Paul Lamar wrote:
Come to think of it I did not see any -4 bleeds on Power Sports engine
from the front of the block (PSRU end) to the pressure can.
Perhaps when they changed the rad to a single pass from a double
pass they inadvertently got all the air out of the system and it
started
cooling better. Things like that can really fool you.
Come to thing of it if you don't have a bleed like this the rotary
engine will
build its own air pocket in this part of the engine block while the
airplane is in a climb . That is because the water inlet and
outlet to the block are located no higher than the rest of
the block. When it is tilted up in a climb it is generating
an air pocket in the front of the block due to normal nucleate
boiling. Not to mention an out and out boiling steam pocket.
Russell Kent wrote:
[ Nitpick mode = ON ]
Paul,
True nucleate boiling won't cause an "air pocket". Actually, nothing
but
entrailed air can cause an air pocket. It is possible to have full-tilt
boiling
cause a steam pocket, however. Of course, if you have full-tilt boiling
(not
nucleate), then you're seconds away from much bigger issues...
But you are correct in stating that the highest point in the block needs
to have a
bleed line.
Russell Kent
I think Heinz Heisler disagrees with you. He calls it "vapor" in
them tiny bubbles and they go to the air space at
the top of the system.
Paul Lamar
Monty Roberts wrote:
Russell is right Paul. There is a steam pocket, not an air pocket. This is
not the same. Nothing in the Heisler text you posted says anything about
air. I did my MS in mechanical engineering on this subject and unfortunately
for me (and now all of you) know more about the boiling process and heat
transfer than I ever wanted, or anyone should have to know about. The main
reason that the distinction is important is that air is a different
substance than the coolant. It remains a gas (for our purposes) and (if you
do not provide a bleed) will not go away no matter what. The steam pocket is
formed over a hot spot from the coolant itself, or in a low pressure area of
the coolant system (vortex, pump inlet, sudden expansion etc). Even if under
normal operations (cruise for instance) you have nice liquid to metal
contact, during high heat flux elevated temp situation (climb on a hot day)
you may form steam pockets over the hot spots (if the steam bubbles can't
escape). If you are using a two phase coolant (glycol, water) the individual
substances can separate, depending on which one has the lower vapor pressure
(boil at the lower temp for a certain pressure). This is undesirable. The
reason we use two substances (other than anti-corrosion additives etc) is to
increase the boiling point of the coolant blend. If you get it hot enough to
cause a steam pocket or boiling you have already lost the battle.
There are many types of boiling, the ones most relevant are as follows:
Sub cooled boiling- The bulk fluid has not yet reached the boiling point for
the pressure it is at. Small bubbles form on nucleation sites on the metal,
expand and then collapse. This happens because the fluid next to the metal
reaches the boiling point, but the vapor bubble rapidly gives up its heat to
the bulk fluid condenses, and collapses.
Bulk boiling-This occurs when the bulk fluid temperature is at the boiling
point (temp) for the pressure that it is at. Bubbles will form throughout
the fluid and expand as they rise. Being less dense than the fluid the are
buoyant and move opposite the acceleration on the system (upwards for a pot
of water on the stove, less predictable in an airplane). When they reach the
surface they burst and release vapor.
Film boiling- the surface is REALLY hot! There is a film of vapor formed
over the surface of the metal which acts as an insulator. VERY BAD!! The
metal gets too hot. This can happen at a sub cooled state or a state where
the bulk liquid is at the boiling point.
In an engine there is not a vapor/liquid surface per se and the steam
bubbles may become trapped in a nook or cranny in the engine block. This
will produce a steam pocket, which in turn produces a hot spot because the
steam does not absorb heat as well as the liquid coolant. Incidentally, the
heat transfer coefficient of a material which is in a phase change (boiling)
is infinite. It will absorb an infinite amount of heat flux until all of the
liquid is changed to vapor, or vapor to liquid, or solid to liquid etc. And
it works both ways. That is why it burns like a @#^%$*&@ when hot steam from
your microwave dinner condenses on the soft pink flesh of your fingers {=0.
Cooling systems which use this principal are extremely compact and
efficient, but they must be designed to work this way. Nuclear reactors,
boilers of all types, and most types of refrigeration systems come to mind,
but not liquid cooled internal combustion engines. They need liquid coolant
under all circumstances. A little bit of sub cooled boiling over hot spots
is acceptable under extreme conditions, but it must never become bulk, or
film boiling. There should never be boiling in a properly designed cooling
system for an engine for this reason. If you have steam pockets, you do not
have the right coolant, or your system pressure is too low.
The end result of either a steam pocket or and air pocket is the same, a hot
spot where heat transfer is diminished. This results in thermal stresses in
the material being cooled due to the large temperature gradient between the
hot spot and the cold areas around it. The difference in thermal expansion
eventually will warp the part, or cause the chrome plating to flake off or
other nasty things like cracking. This is a condition we should not allow to
occur. A bleed will do nothing if you have film boiling, you are going to
have a melt down. If you have a steam pocket, due to bulk boiling, your
system pressure is too low, or your coolant temp is too high, take your
pick. Steam pockets will not form from sub-cooled boiling because the steam
cannot exist in the sub cooled liquid for long periods of time. Sub cooled
boiling is the only type of boiling that is acceptable in our case, it
actually helps remove heat from the hot spots at a rapid rate. So install
bleeds to remove air from your system, but don't try to solve boiling
problems with this method.
Incidentally ( and slightly off subject) it requires several things for
boiling to occur:
1) The substance is at saturation temperature and pressure (the boiling
point)
2) Sufficient nucleation sites exist to cause bubble formation (rough
surface, particles in the substance; things which are not a problem in an
engine block)
If you take away number 2 you can actually heat the liquid above the boiling
point. This is called a metastable state. All it takes is a trigger like
vibration or introduction of nucleation sites and the fluid explodes into
bulk boiling (emphasis on EXPLODE). All of the stored energy above the
boiling point will be released when this happens. Some of you may have
experienced this when heating water in a microwave in a container which just
came out of the dish washer. When you put your tea bag in BOOM. Or sometimes
it just explodes in the microwave due to vibration, and you get your
microwave cleaned for free ;-)
Hopefully I have not included too many obscure and technical descriptions.
If I have, I hope I explained them well enough, If not say so and I will try
to do better next time.
Monty
PS: I'm not trying to be the little boy in Herb's red dog leader post. I
just thought that this might help some of you guys struggling with cooling
issues.
OK it is vapor or steam pocket at the highest point in the system not
an air pocket. I stand corrected.
Imagine what would happen if it were plugs up.
Here is some work Continental did.
Paul Lamar
The AirCraft Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://home.earthlink.net/~rotaryeng/ http://www.linux.org
In the mid-1980s, Teledyne Continental Motors initiated design
of a 9.O-L, six-cylinder, liquid-cooled aircraft engine, with the
goals of attaining higher specific output, lower fuel consumption1
longer periods be- tween engine overhauls, and no weight penalty
when compared with traditional air-cooled aircraft piston engines.
Fol- lowing FAA certification in 1988, production of the engine,
shown in Figure 1, started in 1989.
To achieve an engine installation weight comparable with
existing air- cooled aircraft piston engine installa- tions, cooling
system features include in- creased coolant operating temperatures
and extensive use of light alloys (in areas such as the cylinder
head, coolant-pump housing and impeller, radiator, coolant
circulation piping, and coolant tank). The engine cylinder was
designed for 121 C maximum engine-coolant inlet temperature. The
increased operating temperature reduces the heat rejection to the
coolant (see Pigure 2) and permits a smaller radiator. The smaller
radiator can also result in reduced aircraft drag. For operating at
these elevated tempera- tures, higher system pressures were
specified to suppress boiling and cavita- tion.
The engine is designed with separate cylinder
construction~ach cylinder as- sembly consists of a steel cylinder
barrel and cast-aluminum cylinder head. The barrel and head are
joined via a threaded joint in each part. The combustion cham- ber
is formed in the cylinder head. A coolant jacket surrounding the
combus- tion chamber is cast as an integral part of the cylinder
head. An ethylene glycoll water mixture circulated through the
coolant jacket plus oil cooling provides the he~ transfer from the
combustion section.
The inlet coolant is directed toward the valve bridge, spark
plug, and exhaust port areas. The coolant then flows t~ ward the
lower portion of the cylinder head and exits the outlet port. Oil
spray from a crankcase-mounted nozzle aimed at the piston underside
provides addi- tional heat transfer.
The pump uses a cast-aluminum, closed-impeller design,
operating at speeds up to 8100 rpm. A pressure rise of 207 kPa at a
flow rate of 114 L/min. oc- curs at rated engine speed.
The coolant distribution system works in parallel. From the
coolant- pump exit, the coolant is divided into two branches and
routed to each bank of three cylinders. It is equally divided among
the cylinders by an orifice at each cylinder's exit. From each
cylinder bank, the coolant is collected and routed to a 4.9-L
coolant tank.
An existing commercial, ground-ve- hicle ethylene glycol
formulation was initially selected as the coolant. A 60/40%
solution of ethylene glycol/distilled water was specified to ensure
adequate freeze protection. Typical inhibitor pack- ages for these
coolants include phos- phates, borates, hydroxides, nitrates, sili-
cates, and triazole. Initial coolant change intervals at 250 h of
operation or one year were selected, based on coolant supplier
recommendations.
To operate at the elevated coolant temperatures, higher
system pressures are employed for suppression of boiling and
cavitation. During the engine dyna- mometer development phase, a
cylinder- head temperature profile was obtained to evaluate
temperature gradients by em- bedding thermocouples throughout the
cylinder head. The maximum metal tem- perature measured at the
combustion chamber surface is 210 C between the valves. The typical
coolant temperature rise across each cylinder is 8-11 C.
No unusual cooling system problems were encountered, except
for light coat- ings of a white substance that were ob- served
intermittently in the engine cylin- ders and other components. These
depos- its were analyzed using energy dispersant spectroscopy (EDS)
and determined to contain significant amounts of phospho- rus and
silicon.
Engineers converted a twin-engine air- craft which was
originally powered by air- cooled engines (see Figure 3) to a water-
cooled system. The cooling system is con- figured to fit within the
existing aircraft nacelle, as shown in Figure 4. New cowl- ings are
used to manage airflow through the radiator and engine compartment.
The coolant is routed through the system from engine exit to coolant
tank, coolant tank to radiator inlet or thermostat~ontrolled by-
pass, and radiator exit to coolant pump inlet.
The coolant tank provides additional coolant storage,
coolant/air separation ca- pability, system pressure control, coolant
~vel monitoring, and supply of bleed cool- ant to the pump inlet.
The coolant radiator is an aluminum plat~and-fin~ type unit with
headers welded on each end. The coolant passage cross section is
approxi- mately 1.5 x 2 mm, with a length of 585 mm. Except for
flexible lines at the engin~ to-airframe connection points, all
coolant is routed through components of alumi- num alloy material
with Oring seals at each joint.
The system's initial service experience was satisfactory.
However, after 200-250 h of aircraft operation, excessive coolant
temperatures (greater than 120 C) were reported. Inspection of
several cooling systems revealed two distinct types of de- posits.
Researchers observed a very light coating of a white substance on
most sur- faces throughout the system, as well as a heavy buildup of
a gel-type, off-white sub- stance at the exit side of the radiator
(see Figure 5). The light coating was observed intermittently during
development, but did not present a problem. Several clean- ing
methods were tried without disassem- bling the radiator, but none
was com pletely successful. Many of the radiators required
replacement during this period.
Researchers collected deposits from several radiators and
subjected them to an elemental analysis using EDS tech- niques.
Major elements identified in- cluded phosphorus, silicon, aluminum,
and sodium. Inspection of the cooling system components did not
provide an obvious source of aluminum material loss.
Research into other liquid-cooled en- gines manufactured
with aluminum heat-rejecting surfaces revealed similar deposits.
Previous testing of these en- gines seemed to indicate that some
ethyl- ene glycol formulations may not provide adequate corrosion
protection, particu- larly at temperatures above 107 C.
Possible solutions included operat- ing-temperature
reductions, coolant-sys- tem design changes, and evaluation of other
coolants. Reduction of operating temperatures or coolant-system
design changes were considered undesirable based on the effort
required for system development and certification. Other coolant
candidates considered were pro- pylene glycol and ethylene glycol
with a revised inhibitor package. Although propylene glycol is
readily available, ex- perience with inhibitors for corrosion
protection has not been well established. Based on a coolant
supplier recommen- dation, an evaluation of a newly intr~ duced
ethylene glycol with a revised in- hibitor formulation was selected
as the favored approach.
Earlier analysis of the deposits in the radiator suggested
that some of the com- pounds in the initial coolant formulation were
contributing to the excessive cool- ant-system deposits. The revised
coolant included an inhibitor package relying on low silicate
concentrations (250 ppm) plus specific carboxylic acid combina-
tions for aluminum protection. The for- mulation is phosphate-,
nitrate-, and amine-free.
Initial evaluation of the new coolant was conducted on an
engine dynamom- eter. No undesirable engine performance or coolant
system deterioration was identified.
While the new coolant was evaluated, the previously plugged
radiators were examined to determine if they could be restored
without removing the header tanks. Engineers fabricated a test rig to
circulate hot solutions of several com- mercially available
coolant-system cleaning compounds through the radia- tors. No
successful cleaning method was identified for radiators with
significant plugging. Cleaning of several aircraft coolant systems
was also attempted, with similar results.
The coolant was introduced to the air- craft fleet, using
the same procedures for mixing, installation, and coolant-system
operation as for the original coolant. Many of the radiators
required replace- ment when the new coolant was intro- duced, since
the coolant system could not be adequately deaned.
Researchers initiated a coolant monitoring program
lo?'several of the aircraft. The program consisted of withdrawing
coolant samples at regular intervals, and conducting analyses (see
Figure 6) and inspections of the coolant system.
The analysis revealed a significant decrease in silicate
levels between the imtial fill and the first sample interval, which
was also reported in previous studies. Sufficient aluminum protection
is indicated based on only trace amounts of aluminum with stable
alkalinity, pH, and carboxylic acid concentration. The quantity of
deposits in the cooling sys- tem was significantly reduced.
Based on the favorable test results, researchers plan to
continue monitoring the coolant performance with a sched- uled
sampling program. By also evaluat- ing extension of the coolant
change inter- val, and examining the performance of silicate-free
ethylene glycol formulations containing carboxylic acid combinations,
they hope to maximize the performance of cooling systems for
aircraft liquid- cooled piston engines.
Information for this article was provided by J. G. Wheelock,
Teledyne Continental Motors Aircraft Products, and J. M. Burns,
Texaco Chemical Company.
Reader Feedback
Please circle all the numbers on the enclosed
Reader Sen'ice Card which pertain to this
article.
Circle 181 This article is directlv related to
mv work
Circle 182 Technicallv, a verv useful article
Circle 183 Read for information only
Circle 184 Would like more articles in this
area
Circle 185 ~d not find articlQ usef~