Subscribe to our 

Years ago we built, tested and shipped a double ended pump only to have the customer call and report they had a hot thrust bearing. Checking test reports the thrust bearing temperature was normal during test.

Working with the customer they changed the thrust bearing, checked the oil for proper level, viscosity, and contaminates. After restarting the pump the trust bearing temperature was over 220F.

At that point it was decided to return the pump and place it on the test stand. Having all the confidence in the test report and pump, I was surprised to see the thrust bearing temperature rise over 200F.

During the next four days we changed bearings, checked the bearing fits, oil level, viscosity and everything else we could think of. Each time we ran the pump thrust bearing temperature would rise above 200F. Each time we analyzed the bearings they indicated a high axial load in one direction. However, with the pump having a double suction impeller mounted between bearings it was hard to accept the signs of high axial load.

After checking all the possible reasons for bearings running hot except the load on the bearings I decided to revisit the loads on the bearings. The pump was a double suction/double ended pump which should have had a very low axially thrust. The only axial load should have been from slightly different size wear rings which were intentional to keep the rotor from shuttling during operation.

On the way to work the next morning I realized I needed to “Step out of the woods and look at the trees”. In looking at the trees I remembered an old saying in hydraulic design, “high velocity low pressure, and low velocity high pressure.” Appling this to the pump design I realized that if the impeller was not centered in the volute the pressure acting on the sides of the impeller would be different.

Arriving at work I had the pump disassembled, then added some pattern makers clay to each side of the impeller and reassembled the pump. Knowing the clay would tell us the clearance on each side of the impeller we disassembled the pump and measured the clay. To our dismay we found a significant difference between the clearances on each side of the impeller.

After centering the rotor and reassembling the pump, it was back to the test stand. The results were normal thrust bearing temperatures.

Now that we had found and corrected the problem, we needed an explanation for the difference between the original axially setting and the axial setting of the shipped pump.

Our normal procedure after testing was to disassemble each pump, clean and reassemble the pump for shipment. During reassembly of the pump, the assembler misplaced the spacer between the shaft and the thrust bearing, which sets the axial position of the rotor. He simply got another from stock and assembled the pump.

The problem was twofold. First, the assembler should have checked the axial position of the rotor and machined the spacer to the correct length. Second, the spacer should have been designed so the pump could not be assembled without machining the spacer.

To ensure this did not happen in the future, the assembler was made aware of his mistake and the spacer was revised so it had to be machined.

The moral of the story is to make sure the impeller side walls are centered in the volute. Note: Simply pushing the rotor each way in the pump and centering it may not ensure the impeller side walls are centered in the volute.

A lesson learned sixty years ago. When I was a little boy in the late 1940’s my father drove a 1939 Ford coupe, new cars were not available because of the war effort. The Ford coupes were good cars for the day however the windshield defrost system was almost nonexistent. Everyone added a small rubber bladed fan to the dash that blew air across the windshield acting as a defroster.

 Most fans were powered by 6 volt systems as 12 volt systems were not yet standard on today’s cars. Seatbelts were not even used in race cars let alone in passenger cars and kids road in the front seat when there was room. Many families only had one car which dad drove to work leaving mother home to take care of the house and kids. Many women didn’t even have driver licenses as there was no need. However, women whose husband was fighting in the war did drive and work outside the home to support the family and war effort.

 Enough of history — let me get back to the lesson. One cold winter day the family was going to town on Saturday for the weeks shopping trip and I got to ride in the front seat. Standing on the seat between mother and dad I was told not to put my fingers in the fan. Being a typical little boy what do you think I did? Of course I put my finger in the fan, now remember it was rubber so I didn’t get hurt only a sore finger. Years later I related that experience to NPSH.

 How does a sore finger relate to NPSH? The amount of energy I used to push my finger into the fan. NPSHr (required) is like the energy it takes to push liquid into the impeller eye of the impeller and just past the vane tips. NPSHa (available) is the energy in the piping system that is used to push the liquid into the eye of the impeller and just past the vane tips. That is how a sore finger years ago relates to NPSH.

So for non-engineers you now have an explanation of NPSH that you can use to explain NPSH to an engineer.