Talk about a few common & important processes, show them on the T-s diagram, and talk about GUNNS limitations:

Whenever we talk about fluid processes we have to consider “work”. A common use for fluids is to do work on its surroundings, or have work done on it by its surroundings. Whether or not work is involved in the process is a factor in what kind of process we consider it to be and how we model it.

The simplest way to think about fluid work is whether or not the volume of its surroundings is changing. Examples are:

• Pistons:
  • When the fluid pushes on a piston and causes it to move such that the fluid volume increases, then the fluid is doing work on the piston.
  • Likewise, if the piston is compressing the fluid into a smaller volume, the piston is doing work on the fluid.
• Compressors & turbines:
  • This is similar in concept to pistons but a little harder to visualize.
  • When fluid passes through a turbine and the turbine is spinning, the volume swept by the turbine blade is similar to the volume change by an expanding piston, and this is experienced by the fluid as a volume increase — thus the fluid is doing work on the turbine as it flows through.
  • Likewise, the compressor decreases the fluid’s volume and does work on it.
• Pumps & fans:
  • These are the same thing as a compressor. They do work on the fluid.
• Rocket nozzles:
  • As the fluid flows through the expanding bell of a rocket nozzle, it experiences a change in volume and does work on the bell, which translates into thrust.

Isenthalpic Process: a process on the fluid that doesn’t change its enthalpy. This happens when there is no heat transfer to or from the fluid and no work done on or by the fluid. The “throttling” process is the most common example — it’s essentially any flow through a valve or orifice with no heat transfer or work done. The drop in pressure across the orifice causes a corresponding change in temperature at the same enthalpy.

This is shown by the path from points A to B in the T-s diagram below:

Here we start with water vapor (steam) at temperature of 400 °C, pressure of 5 MPa, and specific enthalpy of 3200 kJ/kg. This steam passes through a valve across which its pressure drops down to 1 MPa. Following the constant enthalpy line at 3200, it intersects the 1 MPa pressure line at a new temperature of 370 °C, and this would be the actual temperature of the steam exiting the valve.

In GUNNS, we can’t model this effect. A → B is how it works in the real-world, but in GUNNS we can only do A → C. Since our enthalpy is only a function of temperature and nothing else, our lines of constant enthalpy are also at constant temperature, and horizontal on the T-s diagram. You can see that at low pressures, the ideal gas already has horizontal enthalpy lines anyway, so the temperature change of isenthalpic processes at low pressures is very small, and the simplification of our enthalpy is appropriate and doesn’t affect us. However at higher pressures, the simplification has a significant impact, and we don’t get he expected temperature change. In most applications, we don’t need to model this affect anyway so this limitation rarely comes up. Here is further discussion of the limitation and how to work around it.

Isentropic Process: an idealized process on the fluid in which:

• there is work done on or by the fluid but it is frictionless and there are no losses due to friction,
• there is no heat transfer to or from the fluid (the process is adiabatic),
• the process is reversible.

We won’t go into the definitions of all of the above terms. Instead, the big takeaway of all this is that the isentropic process is a good way to model processes in which there is work involved, i.e. in pistons, pumps, nozzles, etc.

Consider the above diagram again, at the same staring point A. The gas expands to the same exit pressure as the previous example, 1 MPa, except this time the steam is driving a turbine. This is an isentropic process, so we trace a vertical line along constant entropy to the exit pressure at point D, and end up with an exit temperature of 191 °C.

A good way to think of the isentropic process and the work involved is as follows:

• When the fluid is doing work on something, say driving a turbine, we are extracting more energy out of the fluid than if it were just passively flowing through the passage and the turbine blades weren’t spinning at all (which would be more like an isenthalpic process).
• This extra energy from the fluid is going into mechanical work done on the turbine shaft to spin it.
• Since we extract more energy out of the fluid, the fluid is left with less energy than in the isenthalpic process, and this results in lower temperature.
• The ideal isentropic process represents the maximum amount of energy that can be extracted or added to the fluid for the given pressure change.

In GUNNS, we can model this process quite well for ideal gasses, but not so well for other types of fluids. The modeling of this affect doesn’t happen automatically. Rather, the links that can model it must be configured with their “expansion scale factor” term. This discusses how to use it.

Consider the process going from E to F in the picture below. This is a typical process of simple heat transfer from (or to) the fluid at constant pressure. This could be any form of heat transfer, but the main one that we model in GUNNS most often is forced convection. (More on these later…)

Here we have hot steam that is either flowing or at rest, being cooled by some transfer of heat out of it into colder surroundings. This kind of process usually happens at constant or near-constant pressure, so we follow the constant pressure line down from the fluid’s starting to final enthalpy at F.

Enthalpy is not a measurable quantity (the vehicle doesn’t have ‘enthalpy sensors’), so it doesn’t matter if GUNNS has the correct value for enthalpy at various points, but the difference in enthalpies between points is important because this relates how much heat must be transferred to move from one temperature to another. The important parameters for the user are temperature and heat flux:

• Temperatures can be directly sensed, so the user will usually notice if the temperature is very wrong for a given heat flux to/from the fluid.
• The magnitude of heat flux matters because it affects the relative temperature between the fluid and the surroundings (i.e. vehicle equipment). The equipment generates a certain amount of waste heat, which is generally a known quantity. If we model heat flux to be too small at the expected temperatures, then the equipment will get too hot. Equipment tends to shut down or break when it overheats.

The fluid’s specific heat property, which governs this relationship between heat flux and change in temperature, is optimized for gas and liquid fluid types in the region of pressures and temperatures where they’re typically used for heat transfer applications, i.e. in ATCS and ECLSS THC systems. It tends to be less accurate near the supercritical phase. For example, heat transfer between supercritical fluid in a tank and the tank wall will be less accurate than between an ideal gas and the wall.

Now let’s consider path F to G in the above picture. This is typical of a condensing heat exchanger, where the gas is chilled to the saturation line where it starts to condense. At point F, we have 100% water vapor at 1 atmosphere, specific enthalpy of 2700 kJ/kg, and temperature of 112 °C, slightly above the saturation temperature of 100 °C (boiling point). In this process, which happens at constant pressure, we remove enough heat from the fluid to drop its specific enthalpy to 2000 kJ/kg. Things proceed normally until the fluid reaches the saturation line, and then phase change starts to occur and the process proceeds to the left towards point G. The important thing to remember is that if pressure is held constant, phase change also happens at constant temperature. This means that the energy we are removing from the gas is causing changes in the inter-molecular bonds (condensation) instead of molecular kinetic energy (temperature).

At the final enthalpy, the phase mixture is 30% liquid / 70% gas, as denoted by the dashed blue line that we’re on at point G. To condense all of the gas into water, we have to drop the specific enthalpy all the way down to 418 kJ/kg. This big jump in energy between the liquid & gas phase at the saturation curve is called the latent heat of vaporization. The phase change itself is able to absorb a lot of energy without changing the fluid temperature at all. Coolant loops can take advantage of this to dramatically improve their performance — this is central to the vapor-compression cycle that is used in most household refrigeration applications.

Note that most GUNNS links don’t model this process. Instead of beginning phase change at the saturation curve, a link that causes the specific enthalpy to drop below saturation will keep dropping the gas temperature, continuing further along the straight line path from F to H in the picture above, which is completely unrealistic. Not only do we miss the phase change, but the resulting temperature is much too cold. So be aware of this limitation.

To model the above process from F to G properly, you must use a special GUNNS link that is designed to handle it. Links that can do this are:

• GunnsFluidCondensingHxSeparator: this can only condense H2O out of a gas mixture like air.
• GunnsFluidPhaseChangeSource: you have to give it the power or heat flux into/out of the fluid.
• GunnsFluidPhaseChangeConductor: automatically converts all flow-thru between phases and doesn’t consider energy.

Boiling is the reverse of the condensation process described above.

In this example, we start at point I with liquid water at 1 atmosphere pressure, specific enthalpy of 200 kJ/kg, and temperature of 40 °C. As energy is added, the liquid heats up until it reaches the boiling point J, and then further energy goes into breaking the liquid molecular bonds and freeing up molecules into the gas phase bubbles, instead of increasing temperature. Thus the process proceeds to our final enthalpy of 1200 kJ/kg at point K.

In GUNNS however, any link that keeps adding energy to the fluid will cause it to keep heating above the boiling point while still remaining in the liquid phase. The resulting temperature at the final enthalpy at point L will be much higher than it should, and you’ll end up with 100% liquid phase in a pressure-temperature regime where 100% liquid can’t exist.

The same links that can do condensation, described above, can also do boiling properly.

Previous Page / Next Page