Pump problems.... cavitation?

[jmw]

I have a pump running at around 20g/min and fluid is hydrocarbon at 150cst.
The pump has a 2" strainer in the inlet line but the line appears to reduce to 1" either side so pump inlet is 1" and discharge is 1".
Between the pump and a sensor there appears to be several discontinuities due to Ts, and fittings and one elbow immediately at the discharge.
I'd estimate the distance from pump discharge to sensor to be no more than two - three feet.

The problem appears to be either cavitation or gas break out at the sensor.
I am concerned that the pump suction may be causing cavitation and/or gas break out and that this is then pumped through to the sensor causing disrupted measurements.
I have a rule of thumb which suggests:
"To avoid cavitation the transducer should be installed a distance of 3 x pump velocity downstream of the pump –i.e.if pump velocity is 1 m/s transducer should be 3 m downstream of the pump"

If there is more information required, please let me know. I suspect that the elbow is causing some swirl.

At pump stop the sensor is accurate and stable. With the pump running the readings are seriously disturbed (the sensor is sensitive to gas/cavitation).
If back pressure is increased some improvement is noted.

Would I be right to suspect that I have either or both cavitation and gas break out in the fluid and that I need some more distance from the pump discharge to the sensor and to have 2" suction lines throughout?
Any other suggestions?


JMW

http://www.ViscoAnalyser.com

 

[willhammett]

It may very well be the strainer, I have had problem in the past with improperly installed vortex breakers due to the excess drag forces and low pressure pocket being "pulled". (it was also an aerated liquid at 1.3 sg, fairly viscous) Does it sound like cavitation? Is the discharge pressure swinging?

 

[jmw]

That's one of the things I can ask them to check. I'm not sure why they have a filter in the line, to protect the pump I guess, nothing else needs it. Maybe they should have put in a better filter.
It suggests I can get them to pull the filter element and see what that does. (if the filter is doing nothing much, they can leave it out. If it is doing something then chances are the operators will pull it and leave it out anyway to save routine getting dirty to clean and replace. A "Y" type is probably not intended to be more than insurance anyway.)

JMW

http://www.ViscoAnalyser.com

 

[willhammett]

That's true, we try to stay away from y type strainers except on steam condensate and water applications that pose no chemical health/environment threat for routine blowdown (hourly intervals)

 

[jmw]

I have had some information.
This is actually crude oil with distillate and I am wondering if this combination of strainer and some other suction line issues may be causing the volatiles to gas off in the suction and then, in a 150cst fluid, get carried through.
Any thoughts?

JMW

http://www.ViscoAnalyser.com

 

[willhammett]

That's quite possible. Entrained vapours have a tendency to create havoc on centrifugal pumps when not sized specifically for the liquid in the application.
And having a very viscous liquid means a deaerator typically won't work. We have one on a type of a liquid at 1.25 sg and 89 cp. if removal of the strainer won't work you may have to resize the pump, or switch pumps. Go from a centrifugal to a small moyno? You won't get the same flow rate but you could get a larger pump.

 

[rmw]

Cavitation by definition occurs when a fluid entering the suction of a pump falls below its vapor pressure or the vapor pressure of the lowest vapor producing component in the fluid.

In the case of "crude with distillate" (I don't know if that means crude that will ultimately be refined to produce distillate or crude that has been cut with distillate), I suspect that there are high end volatiles that are coming out of solution as the fluid enters the low pressure zone of the pump.

Crude is like water with respect to vapor pressure, except that with water you have but one fluid to deal with. With crude, well, after it is fully refined, we all know how many different "fluids" are in there, and we (the general public) don't typically even see the light volatiles in the naptha range and above.

Get rid of the strainer. I think that is your culprit.

rmw

 

[jmw]

It is a heavy Crude cut with distillate to meet pipeline targets of density and viscosity.

Thanks for the input guys, I have now suggested that they try running without the filter and see what happens. I doubt it will be a complete cure because there are some other issues with the suction line but if it shows a significant difference, especially with increased back pressure, then we can focus on the pump installation.



JMW

http://www.ViscoAnalyser.com

 

[chief]

I note the suction and discharge pipe sizes are the same. It is normal design philosophy to have the discharge size smaller. Reduce your discharge by fitting an orifice and consider a large hole size on the suction filter screen. Cavitation is caused when there is low pressure on the suction side of pump. too high fluid velocity thru' pump, restriction on suction side etc,

Offshore Engineering&Design

 

[jmw]

Thanks Chief.
I'm waiting to hear the outcome of running without the filter and if this confirms the cause then they will have to redesign the pump pipework.


JMW

http://www.ViscoAnalyser.com