Minimum Flow Issues, Seals Failing Quickly...

[Riley15]

About 6 months ago our client installed a Sulzer APT 53-6(O), a pump I selected, and we have been receiving reports from the field that they are having issues with it. It operates 3 gasoline truck loading arms. The problem is when one arm is operating the pump sounds like a concrete truck mixing...
We have heard that when 2 or 3 arms come on that the loud noises subside; based on that fact, an NPSH issue for this pump was ruled out because a higher flow would make NPSH issue worse, not make it go away.
The pump seems to be suffering from MCSF issues. I understand that with one gasoline loading arm on (600 gpm) it is operating outside the preferred operating region (see attached pump curve) which can shorten seal life but they have been through 3 seals since the pump became operational 6 months ago. The pump operates on a VFD and there is a control valve downstream programmed by an Accuload.
Anyone have any suggestions as to why this is happening?

 

[adammal44]

Based on your pump curve, it seems pretty straightforward that you're seeing issues with over-recirculation. At the minimum, you could be having troubles with vibration and flow irregularities.

You could also be aggravating NPSH(available) issues if the over-recirculation causes overheating of your fluid, probably unlikely though considering that your system is fine at a much higher flow.

Why did your vendor specify this pump for you if 1 arm/600 gpm was an operational case?

 

[JJPellin]

I would look at this in terms of operating flow relative to BEP flow. With three arms in operation, you are running at about 80% of BEP (1800 versus 2250). With two arms in operation, you are running at about 69% of BEP (1200 versus 1750). With one arm in operation, you are running at about 48% of BEP (600 versus 1250). That is as low as we would ever recommend running a pump with a suction specific speed greater than 10,000 (US units). I suspect the pump is experiencing suction recirculation cavitation as a result of operating too far below BEP. This can be very difficult to suppress with excess NPSH margin. Is the suction piping well designed in terms of straight run upstream of the suction flange?

Johnny Pellin

 

[Riley15]

Johnny,

It has basically the minimum amount of straight run pipe; five pipe diameters before there is a 90° bend. Attached is a few screenshots of the piping layout. Can you elaborate on, "This can be very difficult to suppress with excess NPSH margin?"

 

[JJPellin]

Preventing cavitation from suction recirculation might require NPSH available to be several times higher (multiples) in order to stop the cavitation. An NPSH margin of 5 feet might be adequate to suppress standard cavitation. To suppress suction recirculation cavitation you might need an NPSHa that is 3 to 5 times that of NPSHr. In most of our applications, this would be completely impractical.

Johnny Pellin

 

[georgeverghese]

The low speed operation at 685rpm is at about 50% of the full speed operation of 1200rpm, which is about as low as you can enable on an variable speed electric motor without an auxiliary fan for motor coil cooling.
Agreed, it does sound like this pump is operating at below MCSF of 400gpm or so when only one loading arm is in operation,since you've ruled out NPSH issues here.

Either reset the control valve at the loading bay to keep this pump above 400gpm during startup, or install a min flow recyle line with return to source tanks.

The flow transmitters at the loading bays should tell you if the min flow of 400gpm is exceeded at startup??

The short life on the shaft seals may or may not be related to this MCSF problem, given that these seals are in volatile gasoline service.

 

[Artisi]

Something amiss with what you are telling / not telling us.
600 gpm is not outside the operating range at reduced speed. I would suggest the blue envelope is the operating range at the higher speed.
What is the pump running speed at the 3 flow conditions?
How is speed controlled for the 3 different operations?
Are the operators reducing speed for the lesser number of loading arms or are you assuming they do?
Can see no good reason for failures if the unit is installed correctly and operating at the speed selected for each condition.
Without knowing all facts I would say the pump is not seeing any speed reduction during operation.

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.)

 

[LittleInch]

Please remember that NPSH relates to 3% headloss. Onset of cavitation can occur at much higher values and tends to be higher at both left hand and right hand end of the curves. Ask sulzer for the 0% / onset of cavitation curves.... So don't be too fast to dismiss cavitation until you get the data.

might be something like this

I'm also puzzled by the following:
What is the pump VFD controlling to - flow? pressure?, fixed speed?
What is the control valve controlling to? flow? fixed pressure?
The flow friction seems to be quite high??

Do you have confirmed operating data ( flow rates, temperature, pressures etc?) How does the arm and pump operate? Often flow is reduced at the start and end to avoid splashing and overfill. If the pump also comes on and stays on for a while at the beginning and end this won't be good.

Basically though it seems you have specified a pump which cannot really do the 3:1 turndown as it is just a bit too big and the turndown is hard to achieve.

So either vary how it works, add a re-circulation line to up the minimum flow when only one arm is loading / shut in head.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Artisi]

The pump is more than capable of operating at the flow / head duties as shown on the 3 curves when operating at the nominated speed as indicated on the curve/ data sheets.

Would say there is absolutely nothing wrong with the pumps, with the selection, and is not an NPSHa/r problem, the problem is in the operation and the way the pumps are being forced to operate. Are they running at reduced speed for reduced flows, are they running against a closed valve during change-over of the trucks (more than likely YES)

It's the old never ending always touted story - "there is something wrong with the pump" wish I had a $100 for every time I have heard that utterance when in fact the pump was innocent of the charge - I could now be living like a king in my retirement.

CHECK THE OPERATORS, DON'T BELIEVE THEM, CHECK YOURSELF.

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.)