The Fundamental Myth

[Gary]

People throughout our industry are taught that superheat cannot exist in the presence of liquid, and that subcooling cannot exist in the presence of vapor.

Imagine a container of refrigerant sitting on a block of ice, with a heating pad on top. We have superheated vapor at the top, subcooled liquid at the bottom, and saturation only where there is direct contact between liquid and vapor.

Next, imagine subcooled liquid droplets in a suction line, saturated at their surface and surrounded by superheated vapor.

Further, imagine superheated vapor bubbles in a liquid line, saturated at their surface and surrounded by subcooled liquid.

That's my theory. What do you think?

Gary

[Gary]

Hey Marc. Good to see you again too, friend. It has indeed been a long time.

I don't disagree with what you are saying. The vapor inside those liquid line bubbles could vary from superheated to saturated depending upon the variables.

My main point is that the notion of "in the presence of" should be relegated to the status of ancient mythology. It is a fundamental misconception which confounds everyone's understanding of subcooling and superheat.

We measure subcooling and wonder how there can be bubbles in the sightglass. We measure superheat and wonder how the compressor can be flooded.

Why? Because "in the presence of" just ain't true.

Gary

[Gary]

Been keeping very busy selling my trouble-shooting books.

<url>http://www.gatecom.com/~tmethod/</url>

Much as I try, Marc, my mind zeroes in on the more practical aspects as they relate to trouble-shooting, and I fail to get excited about the finer technical points. Can't help it. I'm a wrench spinner to the core, and an old geezer to boot.

I'll leave the technical explanations to you and the Prof. (Unless/until I disagree, of course)...lol

Cheers.

Gary

[Prof Sporlan]

People throughout our industry are taught that superheat cannot exist in the presence of liquid, and that
subcooling cannot exist in the presence of vapor.

Generally a good dictum, though there are instances where it is technically not correct, e.g., the presence of droplets within superheated vapor flowing in a suction line. <br>

Imagine a container of refrigerant sitting on a block of ice, with a heating pad on top. We have superheated vapor at the top, subcooled liquid at the bottom, and saturation only where there is direct contact between liquid and vapor.

By inducing a temperature gradient across a column of refrigerant, you could have superheated and subcooled refrigerant in a vessel.
But the above dictum assumes all of the refrigerant is at the same temperature. Interestingly, having a column of liquid refrigerant at the same temperature, by definition, allows one to have both saturated and subcooled liquid. At the liquid vapor interface at the surface of the column, you would be saturated. At the bottom of the column, the refrigerant would be subcooled by the equivalent pressure caused by the mass of the refrigerant. For example, if we have 4 feet of R-22 liquid height at 70°F:<br>
<font face="Symbol">r</font>_70°F = 75.3 lb_m/ft³
static pressure = 75.3 lb_m/ft³ / 144 in²/ft² * 4 ft = 2.1 psi or 1.0°F subcooling<br>

Next, imagine subcooled liquid droplets in a suction line, saturated at their surface and surrounded by superheated vapor. Further, imagine superheated vapor bubbles in a liquid line, saturated at their surface and surrounded by subcooled liquid.

The Prof has witnessed liquid droplets in the suction line surrounded by superheated vapor. This can be accomplished by placing a sight glass at the evaporator coil outlet, and accurately measuring superheat and observing refrigerant flow. As he recalls, he saw small droplets up to about 5°F superheat on a residential R-22 a/c system. How can this be? A worthy discussion topic in itself...

[Gary]

Generally a very misleading dictum which is technically incorrect as soon as you start the system, and incorrect where it matters most to the service technician, "e.g., the presence of droplets within superheated vapor flowing in a suction line."

Other than that, it's just fine.

Gary


[Edited by Gary on 17-04-2001 at 03:04 AM]

[Gary]

I have observed liquid droplets in the suction line in a variety of systems (dare I say this occurs in all systems?) up to at least 5°F. That's why compressor manufacturers want at least 15°F superheat entering the compressor.

Marc, in your example, had you replaced the compressor to match the new coil capacity, I suspect the coil vapor leaving velocity would have been even greater, and that the new TEV was just simply too much for the undersized compressor. Possibly an in-between TEV and/or a CPR to safeguard the compressor would have been appropriate.

Just like old times, isn't it?

Gary

[Prof Sporlan]

<i>So how can refrigerant drops exist in the suction line when they are known to be surrounded by superheated vapor?</i>

General refrigeration theory states that for a single constituent refrigerant, e.g., R-22, liquid begins to change to vapor when the temperature of the liquid is raised to its saturation temperature. Above saturation temperature, only vapor remains.

The problem with this theory is it conficts with the above observation. What is a possible explanation?

Vapor pressure at saturation refers to the partial pressure of vapor in equilibrium with the liquid whose surface is reasonably flat, which is the case with a refrigerant cylinder. In a drop of liquid or a bubble of vapor in a liquid, however, the surface of the liquid is not flat, but curved. For drops of reasonable size, this does not make much difference.

But first, why do drops form anyway? A molecule within bulk liquid experiences attractions to neighboring molecules in all directions. Since these attractions average out to zero, there is no net force on the molecule, and it is energetically as happy in one location in the liquid as another. If the molecule happens to be located at the surface of the liquid, however, then there is no attraction operating in a 180° solid angle above the surface. As a consequence, the molecule at the surface tends to be drawn into the bulk of the liquid. Intermolecular forces will tend to minimize surface area since matter tends to move to its lowest energy state. And the geometric shape that has the smallest ratio of surface area to volume is a sphere, so very small quantites of liquid form spherical drops. As the drops get bigger, their mass deforms them into into the typical tear shaped drop.

Interestingly, since the outermost molecules of the liquid are bound to the droplet less tightly, a very small drop will have a larger vapor pressure than the partial pressure of vapor in the gas phase. In other words, the fact that refrigerant is in small droplets doesn't in itself make the refrigerant "harder" to evaporate. The opposite is true, in fact. So the answer lies in the fact that the drop has velocity.

[Gary]

Would you care to take on the other (similar) proposition; That of vapor bubbles existing in the liquid line when they are known to be surrounded by subcooled liquid?

My testing has shown them to be even more persistant. While liquid droplets in the suction line disappear between 5-10°F superheat, vapor bubbles in the liquid line disappear between 10-15°F subcooling.

Gary

[Gary]

Here are some further thoughts on this:

In a stable closed loop system, assuming no parallel paths, the flow rate past any point in the system must necessarily be identical to the flow rate past any other point in the system.

The velocity through the piping would then depend upon it's diameter. The liquid line being smaller piping than the suction line, would have a higher velocity. So perhaps this would account for the bubbles lasting longer than the droplets.

Gary

[Prof Sporlan]

<i>The velocity through the piping would then depend upon it's diameter. The liquid line being smaller piping than the suction line, would have a higher velocity.</i>

Conservation of mass does dictate flow rate must be the same from one point to the next when refrigerant flows along one path. But velocity is a function of both the cross-sectional area of the pipe <b>and</b> refrigerant density. Since liquid refrigerant is much more dense than its vapor, you will rarely find a situation where refrigerant velocity in the liquid line is greater than the suction line. In general, suction lines are generally sized to maintain a minimum 700 ft/min velocity for horizontal sections, and about 1500 ft/min for vertical sections. The purpose for maintaining these velocities is to assure oil return.

With liquid lines, 400 ft/min is perhaps the highest velocity one might find, and 200 ft/min or less is more typical of a refrigeration unit.

But the Prof is stewing over the problem of the presence of vapor in subcooled liquid....

[Gary]

Perhaps some sort of insulating effect/thermal inertia?

The one thing that seems clear is that static gas laws are inadequate to properly describe dynamic processes.

Gary

[Brian_UK]

One thought occurs to me which is:-

Does that fact that lubricating oil is travelling around with the refrigerant vapour/liquid have any bearing on the formation of droplets or their shape?

Brian

[Prof Sporlan]

<i>Perhaps some sort of insulating effect/thermal inertia?</i>

We could consider the refrigerant's specific heat c_p as the source of its thermal inertia. For R-22 liquid at 20°F, c_p would be 0.276 Btu/lb_m-°F. But the mass of a drop is very small, and it would seem micro-Btus would only be necessary to cause a drop to vanish.

<i>The one thing that seems clear is that static gas laws are inadequate to properly describe dynamic processes.</i>

But the advantage of static gas laws is they are easier to teach and understand....

[Gary]

An interesting question, Brian. Perhaps an oil film could provide an insulating effect.

Prof, it's easier to teach and understand that the world is flat, but technically incorrect.

Gary

[Edited by Gary on 19-04-2001 at 02:53 PM]

[Dan]

Professor, you are bringing up fascinating points regarding droplets. Question, though. What about surface tension? Doesn't that come into play with its own peculiarities?

Dan

[Jorgen Bargsteen Moller]

I have with interest read your thoughts about this subject.

It was an issue here in Europe in the late 70'ties and the early 80'ties because of the heat pump fewer.

Many would be manufactures found it difficult to achive the COP expected from the chosen compressores, TEV's ect. We at Danfoss therefore had to explain why COP was less than expected.

We ahd the knowledge about the work done by Doctor Engineer J. Reichelt from the university in Stuttgart, Germany analysing this phenomenon. Danfoss suported the investigation so we were allowed to publish part of the work in the Danfoss Journal including some very interesting photos taken of the droplets in the pipe and the amount of superheat required before the unevaporated part of the refrigerant is acceptable (test refrigerant R 12).

Unfortunately I have a Danish version only, number 3-79 meaning the English version is either 4-79 or 1-80 (number and year). I am searching for an English version for later use.

[Gary]

Welcome to the discussion and the group.

Unfortunately I have a Danish version only, number 3-79 meaning the English version is either 4-79 or 1-80 (number and year). I am searching for an English version for later use.

This would be very helpful. I have been waiting for 16 years for something official to show the naysayers.

[Prof Sporlan]

Perhaps adiabatic influences can contribute too

Hmmmm... interesting thought. If refrigerant flow thru the suction line were adiabatic, and the refrigerant vapor was marginally superheated, then an unloading compressor could, in theory, recondense vapor.

Of course, refrigerant flow would not be adiabatic due to friction and some heat transfer thru the suction line. But it may be close enough to cause recondensing to possibly occur.

But the Prof is certain liquid droplets can exist in marginally superheated vapor without the need to increase presssure.

[Prof Sporlan]

For those of you who wish to peer into this arcane arena called thermodynamics, the term "adiabatic" simply refers to a process involving a working fluid (refrigerant in our case) where heat does not cross a predetermined system boundary.

To complicate matters a bit (there is always complications with thermodynamics ), one might think a constant temperature (isothermal) process is adiabatic. This is not the case! An example of an isothermal process is ice melting, or allowing a refrigerant to boil in an evaporator having no pressure drop. Clearly, these processes are not adiabatic since heat is being absorbed from the environment.

Adiabatic processes are a general category for the following processes: (1) isentropic; (2) isenthalpic (throttling); and (3) polytropic.

An isentropic process is a process in which we have no change in entropy. An example of this process is refrigerant vapor being compressed by an ideal compressor. Of course, real world compressors do not perform isentropic compression, but often it can be assumed they do to estimate required heat of rejection.

An isenthalpic process is a process in which we have no change in enthalpy. The common example of this process is refrigerant flow thru the metering device. That right, refrigerant flow thru that ubiquitous thermostatic expansion valve is assumed to be adiabatic!

A polytropic process is a general type process in which the state of the working fluid can be predicted by the following equation:

PVⁿ = constant

where:
P = pressure
V = volume
n = an index which depends on the process type

Any friction due to flow (as with pressure drop in the suction line) would seem to move us away from an adiabatic process, even if the suction line were heavily insulated. Refrigerant flow thru a suction line which is perfectly insulated (no heat transfer), and where no pressure drop exists would be adiabatic, by definition.

[mark_harris_nz]

Could this be the reason for TX valve bulb placement on a pipe (ie 4 oclock on a large pipe).

[superheat]

I leave for a few weeks and the prof starts up on thermodynamics. Look at the device used to track subatomic particles (forget the name). It uses a gas below its condensing point. The particles leave a trail by condensing the gas into droplets. I do not know how well this supports Gary's theory, but I bet he studies up on this and generates concrete proof based on his studying.

Just mention my name in the contributors section when you publish Gary.

[Gary]

I'M thinking maybe a little more than a few weeks.

I try to avoid thermodynamics except as it relates directly to the service techs ability to do his job. Here I presented carryover as if it were a theory, but it is FACT. I will let others explain it.

I published in 1986, but rest assured that you were mentioned, along with your alter ego "subcooling", ad infinitum throughout my books.

[Jorgen Bargsteen Moller]

Long time back I promised to find an article in a Danfoss Journal from late 1979.

I was about droplets in the suction gas as a function of the super heat.

I have article with black and white pictures.

All interested please e-mail me at: jbm@danfoss.com

Besr regards

Jorgen Bargsteen Moller
Danfoss Nordborg, Denmark