Tubes which meet all normal tests, but fail to work satisfactorily in certain critical applications, are called "subjective" failures. In some other application, they may prove to be entirely satisfactory. In this section, we are going to limit our discussion to those failures which come under the broad headings of hum, microphonics, and noise. There are, however, additional types of subjective failures which will be discussed in another section under the heading of selected tubes.
One of the major difficulties involved in studying subjective failures is the fact that their characteristics are so intimately associated with circuit considerations. This makes their measurement and evaluation extremely difficult. In fact, there is almost no way of defining these characteristics except in terms of their effect upon a particular circuit. If you place these tubes in some other circuit, their performance will be completely changed; some times for the better, sometimes for the worse. It is this inability to predict performance on the basis of standard tests which makes the control of these characteristics so difficult for the manufacturer and the user.
HUM
Hum is the name given to that form of unwanted signal which originates from the power line. It may be at the power line frequency or any multiple of it. There are several methods by which these unwanted signals can become superimposed upon the wanted signal. An understanding of each of them is essential before taking up the subject of corrective measures.
As was mentioned in an earlier section, leakage current which passes between the heater and cathode can be one source of hum. This is usually the result of a diode action in which the very hot heater wires actually emit electrons to the cathode during the half cycle when the heater is more negative than the cathode. The path for this emission current may be from the bare ends of the heater folds, or from cracks or chips in the heater coating inside the cathode sleeve. This particular form of leakage is very hard to detect because it often changes with each new heating cycle or after very slight jarring of the tube. This form of interference can be identified from the several other possible sources by isolating the suspected heater and placing about 50 volts between it and the cathode return point. The positive terminal should be at the heater, and the negative terminal at the cathode return (Fig. 3-1). Another possible source of hum is emission from the heater to the grid. This is particularly troublesome in circuits which have a very high grid resistance and depend upon this for tube bias. Emission current, from the exposed ends of the heater, either at the folds, or at their ends, may go directly to the grid side rods. This form of hum will frequently be combined with and difficult to distinguish from heater-to-cathode leakage hum. A relatively simple test is to ground the grid at the socket.
If the hum disappears, it is being picked up on the grid.
Heater-to-grid is not the only method by which hum can be introduced into the grid circuit. There are many electrostatic and electromagnetic fields present in most all electronic equipment. There are power-line fields just about everywhere that electronic equipment is to be found. These fields may be very weak; however, they are often strong enough to modulate some of the weak space currents that reach the tube grid, and cause a hum signal to be superimposed upon it. Evidence that the cause lies outside the tube can generally be established if shielding the tube reduces its effect. Of course, all high impedance leads connected to the grid must be meticulously shielded also.
Fig. 3-1. Placing 50 volts between filament and cathode-return point
to identify interference.
Electromagnetic interference is not always recognized for what it is and, consequently, may be difficult to cope with. Wherever there is electromagnetic interference, there is usually also electrostatic interference. The latter must be completely minimized before the effects of the former can be evaluated. This means that the affected stage must be completely shielded from electrostatic interference first. The effect of heater-cathode interference must be eliminated by heater biasing. The test for electro magnetic interference is a permanent magnet. When this magnet is brought near the suspected tube, the hum will be found to increase or decrease, depending upon the position of the magnet. Hum which can be affected by a DC magnetic field is magnetically caused and is usually the result of uncanceled heater current fields within the tube.
Hum Bucking
Frequently, it will be found that making one or more of the suggested tests in an endeavor to isolate the cause of hum in a circuit actually causes the hum to increase.
This is a very confusing situation until one understands more thoroughly just what is going on. Because of the many methods by which power line interference can take place, considerable distortion of the sine wave source usually develops. This takes the form of both phase and amplitude distortion. The result is a net signal which is often very complex in structure. If one element of this complex waveform is altered, as for instance, by biasing heaters positively, the resulting elimination of one signal source, with its particular amplitude and phase, may now unmask another source. The net effect is an increased amplitude of the interfering signal in the output.
There are two ways of working with this problem. One is to systematically eliminate each possible source of the interference and arrive at a design that is both theoretically and practically correct. This is the course most often followed in designing the better electronic equipment.
But in the mass produced equipment, where price is often a major consideration, certain short cuts are sometimes resorted to which introduce complex problems.
One of these is the technique known as "hum bucking." Hum bucking is the intentional introduction into the signal circuit of a small amount of hum voltage of such magnitude and phase that it tends to cancel the residual hum.
This is mainly a circuit consideration, but inasmuch as it has a considerable effect on how tubes appear to per form, it is important that we discuss it.
Several methods are used to introduce a hum bucking voltage; however, we will discuss only the two most common ones. One of these is the use of a small air core coil located near some heavy power frequency field, such as a power transformer. The coil is made movable for orientation so that almost any amplitude or phase can be picked up from the field around the transformer. The coil is placed in the position which results in the least hum in the output circuit.
Another method of hum bucking is the use of a hum balancing control, usually a low-resistance potentiometer, connected across the heater winding with its center arm connected to ground (Fig. 3-2). Its effectiveness depends upon the introduction of a small amount of unbalanced power frequency into the input stages. This unbalanced power frequency is used to buck out additional hum volt ages that are picked up elsewhere in the circuit. If the adjustment for minimum hum comes somewhere near the center range of the control, it is not being used to intro duce a bucking voltage. However, when the hum is mini mum at some point near either extreme end of the control, it is being used as a hum bucker.
Wherever hum bucking is resorted to, it becomes very difficult to analyze tubes and their relation to the hum problem. The reason has been suggested before. If a tube having a given amount of hum leakage is used in a circuit that has been adjusted for minimum hum, the substitution of a new tube having less hum leakage will result in an apparent increase in the hum level. If the average level of the hum introduced by several tubes is sufficient to cancel each other, the introduction of a new tube having either more or less hum leakage will result in a notice able increase in the over-all hum level. Very often the adjustment range which is possible within the hum bucking circuit is insufficient to compensate for the large change introduced by a replacement tube, be it better or worse than the original tube. This makes a much better tube actually perform no better than a much worse tube as far as the net effect is concerned.
Fig. 3-2. A circuit with hum bucking and hum balancing.
Corrective Measures
There are certain accepted methods for minimizing the hum problem and a knowledge of these techniques can be of great assistance in trying to separate those situations that are brought on by poor circuit design from those that are strictly tube problems.
The heaters of all high-gain audio equipment should always be powered from a center-tapped transformer winding. Grounding the center tap reduces the heater-to-cathode voltage. This will more than halve any leakage current which might otherwise flow in this circuit.
When one side of the heater circuit is grounded, not only is the heater-to-cathode voltage increased by a factor of two, but another source of power line interference is introduced. This interference results from the creation of large chassis currents which are in series with the normal signal ground returns. These chassis currents make the selection of ground points very critical and al most inevitably lead to a system of hum bucking.
Wherever possible, the cathodes of very high gain input stages should be grounded and grid bias should be obtained by means of relatively large grid resistors which make use of the normal grid current effect. Where this method is not used, adequate cathode bypassing should be employed. The effective impedance of the cathode by pass capacitor should not be more than 10 percent of that of the cathode resistor at the power line frequency.
There are special circumstances where bypassing is out of the question, as for instance, when the feedback loop is introduced across a fairly large cathode resistor in an input stage. In these instances, serious problems may arise from even a very small amount of heater-to-cathode leakage, the exact magnitude of the problem depending upon how large a value of cathode resistor is used.
One method of dealing with this problem is to isolate the heater supply to these tubes from ground and return it to a voltage point of approximately +50 volts. This is called heater biasing, and it depends for its success upon the fact that heater-to-cathode leakage ( either direct or via an emission route) is much less when the heaters are positive than when they are negative.
Another method of solving the leakage problem in critical sockets is to operate the heaters of these tubes from a DC source. Sometimes the AC source voltage is rectified and smoothed, using a large, low-voltage capacitor, and then fed directly to the tube heaters. This is an excellent solution, but may involve some added expense. Most high gain audio equipment terminates in a power amplifier stage. When this is true, a very simple source of DC cur rent is available for lighting the heaters of the input stages, practically for nothing.
The output stage of a high-gain audio amplifier often consists of two power tubes in push-pull. There is usually a common cathode resistor which provides bias for the push-pull tubes (Fig. 3-3). The current flowing through this resistor is essentially a DC current because the current flowing to the individual tubes in the push-pull stage is at all times 180° out of phase. The signal current is thus canceled out and only a steady DC current flows through the common cathode resistor.
If the heaters of the input tubes are substituted for this cathode resistor, they can be lighted by this current while their voltage drop provides the necessary bias for the output stage. This is a case of getting something for nothing because the current is there anyhow, so why use it to heat up a cathode resistor when it could be used to light the heaters of one or more of the critical tubes in the circuit.
It is surprising that this system of hum control is not resorted to more often because it has other advantages that are not so immediately obvious. What better way is there to improve the life of those small input tubes than to protect their heaters from sudden current surges when they are turned on? By using this DC current to light them, these critical tubes enjoy a very gradual warm-up and cooling-off period. This has the effect of greatly in creasing their heater life-their principal cause of failure in most applications.
Fig. 3-3. Using amplifier cathode current for heaters. The heater voltage
drop provides the necessary bias for the output stage.
Where DC on the heaters is not a practical solution, a very economical method of reducing heater-to-cathode leakage in high-gain audio stages is simply to operate the heaters at about 10 percent below their normal ratings.
This is possible with most applications of this type be- cause they usually use the tube in a resistance-coupled amplifier, where plate loads are so high that very little actual current flows from cathode to plate. These are voltage amplifier stages and do not require very much current from the cathode in order to perform their functions adequately. The tubes used for this purpose are designed with much greater emission capabilities than they are usually called upon to deliver. Therefore, operation at reduced heater voltage will provide all the current that is needed for a voltage amplifier stage. Also, the lowered heater temperature will often result in a very great reduction in leakage and hum. An added advantage will be improved heater life due to the lower heater temperature.
The use of tubes in series strings in high-gain audio equipment is not a recommended practice. The problems introduced by this form of operation are quite large, especially when consideration is given to the connecting together of several elements into a complete system. The danger of introducing heavy chassis currents, with the consequent high hum levels resulting from them, keeps most designers away from this type of circuitry. Where it is resorted to, in spite of recommendations to the contrary, there are certain precautions that will prove to be advantageous.
The most critical tube should be wired so that its heater is nearest to the grounded end of a series string. The second most critical tube should be the next tube in the string, and so forth. The most critical tube will be the tube nearest the input, or the one nearest the input which uses the largest, unbypassed cathode resistor.
In order to reduce the heater temperature of critical tubes in a series-string circuit, it may be necessary to shunt them with a low value of resistance. This should be accounted for in the design so that the remainder of the tubes in the string do not operate at above normal cur rents.
Special Tubes
There are available certain special tubes designed to operate in the critical sockets of high-gain audio equipment, and to minimize those causes of hum which originate within the tube itself. These tubes have not been mentioned until now for the following reason. The user must realize that they may not always improve the over all hum level when substituted for certain older types, because of the many interrelated causes for hum interference discussed earlier.
A low-hum, high-gain tube must have very low heater to-cathode leakage-either by direct current flow through the insulation, by emission from uninsulated areas, or by induction and consequent modulation of the electron stream. All of these requirements are best achieved by the use of a "coiled heater." The ordinary heater used in most vacuum tubes consists of a length of insulated tungsten wire folded into a neat bundle, and then inserted into the cathode sleeve. In order to fold the insulated wire, the insulation must be removed at the end of each loop. These exposed loops of bare wire are the trouble starters, because they can emit directly to the cathode. They move as a result of expansion and contraction and eventually short to each other, or to the inside of the cathode sleeve. Finally, because the heater bundle is a randomly dispersed mass of folded wire, it does not effectively cancel out its own magnetic field.
Some may do better than others, purely by accident. Those that are satisfactory at first may become re-dispersed due to differential expansion and contraction and cause trouble later.
The coiled heater is made like a long spring. It is coiled and then folded once, in a gentle bend at the top, and then the entire heater is coated with insulation. It has no open or exposed ends to emit or eventually short out. It expands and contracts around the turns of the spring and, in this way, almost eliminates longitudinal stretching. Finally, it is a perfect coil and tends to cancel its own magnetic field very uniformly.
Tubes of this general construction are known to have much lower leakage paths between the heater and all other elements. They can be used to provide greatly improved performance in critical sockets where hum is a major problem. They should provide longer life and greater over-all reliability. But it should be borne in mind that when substituting this type of tube in older equipment not designed around them, the immediate results may seem disappointing. If the hum problem is isolated and reduced at its source, as suggested in the foregoing paragraphs, there is then every reason to expect improved performance through the use of these improved tube types.
MICROPHONICS
Microphonics is defined as any signal appearing in the output of the vacuum tube which originates within the vacuum tube. It is caused by some mechanical stimulation of the tube. Microphonics is classified as a subjective failure because it too depends very much on the use to which the tube is put and the physical nature of its environment.
An outstanding example is often found in phonographs or compact, miniature radio sets. A particular tube will be found to cause the equipment to break into a sustained howl when used in one model; however, the same tube will perform quite normally in another model. It is not uncommon to find that a tube which causes sustained howling in one unit may be perfectly normal in another unit of the same make or model. The answer to this apparent contradiction lies in understanding the nature of this particular tube characteristic.
There are three ways the electron stream within the tube can be modulated. One is through the use of a magnetic field; this is demonstrated in the picture tube. The second method is electrostatic; this is the most commonly used in controlling all vacuum tubes. The third method, seldom used but just as valid, is to change the positions of the elements while holding the voltages constant. This is what occurs when a tube responds to mechanical vibration and reproduces the waveform of the stimulating energy as an AC component in its plate current. This third method is also the cause of microphonics.
Since the moving element within the tube may be an entire structure (the cathode assembly) or only part of a structure (a grid side-rod or a single grid turn), the frequency at which the element vibrates may cover a very wide spectrum. It them becomes almost impossible to measure a specific tube in terms of its susceptibility to microphonics. Some effort has been made to analyze the spectrum response when a vacuum tube is stimulated by a sharp impact. Although there is a correlation between tubes that are very tight and those that have very little output at any audio frequency, the fact also remains that many tubes having relatively high outputs at certain specific frequencies do not appear to be microphonic. Apparently the mechanical stimulation they encounter does not fall into that portion of the spectrum.
There are certain limited ways of dealing with micro phonics. All forms are affected by the amount of mechanical stimulation delivered to the tube elements. Therefore, any technique which lengthens or raises the impedance of the path by which the stimulation will be fed back to the tube will reduce its susceptibility to microphonics.
Thus, sensitive sockets should be positioned as far from mechanical vibrators as possible. Some vibrators, such as speakers, convey some of their energy directly to the tube via the air space between them. In such instances, it is desirable not to place the socket where the maximum sound pressure can be developed against the tube envelope.
Most mechanical energy is fed to the tube elements by means of the common supporting member, such as the chassis. Standing waves are often present which effectively couple a high percentage of the energy directly to the tube elements. Two methods are very effective in breaking up the standing-wave pattern and hence in reducing the feedback loop. One of these is to add a weighted shield over the tube, thereby changing its vibration period and eliminating the mechanical resonant circuit. Another method is to isolate the socket from the chassis by using rubber grommets.
In a few cases, low microphonic tubes are available for some of the more critical applications. These tubes at tempt to solve the problem through the use of tube structures that are known to be less susceptible to micro phonics. These techniques cannot be applied to very many tubes because they will alter the performance of these tubes too greatly. Therefore, they are appropriate for only a limited number of tubes.
In general, the technique is to shorten the unsupported distance between the top and bottom micas as much as possible, thereby reducing bending of elements. Heavier micas are also used ; they are sometimes forced into the envelope under tension in order to stiffen the entire mount. These tubes often are effective in reducing microphonics, although depending upon them to always solve the problem is not recommended because some of them are inferior to their prototypes or to other, more efficient tubes.
NOISE
Noise is defined as an output signal originating within the tube and is not specifically hum or microphonics.
There are several possible sources of noise signals.
Whether they constitute a defect depends upon how the tube is used. Hence, noise is considered to be a subjective failure since the actual application of the tube is the significant factor in determining whether or not it actually becomes classified as a failure.
The most common form of noise within a vacuum tube is that caused by intermittent shorts or opens. The nature of these intermittents is such that it makes a big difference how much voltage is applied between elements and at which source impedance, as to whether they exist or not. Also, it is important to define what kind of detection device is going to be used before you can decide whether you have an intermittent or not.
The classic short tester is a neon lamp in series with a voltage source and a resistance to limit the current flow.
It is assumed that if an intermittent short occurs, a circuit will be formed and a current will flow. At the same time, the lamp will flicker and the user will know that he has an intermittent short. But will he? The neon bulb is often used in a self-rectifying circuit with an AC volt age applied to it. The voltage required to cause the lamp to ionize is not present during the entire cycle. During a very significant portion of each cycle, the lamp cannot light even if a dead short occurs because the voltage across the circuit is insufficient. Suppose that the voltage is just sufficient, but falling, when the short occurs. The neon lamp requires a finite time to ionize and fire; hence, if the short occurs at the time the voltage is approaching the critical value there will be insufficient time to record it and it will pass unnoticed.
There is also the matter of non-repeatable shorts and intermittents. Particles sometimes shake loose from within the tube structure and drop down between the elements.
These occasions produce momentary shorts which in all probability will not occur again. Nevertheless, in some pulse-triggered applications, as for instance, in counting circuits, these random noise pulses can cause false operation.
Repeated testing of tubes in an endeavor to eliminate this failure has been unsuccessful. An exact explanation for this involves problems in statistical probability that are not within the scope of this discussion. However, it can be mentioned that the use of a statistical approach to this problem can be quite successful. For instance, if 100 tubes are measured and the number showing momentary shorts is noted and compared with another lot of 100 tubes similarly tested, the lot having the significantly lower number of random intermittent shorts will always show this characteristic, no matter how many times it is re-measured. This means that statistical sampling can be used to indicate the probability of the occurrence of intermittent shorts in future use.
The second and most common form of noise in vacuum tubes is sometimes referred to as "frying noise." It is most often the result of leakage paths across the micas.
The measurement of this characteristic is complicated because it has impedance and frequency characteristics that make a universal test very impractical. For example, there are tubes which will produce considerable noise when tested in a high-gain RF amplifier, but none when tested in a high-gain audio amplifier. The converse is also true. Tubes which produce extraneous noises in an audio amplifier may produce no noise at all in a high-gain RF amplifier. The reasons for this observed phenomenon are not well known. Neither is it thoroughly understood why the relation between the amount of noise detected and the sensitivity of the amplifier used to detect the noise is not a linear function.
Noise in vacuum tubes is a problem which depends al most entirely upon the users' requirements. It does not lend itself to accurate definition and there are very few methods, if any, by which the user or the designer can protect himself against it. About the only method is to make use of the laws of probability in some way or an other. In fact, so many vacuum-tube characteristics resolve themselves into a matter of statistical probability, it is felt that some understanding of this subject is essential before proceeding into some of the other areas of vacuum tube knowledge. For this reason, the subject of characteristic variables and their normal ranges and limits is discussed in Section 4.