Engineering investigations into the causes of wear, erosion and metal loss on the back-plates behind the impellers of two year old centrifugal pumps pumping water about an irrigation system
We have just commenced our winter maintenance schedule in the irrigation water pump stations and have found some serious wear issues occurring behind the pump impellers. I have reviewed documentation on pump cavitation and cannot see anything that looks similar to what we are seeing on our pumps. Photos of the problem are in this PDF document of the typical condition of centrifugal pump back-plates. There is very little wear occurring on the brass impellers, the volutes or on impeller vanes, however, we are seeing extreme gouging and wear to the cast backing plates on the motor side of the impellers. Are you able to shed some light on this ?
The badly affected outer annulus showing on the back plate under the wear ring end is most likely due to high differential pressures between the discharge and suction sides of the impeller. The balance holes in the impeller are at the low suction pressure of the impeller inlet. Water will always flow from the high pressure discharge side of the impeller and past the wear ring then out through the balance holes back to the low pressure suction. The wearing ring combines with the backing plate, impeller wall and shaft to form a chamber and water will swirl about the chamber at faster and faster speeds as the differential pressure between discharge and suction rises. Recirculation erosion and cavitation are both definitely possible, especially at low pump flow conditions.
If the gap between the backing plate and the end of the wear ring is small (say less than 2-3mm) I would expect erosion from high liquid velocity as well as cavitation when a pressure differential causes recirculating water in no flow and low flow situations. It would be worth measuring the distance of the gap between the back plate and the end of the wear ring. The fact that there is a concave scallop on the end of the wear ring supports a high liquid velocity cause.
I would also expect to see evidence of wear/erosion on the back plate wear ring surface where the impeller wear ring runs inside it due to high water velocities. Please also check the clearance distance between the impeller wear ring and the back plate wear ring sealing surfaces to see what they are when the pump is assembled. The gap needs to be tight—around 0.1 mm. If there is more than 0.1mm gap there will be a great amount of recirculation during dead-head and low flow events. Also look with a magnifying glass at the back plate wearing ring surface for signs of erosion or preferential tracking from high pressure to low pressure.
The deeply affected annulus around the shaft looks like erosion. It is likely that localised eddy flows swirl in the chamber corner around the shaft as the water moves from high pressure at the wear ring to low pressure at the balance holes. If the shaft is made of stainless steel then preferential galvanic corrosion of the cast iron becomes another contributing possibility.
Since the irrigation pumps pressurise to 650kPag and maintain that pressure in a water distribution system there will always be high pressure on the discharge side of the impeller. If you run the pumps against that pressure you will always create recirculation across the wear ring and get the wear patterns seen in the back plate. To stop the back plate erosion and/or cavitation you must stop running as many pumps as possible when there is little or no flow in the distribution system. You only want to run a pump when there is draw-off from the distribution piping.
If there is a galvanic problem present as well, regulating the flow characteristics will not change that failure mode. Next year sample a pump from each station and strip it down as you have done in the current refurbishment. If there is still damage to the back plate after making process control modifications to prevent pumping at no flow and low flow conditions then galvanic problems are a real possibility.
I would also send the photos of the pump wear to the pump OEM design engineers and get them to make a determination of the possible causes of the problem. They designed the pump and will know the likely causes of such problems.
Ask the OEM for replacement back plates with better wear and cavitation resistant properties.
You may still have another problem. This one is financial. If you keep the pump configuration as is, then potentially every two years you will spend something like $2,000 (for parts and labour) on every pump in your system to refurbish it with a new back plate. Over the life of the pumps it will total to a lot of money. You might want to think about using alternate materials in future (both galvanically compatible and harder wearing against erosion) if next years inspection shows that a problem is still present.
My best regards to you,
Lifetime Reliability Solutions HQ
P.S. If you have maintenance engineering advice on industrial equipment maintenance, especially defect elimination and failure prevention of plant and equipment, or have made successful equipment reliability improvements, please feel free to send me your articles to post on this website. You can contact me by email at