Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise

Many readers have written in asking for information on the methods of making the vacuum inside of the radio tubes considered the heart of radio transmitting and receiving apparatus. The author, in this article, tells how it is done and explains just what is meant by the general term "vacuum"

Putting “Empty Spaces” in Vacuum Tubes
by Irving J. Saxl, Ph.D. Radio News October 1933

The conquest of "empty space," otherwise called a "vacuum," is perhaps the characteristic that defines best the trend of the present physical sciences and their most interesting part--the radio tube and the electronic sciences.

In the last sublimation, Einstein's relativistic universe is a cosmos, defined through its co-ordinates of space and time. As a practical working hypothesis, we find that wherever we try to go down to the fundamental principles of physical phenomena, we have to deal, according to our present theories, with a Few "points" in space that act as though matter would be concentrated in them. Such points are many times their own diameters apart from each other, as postulated by the wave theory of matter. Still more primitive, mechanistically speaking, we have to deal with molecules which are separated from each other by a space several times the diameter of the molecules.

The immediate importance of the physics of space is particularly clear in the latest wonder tool of the scientific magician: in the radio and electronic sciences. The very nature of radio broadcasting necessitates a space capable of transmitting electromagnetic oscillations.

On the other hand, it is the radio tube--a vacuum, device--which makes possible the amplification of impulses which, as such, could not be directly perceived by human beings.

What is a vacuum? Vacuum--absolute emptiness--has as yet never been produced by man. Even with our most refined evacuating devices there still remains a vast number of molecules in each cubic inch of air. On the night of April 9, when KPO, the new 50-kilowatt broadcasting station of NBC at Belmont, was christened with a "bottle of nothing," this vacuum still contained 25 sextillion molecules of air. Though considered 99.999999 percent complete, there still remained in this pearshaped lamp bulb of approximately 5 inches in diameter 250 trillion molecules of air. If these were converted into drops of water, sufficient water would be produced to raise the level of the Great Salt Lake about one-sixteenth of an inch.

This bulb, evacuated by the best present means by Dr. Irving Langmuir, who was recently awarded the Nobel prize in chemistry and who is a specialist in high-vacuum research, still contained 250 trillion molecules of gas. This will give an idea of how far away we still are from a perfect vacuum.

However, it is an assured fact that this incomplete vacuum we are able to produce is, nevertheless, quite sufficient for exhausting radio receiving and transmitter tubes, X-ray tubes and other forms of vacuum equipment and for bringing about the miracles of modern electronic techniques which are so close to the borderline of the unbelievable.

How is a vacuum produced? While we have just seen that a perfect vacuum has not as yet been attained, we must make a distinction, even in this limited realm of vacua, that is within the reach of our instruments, between the regular vacuum, the high vacuum and the extreme vacuum.

In considering vacua, the best comparison is with the air pressure at sea level. This pressure averages about 760 millimeters of mercury. If we wish to go down somewhat, for producing pressure that is sufficient for ordinary chemical work, water jet pumps of the aspirator type are sufficient for work to about 14 millimeters pressure. However, as these pumps do not approach the region of vacuum necessary for electronic work, no particular details will be given about them here. Forms of reciprocating and other old constructions of vacuum pumps will also not be discussed here, as they, too, do not enter present-day production methods.

If we wish to go to pressures of a small fraction of one miliimeter, which is the type most frequently used in vacuum work, relating pumps of the Caede type are used in modern vacuum techniques.

Figure 1. The Hypervac electrically driven pump more or less well known to people in the art.

Figure 1 shows a rotating pump of this type, driven by a small electric motor. From the cross-section at the right side of Figure 2, the operation of this pump becomes clear. Within the outer cylinder (a) moves a smaller cylinder (b) which is attached eccentrically and touches, with its widest elongation, the wall of the outer cylinder chamber. As shown by the arrow, this inner cylinder rotates clockwise. A valve (c) is pressed down upon the inner cylinder by an arrangement of springs in such a way that it is moved up and down if the inner cylinder rotates.

Figure 2, left, shows the cross-section vieww of the Hypervac pump. Figure 3, above, shows diagrammatically another type of high vacuum pump, while figure 4, right, is a patent diagram of a multi-stage high vacuum pump.

It will be readily understood from this picture that the volume of air which has entered through the big pipe (p) at the upper right side of the diagram will be moved around clockwise before the rotating inner cylinder, and finally is pushed out through another valve (v) immediately behind valve c. The valve, of course, must close tightly, and similarly the connection between the eccentric inner cylinder and the wall must be vacuum tight.

Pumps of this type, if operated properly, give vacua which are probably better than 1 micron (1 micron is one-thousandth of 1 millimeter mercury pressure). This is within the dimension which is necessary for the production of regular radio ampiifier tubes. The usual procedure is that they are exhausted in several states with rotating oil pumps, similar in their general physical construction to the one discussed above. After the last amounts of gases have been properly driven out, a small amount of material consisting mostly of magnesium, calcium and some alkaline earths is placed within and evaporated in the bulb. This material, known as a getter, covering the walls of the bulb with a mirror-like deposit, absorbs the remaining gas in the tube and brings the vacuum within the bulb down to a point where electrons can move freely and the tube operates efficiently as an amplifier for weak impulses.

While this method of using rotating oil pumps is satisfactory for most receiving tubes which do not run under too heavy a load, difficulties arise in the production of vacuum in transmitter tubes, X-ray tubes and other high-powered electronic devices.

Gas may suddenly break out under the severe strain of energy transformation taking place in the tubes which carry many kilowatts of power. A higher degree of vacuum and better degassifying methods are necessary in such tubes and use must be made of other pumps, capable of producing pressures below 10-millimeters.

For bringing about a lower vacuum than can be obtaihed commercially with present-day rotating pumps, molecular pumps of the condensation and vapor-jet types are used.

The first one of this type is the diffusion pump. It goes back to an old principle worked out by Dr. Wolfgang Gaede as early as 1913. His invention concerns itself with an apparatus for production of a high vacuum with the aid of diffusion. The vessel which is to be evacuated is connected with a pump through a porous wall. The openings in this wall must be so fine that their diameter is not bigger than the dimension of the average free mean path of the gaseous molecules. While one side of this porous diaphragm is in contact with the space which is to be evacuated, the other side of the diaphragm is filled with mercury vapor which is as free as possible of air. Such a vapor may be either mercury vapor or certain oils of low vapor pressure. Gas of the vessel to be evacuated thereafter diffuses through the vessel and is carried away by the vapor. If the vapor which is diifusing in the opposite direction is condensed by a radiator, or if it is absorbed by chemical materials, a vacuum is produced in the vessel.

High-Vacuum Pumps

As the average free mean path of the molecule decreases with increasing pressure, the openings of the diaphragm must be finer the higher the vapor pressure. On the other hand the openings may be so large that wire mesh may be used if the diffusion is aided by a relatively high "fore" vacuum. The bringing about of condensation and diffusion has also been used in the construction of Dr. Langmuir's condensation pump. While diffusion pumps are able to go down to small fractions of a micron, their speed is not quite what is necessary for actual production work. For bringing about quicker action, mercury vapor pumps have been constructed which, working in several stages, bring about a successive acceleration of molecules which are to be carried out from a vacuum container.

Figures 3 and 4 show a design of a multi-stage high vacuum pump as constructed by E. Leybold Co. Referring to the view shown in Figure 3, the suction member consists of a nozzle (a) from which a jet of mercury vapor may be impelled directly downward, a metallic member (b), having an axle bore (c) in alignment with the nozzle (a), spaced at an interval from the nozzle (a) and constituting, with the nozzle, a vapor ejector. The member (b) is fitted within a water-jacketed cylinder (d) so as to be in thorough cooling contact therewith.

The jet of mercury vapor rushing from the nozzle (a) passes into the bore (c) of the cooling member, where it is immediately cooled.

As long as the air passing through the chamber above the member (b) exceeds that of the jet of vapor passing into the bore (c), and the air is drawn by suction through the narrow space betwen the rim of the nozzle and is carried along by the jet, the pump acts as an ordinary ejector.

If the air pressure has gone down considerably, the tube works thereafter as a diffusion pump, the molecules of the gas diffusing into the space filled with mercury vapor, which is condensed thereafter.

Figure 5, extreme left, shows the outside view of a multi-stage metal pump (combined ejector and diffusion). In this same photograpp, Figure 6, on the right-hand side, is a close-up of the inside assembly of this same pump. Figure 7, photograph at the right, shows the extreme vacuum type of pump of Schirman, with which it is claimed pressure as low as .0000000000000000001 mm. Hg. can be reached at a velocity between 20 and 30 liters per second.

Figure 5 shows a photograph of a complete pump of this type, while Figure 6 gives a close-up of the multi-stage jets within the pump.

While pumps operating along this physical principle are very effective in their operation, there may be mentioned one simpler pump which, though little known to industry, brings about a high-speed production of extreme vacua. In the vacuum pump of Mr. M. A. Schirmann, shown in Figure 7, gas coming from the vessel which is to be evacuated is brought into contact with a stream of mercury vapor along a large surface. This is done in three different ways:

1. The vapor is ejected from a nozzle of considerable diameter, thus having a great circumierence and a large active surface.

2. The speed with which the vapor is propelled is considerable; therefore the mercury vapor is projected for several inches into the space within the pump.

3. For still further straightening out of this vapor over distances as large as possible, a metal mesh is inserted, which acts somewhat similar to the guides in a vapor turbine.

Along this pre-exhausted metal net the transportation of molecules from the vacuum towards the "fore" vacuum takes place with such speed that a vacuum of .000000001 can be reached in 10 minutes, while it takes over an hour to get the same result with most of the condensation pumps.

There is still one more thing to be mentioned in the pumps described above, which is that practically only mercury was used. For liberating the vacuum from mercury vapor, it is necessary, in such pumps, to condense the mercury vapor in liquid air traps. An improvement has been made recently by Dr. K. C. D. Hickman of the Research Laboratories of the Eastman Kodak Company, by introducing Butyl phthalate instead of mercury.

Butyl phthalate and certain oils of the apiezoil type have a vapor pressure below .00000001 millimeters of mercury at room temperature and thus are able to replace mercury under certain conditions.

The allthor suggests that high vacuum might be produced by having attached to the chamber being exhausted a glass tube with a high-speed fan run electrically with an outside armature in which would flow a rotating electric field. The Purpose of the fan would be to push gas molecules in the direction of the fore vacuum.

Figure 8 is a schematic diagram of an idea for a new design of simplified high-vacuum pump.

According to this theory, the motion of the molecules from the high vacuum toward the fore vacuum is not effected by a stream of vapor (which always necessitates provisions for cooling, condensation and heating), but by inserting into the vacuum a small, pre-exhausted turbine wheel, (W). It would be driven by a rotating electromagnetic field (F) placed on the outside of the glass tube. As half of the turbine is encased in a container, only the upper half, which moves constantly in one direction, is protruding. Its motion would be impressed upon the remaining molecules, thus initiating a stream of gas towards the "fore" vacuum. This would eliminate the necessity of vapor traps in the high vacuum part of the apparatus.


Back to Antique Radios