Problem

Is there a rule of thumb for deciding if the volume of vapor or number of collapses is within the scope of what can accurately be simulated in
HAMMER?

I'm seeing large vapor pockets forming - how do I know if HAMMER can accurately simulate this?

Problem ID#: 72094

Solution

The literature on the Discrete Vapor Cavity Model references a rule of thumb: "the ratio of maximum cavity size to reach volume should stay below 10%". Given that HAMMER uses the Discrete vapor cavity model and that the literature supports this 10% rule of thumb, this is a reasonable rule if you are comfortable with it.

However, generally speaking it is probably best to avoid recommending any rule of thumb with regards to acceptable vapor pocket size or number of collapses. Due to the chaotic behavior of liquids in a cavitating system, it is very difficult to measure the high-frequency pressure spikes with any kind of accuracy; moreover, operators of such systems are generally reluctant to subject their pipe networks to these potentially damaging conditions. Meaning, the presence of large vapor pockets or multiple collapses usually tells you that you need to add some protection strategy.

Unfortunately, it is equally hard to formulate rules of thumb for determining acceptable sizes for vapor pockets in the context of a HAMMER run. The 10% rule of thumb referenced above seems quite conservative, but we just don't have enough physical evidence or case studies to say that it will always be okay.

Therefore, we ultimately concluded that you'll need to apply your own judgement in this case. It is OK to push the software to its limits, but we feel that if you do that, you'll need to assume the risk.

Note that in the presence of multiple vapor pockets the Wave Speed Reduction Factor may be appropriate. In real systems it is understood that entrained air is released at low pressures slightly above vapor pressure. This phenomenon has the effect of reducing the wave speed during the period of time that the pressure is low. A reduced wave speed will normally mitigate the severity of upsurges upon collapse of cavities versus the basic cavitation model (full wave speed). By neglecting the decline in wave speed (the default factor of 1.0), the extreme pressures generated by vapor pocket collapses are conservatively overestimated. For more information on this, please see the help topic for wave speed reduction.

Also note that the "reach volume" referenced in this rule-of-thumb would be the full volume of the pipe in which the pocket formed. Find the pipe length (consider including any "length adjustment" if that is significant, which you can view in the pipe FlexTable) and multiply by the pipe cross sectional area to get the reach volume.

The "Vapor Volume (Maximum Transient)" or "Air Volume (Maximum Transient)" fields in the Transient results section of the properties of pipes should give the max volume in that reach. So you can typically divide this number by the reach volume.

If a vapor pocket forms at a high point at the intersection of two pipes, or in the case of an air valve, you would need to treat it as two separate reaches/checks since there may be vapor/air in each pipe (on each side of the high point.)

The one problem with using the "Vapor Volume (Maximum Transient)" or "Air Volume (Maximum Transient)" fields is they won't tell you much if you have multiple vapor/air pockets along a pipe. If that appears to be the case after inspecting the profiles it may be necessary to check the "*** LIST OF SORTED VAPOUR AND AIR POCKETS ***" section in the Transient Summary report (Reports > Transient Analysis Reports > Transient Analysis Output Log) to see the max vapor/air volumes along the pipe. You may wish to add these volumes up and check that they don't exceed the 10% rule of thumb (it is possible that these maximums don't all occur at the same time, but we don't provide results of volume versus time at interior points along a pipe).

See Also

Assumptions and limitations of tracking air or vapor pockets in HAMMER