

Dear Chpal1984 ,
Problem still existing. Seal manufacturer has come with new design with some modifications. mainly playing with the balance ratio and seal face geometry. new seals still under installation and performance yet to be evaluated.

here I want to ask... if our problem is mainly with the higher DP across the seal faces (contacting), how would changing the barrier fluid enhance that???

 

[Chpal1984]

I think the main reason we tend to use oils is because the water has a narrow temperature range it can be used for. So using an oil as a default, generally means you don't have to worry about it as much.