A WORD ABOUT VACUUM
PUMPS
For
this discussion we will use the example of an old style Welch Model
1405, although this information may be applied to any rotary vane, oil
sealed mechanical vacuum pump, whether belt drive or direct drive. The
old style 1405 was rated at 60 L/min. (2.1 CFM) with a guaranteed
ultimate vacuum of 0.1 micron (0.1 millitorr or mtorr). Many mom & pop
backyard-type wholesale neon shops as well as small retail shops used
these pumps because they were quality pumps, relatively inexpensive,
small, compact, quiet and readily available from most sign supply
distributors and vacuum equipment suppliers. However, this does not
mean they were the best choice.
First, it is important to understand that an oil sealed mechanical
vacuum pump, whether belt drive or direct drive, is not efficient at
pumping water vapor. It is also important to understand that at a
bombarding temperature of 200°C roughly 70-80% of the total gas load
from a unit is water vapor. However, the ratings of a vacuum pump,
that is the capacity in liters per minute (L/min.) or cubic feet per
minute (CFM) and the ultimate pressure (vacuum) capability, are
relative to pumping dry air. These numbers are often misunderstood as
to how they apply to real-world conditions: In this case, processing
luminous tubing.
PUMPING SPEED RATING
Mechanical vacuum pumps of this type are rated for their free air
displacement capability. This is generally stated as L/min. or CFM*,
although some pumps manufactured offshore are rated for cubic feet per
hour when sold to the U.S. market.
Free air displacement is defined as: The volume of air passed through
the pump per unit of time when the pressure at the intake and exhaust
of the pump is equal to atmospheric pressure (ATM). In other words,
how much air the pump can displace when the intake and exhaust are
wide open to ATM. In our example the free air displacement of the old
1405 is 60 L/min. as previously stated.
Free air displacement is also the pumping speed of the pump under the
above conditions; i.e., when the intake and exhaust are at ATM.
However, the pumping speed changes as the pressure at the intake of
the pump decreases. Therefore, pumping speed is defined as: The volume
of air per unit of time that the vacuum pump is able to remove from
the system.
When the pump is connected to a vacuum system and begins to pull a
vacuum the pumping speed changes. As the pressure in the system is
reduced and the vacuum gets better the ability of the pump to displace
air (or other gas load) progressively gets worse. For example, the
same 1405 pump that has a free air displacement of 60 L/min. at ATM
only has a pumping speed of about 35 L/min. at a vacuum level of 10
microns. A vacuum of 10 microns is not an adequate vacuum for neon
work and the pumping speed of the pump has already fallen off to
almost half of its free air displacement. At a vacuum of 5 microns,
which is considered by most experts in the field to be the bare
minimum for neon work, the pumping speed is roughly 20 L/min. As the
pressure is reduced further the pumping speed continues to get worse
until the pump reaches its blank-off pressure, or ultimate vacuum
capability. (In real-world situations the actual blank-off pressure is
much higher than the advertised “guaranteed ultimate vacuum”). At the
blank-off pressure the pumping speed is only a few liters per minute.
Even though it is roughly ⅓ of the free air displacement, the pumping
speed of 20 L/min. at the 5 micron vacuum level may sound adequate.
However, this equates to just ⅓ liter per second. At a bombarding
temperature of 200-300°C an average size 15mm unit can release several
liters per second of gas load. Therefore, the ⅓ liter per second
pumping speed is not adequate to remove the gas load fast enough,
before the unit cools to the critical re-absorption temperature of
175°C. Further, the pumping speed for water vapor is much worse than
it is for air. Although it would be very difficult to calculate the
percentage decrease in pumping speed for water vapor for a given pump
due to the variables involved, such as tubing diameter, length, coated
or uncoated, how the glass has been stored, etc., it is certain that
the performance of the pump falls off considerably when pumping water
vapor as opposed to dry air. The majority of gas load in this
situation is water vapor, so the ability of the pump to remove the gas
load is greatly reduced.
Larger pumps can be used in an attempt to compensate for the lack of
performance of smaller pumps in the low micron range. To some extent
this may be helpful as a larger capacity pump will rough pump the unit
quicker, down to a certain point. But regardless of how big a
mechanical vacuum pump is it still suffers from the same inability to
pump water vapor in the low micron range, though it will do it
marginally better than a smaller pump.
* To convert CFM to
liters per minute multiply the CFM by 28.32. To convert liters per
minute to CFM multiply the liters per minute by 0.0353.
ULTIMATE PRESSURE / VACUUM RATING
This heading may sound confusing to some. The correct terminology is
“ultimate pressure”. However, most think of anything less than
atmospheric pressure as being a vacuum, which is also correct. The
terms seem to contradict each other. But for our purposes, the terms
ultimate pressure and ultimate vacuum are the same thing.
The reasoning behind the term ultimate pressure: If there were a
perfect vacuum, which does not exist, it would be absent of all gas
molecules. This is considered absolute zero, or zero pressure. If even
one molecule of gas existed it would have the ability to exert a
pressure inside a sealed chamber – no matter how small of a pressure.
Therefore, the chamber would have in it a pressure above zero.
The reasoning behind the term ultimate vacuum: A device that has the
ability to remove some of the air or other gas from a chamber is
considered a vacuum pump, mechanical or otherwise. Once the pump
begins to remove the gas from the chamber and the pressure inside that
chamber falls below ATM a vacuum is created. The pump will continue to
remove gas molecules from the chamber until the pumps ability to
remove gas and the rate of outgassing and/or continued gas load caused
by leaks reaches equilibrium. The vacuum inside the chamber does not
improve regardless of how long the pump continues to pump. This level
of vacuum is the best, or the ultimate, that the pump can achieve.
Therefore the term ultimate vacuum applies.
Ultimate pressure is defined by one vacuum pump manufacturer as: “The
lowest attainable pressure in a vacuum system; The lowest attainable
pressure by a vacuum pump; Ultimate pressure is limited by the pumping
speed of the vacuum pump and the vapor pressure of the vacuum pump
sealing fluid, among other factors.”
The advertised ultimate pressure (or vacuum) for 2 stage rotary vane,
oil sealed high vacuum mechanical pumps is typically between 0.1 and
0.5 microns. However, this number is extremely misleading in
real-world situations. To understand why one must know how the pump
manufacturers test their pumps.
First, pumps are tested in a laboratory setting and not in a
manufacturing plant somewhere. The lab will have closely controlled
ambient air conditions. A temperature of 68-70°F with relative
humidity as close to zero as possible is not uncommon. Prior to
testing the test pump is meticulously cleaned, much more so than an
“off the assembly line” pump is. The highest quality high vacuum pump
oil available is used. The pump is run with this oil for extended
periods of time and flushed as many as 5 or 6 times to remove any
remaining volatiles and contaminants from the pump. For the final
“vacuum reading” test the pump is fitted with an LN2 trap (liquid
nitrogen trap) directly on the intake of the pump, which prevents any
oil vapor from backstreaming to the vacuum gauge. The pump will run
for at least 48 hours before the vacuum reading is accepted. The
vacuum gauge used for the test will be the most favorable for the
desired indicated vacuum. For example, when a McLeod gauge takes a
pressure measurement it compresses the gas in the instrument in order
to take the reading. Therefore, this type of gauge cannot measure
condensable vapors, as do electronic gauges. If condensable vapors
such as water vapor or oil vapor are present in the test chamber (in
this case the vacuum pump intake) the indicated level of vacuum is
different than the actual vacuum when using a McLeod gauge. Therefore,
this type of instrument may be used if it indicates a better vacuum
than the actual vacuum under the test conditions.
IN CONCLUSION:
Immediately following bombarding the unit has to be evacuated to the
lowest possible pressure before the glass cools to 175°C as this is
the temperature that the released gases and vapors will begin to
condense and redeposit on the walls of the tubing. Referring to the
preceding information on mechanical pump performance we know that a
mechanical pump cannot accomplish this by itself. Therefore, in order
to achieve the required vacuum in the time frame necessary a secondary
vacuum pump must be used in conjunction with the mechanical vacuum
pump. It is our strong belief from our many years in the neon
equipment business that the overall best choice of secondary vacuum
pump for our purposes is the single stage glass oil
Diffusion Pump.
They have no moving parts, are isolated from electrical ground, are
easy to maintain in-house and provide fast pumping speeds in the low
micron range. A reasonably sized mechanical vacuum pump will provide
sufficient fore-pressure for the diffusion pump to function properly.
The type and size of mechanical pump necessary is a two stage oil
sealed rotary vane pump with a free air displacement of at least 100
L/min., but preferably larger.
|
