NPSHR reduction with rising temperatures

[25362]

The relevant chart appears on:

http://www.mcnallyinstitute.com/Charts/NPSH_reduction_chart.html

Karassik et al., in their Pump Handbook (McGraw-Hill) show this chart (constructed from test data) saying that it could be used subject to some listed limitations.

I'd appreciate to read your opinions about the pertinent reason(s) for those NPSHR reductions (meant for pure deaerated liquids).

 

[BigInch]

Experience has confirmed that (esp.) hydrocarbons do not cavitate precisely according to what one would expect when deriving NPSHR based primarily on their vapor pressures.

Independent events are seldomly independent.

 

[25362]

True and thanks. I'm after the underlying reason (s). I assume it has to do with the vapor specific volumes (L/kg) decreasing as temperatures rise.

For example with butane, when comparing its (saturation) properties at 30 and 50^oC:

Vapor, L/kg; latent heat, kcal/kg; sound speed in liquid, m/s; pressure, atm; are respectively:

30^oC => 140.1, 85.2, 874, 2.8
50^oC => 81.7, 79.8, 767, 4.9

We see a reduction of more than 40% in specific volume, and only about 6% in latent heat. A small reduction of sound speed in the liquid of some 12% may also help in reducing the impact of shock waves when bubbles implode.

I'd much appreciate reading yours and other experts' opinions on the subject.

 

[25362]

BTW, this strange effect doesn't mean that by heating the fluid stream we avoid cavitation, since the NPSHA drops (vapor pressures rise) by rising temperatures, probably even more than the allowable decrease in NPSHR.

 

[BigInch]

I have never seen this topic discussed to the point where reasons were given, but I would agree that the latent head and sonic velocity reductions are behind at least some of it. Both are always mentioned as contributors to cavitation damage and their reductions would lessen the amount of available energy. The relatively lesser sonic velocities for petroleum liquid products in relation to water would be consistent with the general practice of reducing NPSHR by 2 to 4 feet or so. What are the corresponding reductions for water?

Independent events are seldomly independent.

 

[25362]

This is what Karassik et al, say in their Pump Handbook, about the subject:

"…If the suction system may be susceptible to transient changes in absolute pressure or temperature, a suitable margin of safety in NPSH should be provided. This is particularly important with hot water and may exceed the reduction that would otherwise apply with steady-state conditions."

An exercise with water between 30 and 90°C shows, respectively:

Vapor pressure, atm: 0.0419 & 0.0693;
Saturated vapor, m³/kg: 32.9 & 2.36;
Latent heat, kcal/kg: 600.7 & 545.5;
Sound speed, m/s: 1509 & 1556.

Note that sound speed rises with temperature.

For 90°C the allowable reduction would be about 0.7 ft. According to the graph, for 150^oC, it would be about 4 ft.

 

[BigInch]

I would be very careful reducing NPSHR with hot water. It seems that there are always more than enough problems with cavitation in hot water pumps without applying any reduction.

Water is indeed a strange liquid. This shows a slight reduction of sonic velocity between 150 and 212F.

http://www.engineeringtoolbox.com/sound-speed-water-d_598.html

Independent events are seldomly independent.

 

[25362]

You are absolutely right, indeed, water properties are weird.