Corrosion / Errosion within Seal Flush Pump?

[kmg1983]

Hello everyone! I'm a young, relatively new engineer and have encountered a problem with one of our seal flush pumps. It is a single stage, horizontal centrifugal pump operating @ 285 gpm and 185 psi. It receives suction off of a potable water tank and is used to flush the mechanical seals of several other larger pumps. We tore the pump apart noticing increased amount of noise and saw extreme corrosion and pitting within the volute case, impeller, and stuffing box, as well as within the suction pipeline. My question is, could this also be a case of cavitation even though there is more than enough NPSHa comming off the potable water tank? All the parts are carbon steel.

 

[GPRnD]

The first question would be whether the pitting/corrosion is over the whole wetted surface or just in specific areas (such as the impeller vane surface, just back from the leading edge)? Cavitation will cause the latter.

Posting a picture of one or more of the parts would help.

 

[pumpking]

How many pump seals are you flushing ? What are the stuffing box pressures of thge pumps (rule of thumb 80% of discharge pressure).

Immediately from your post, the flowrate and pressure sound high, meaning that what may be happening is the pump is hanging on the end of the curve. Alternatively, it could be that the circulating pump is not generating sufficient pressure and the seal faces are weeping product into the flush liquid - normal recomendation is to have seal flush liquid 1.5 bar above stuffing box pressure to avoid faces lifting off - seal and pump manufacturers can advise more specific figures for these !!

It may be worth looking into how the pumps operate, are they working at constant flow, or do they move up and down the curve during operation ? If they do move, you may want to investigate and ensure you are not putting any 'hammer' back to the pump - the effect of this again is to lift process end seal faces allowing product into the fluch system.

Hope this will be a starting point for you - alternatively, buy a mag-drive from

http://www.cdrpumps.co.uk

????????????

 

[kmg1983]

The pitting / corrosion appears to be over the entire wetted surface. This particular seal flush pump is providing seal flush for 11 other larger pumps, with suction pressure around 160 psi. As for the flow and discharge of the seal flush pump, those are what the pump was designed at and it operates at constant flow, therefore, it is not moving on the curve.

 

[Artisi]

some corrosion is probably fairly normal in a carbon steel pump, however if you have erosion taking place within the pump this must mean that there are abrasive solids in the pumped water. If this is the case, then the erosion coupled with the corrosion can be a vicious cycle - the erosion continually removes the protective oxide layer(rust) which leads to accelerated erosion / corrosion damage especially if there is also a pH problem as well.
Check the water for fine solids and also check pH.

 

[kmg1983]

Thanks guys for all your help and input! It is greatly appreciated.

 

[PumpSmart]

kmg1983,
I recommend to check all possible reasons as follow:
Pitting corrosion could be option, you should check pump materials for suitable seleccion and realize if there is any corrosion agent, ph and temperature in your pumping fluid. From Ductil Iron to SuperDuplex materials are avilable for Corrosive hanling fluids.
If it is erosion could be option, you should check tip velocity, max RPM, % in solids by weigh , D50 and Casig/Impeller hardness to determine if your type of pump could handle this fluid.
In another hand, it is an standard rule that if NPSHa > NPShr then there is no cavitation (that is not always true).
Cavitation for recirculation at discharge or for high velocity are present some times.

 

[NVAPORMAN]

I have a quick question,Is the potable water spiked with Chlorine?

 

[bk19702]

You can calculate suction energy of the pump in question, which is velocity inside the impeller more or less. This can result in cavitation even though NPSA>NPSHR.

 

[hbw]

Check the pH of the pumpage. It sounds like you have some acid entering the fluid stream. A ph of less than 7.0 is acidic and will cause corrosion. Velocities in a pump are generally sufficient to wear of the passivating patina on metallic surfaces, exposing fresh metal to corrosion.

Try API S6 materials.