Spanish Italian
17447 Users    

Vibrator Power Supply Design - Vibrator Power Supply Construction - I

  Download PDF version of the Article

Vibrator Power Supply Design

Vibrator Power Supply Construction and Interference Elimination - I
 
The discussions in the previous chapters have been limited to the three primary components of the vibrator power supply system and to the influencing factors of the applications in which a majority of the vibrators are used. These applications require the conversion of a low voltage direct current to a high voltage direct current and are usually radio receivers of the automotive, home, or portable types. The general construction of the power supply, the methods of interference elimination, and the values of the associate components are important to the successful conclusion of the design. There are a few similar applications which require additional facilities from the power supply. These facilities will be discussed in this chapter.


There are some variations in the power supply requirements that are fairly simple, while others will require additional skill gained by experience in the field of vibrator and power supply design. The services of the P. R. Mallory Company's Engineering Department are available for those who wish to consult with them. These variations include such requirements as additional windings on the transformer for tube heaters, for bias voltages, etc., or even an extra winding to enable the apparatus to operate from an AC line. The factors involved where any AC load is supplied by the vibrator transformer, include the differences in the wave-form and frequency between the 60-cycle sine-wave AC, for which most components are designed, and the 115-cycle square wave of the vibrator supply. Since this AC load will probably be permanently connected to the transformer, its character will affect the starting characteristics of the vibrator. For instance, tungsten filaments such as are used in radio tubes, or lamps, have exceedingly high ratios of hot-to-cold resistance values. While their operating load may be low, the starting load is very high, and special precautions are usually required to protect the vibrator from these starting surges. One solution would be to turn the heaters on after the vibrator has been started. Another would be to provide for a limiting resistor in series with the heater supply lead, etc., and then supplying a somewhat higher output voltage to overcome its normal drop. Other variations should also be given special consideration.
All of the power supply components, including the smoothing filter, shields, and interference filter parts, must be arranged and assembled on the chassis in the most advantageous manner. While the application requirements may have considerable influence on the lay-out, there are certain procedures that can be followed to advantage.
The amount of smoothing filter required will depend upon the requirements of the apparatus for purity of the DC output. Because of the sharp discontinuities of the rectified DC, a capacity input type of filter is required. Most power supplies do not supply a widely varying load, so the regulation advantage of the choke-input type of filter is not of such great importance.
Where the power supply requires efficient filtering, a capacitor-input filter with a reactor should be used. However, where size and cost of the component parts are important, a resistor type of filter may be used. If the entire load current needs filtering, the value of the resistor must be kept low or the voltage drop in it will be excessive. However, if a large portion of the load current can be used with only a capacitor filter, this portion can be connected to the input capacitor of the filter and the remainder of the current can be supplied through a higher resistor value without incurring too great a voltage drop. This system is often employed in low-priced auto radio receivers, where the output power tube, which constitutes the large share of the load, is supplied from the input capacitor. The voltage ripple appearing on the DC will have little effect, since there is no amplification following this tube. The other tubes, which provide high amplification in the receiver, are well filtered through a comparatively large resistor and an output capacitor.
The minimum capacitance that should be used in the input section of the filter to secure proper vibrator operation is about 5 mfds. Normally, a minimum value of 10 mfds. is recommended. The output section can be any amount desired, although a value larger than 40 mfds. is seldom required. These values refer to smoothing filters such as are normally used in the high voltage supply of radio receivers. If a smoothing filter is required for low-voltage, high-current DC, such as used with tube filaments, much higher values of capacitance are necessary. In general, the required values of capacitance are approximately 1000 mfds. per DC ampere for the input to the filter, and 5000 mfds. per ampere on the output.
The power supply unit should be assembled in a compact manner to keep the leads as short as possible. Care should be taken to keep the vibrator and the electrolytic capacitor as far away as possible from any large sources of heat such as the rectifier and power output tubes. This is sometimes difficult, but by taking advantage of the "chimney-effect" in ventilation, and by placement of these parts near outer-case walls, the ambient temperature effects on these components can be greatly reduced.
Where the power supply is constructed on the same chassis as is. the radio receiver, care should be exercised to locate such interference producing components as the vibrator, rectifier and transformer as far away from the antenna and radio frequency portions of the set as is possible. These components must be well shielded and securely grounded to the chassis. The rectifier tube, if used, can be placed in such a location that it may be shielded sufficiently by other components so that an individual shield will not be required.
Numerous receivers have been manufactured in which the power supply was completely isolated from the remainder of receiver by a metal partition. This arrangement has been very satisfactory wherever close-fitting and securely fastened joints between the partition, chassis, and case were used.
Present assembly methods have eliminated the use of partitions above the chassis and only a shallow metal box is used below the chassis to shield the power supply wiring and RF filter components. Transformers are usually mounted in steel cases and are well shielded since unshielded transformers are potential sources of interference. There are several other very effective methods of shielding the transformer.
The vibrator is the major source of interference, but its drawn metal container is a very effective shield. It should be well grounded to the chassis, however. The most effective method of grounding is by the use of a "ground cup." This cup is riveted to the chassis with the same rivets which hold the socket. The vibrator case slides into the spring fingers of the cup as the plug base enters the socket, thereby providing an excellent RF ground. The vibrator container can be grounded through one of the base pins but this is a very poor method because of the high RF impedance through the pin and socket connections.
The RF interference filter components, especially the RF chokes, should be so located that they do not radiate interference to other portions of the receiver. Also, the output leads should not pick up radiation from the input leads or from other exposed wiring. Transformer leads should be twisted to prevent loops from being formed. Any leads carrying RF interference should be as short as possible, and close to the chassis. Critical leads should be shielded. These are some of the important considerations in the elimination of RF interference, more commonly known as "hash."
Ground currents in the chassis and outer-case often introduce "hash" in the RF section of the receiver. These ground currents are transferred to the RF parts by currents flowing through a common path that includes at least a part of the chassis, etc. The positioning of parts and location of grounds is probably the greatest contributing factor. This is a very difficult trouble to avoid, and even more difficult to overcome. There are no rules or formula to follow for this problem. However, a few precautions may be of assistance. Avoid "loops" of any wire carrying vibrator current from being formed around any portion of the chassis, no matter how small; re-position transformer or improve shielding; check every chassis and outer-case weld, rivet and screw for good RF connection; relocate vibrator circuit grounds and also RF circuit grounds; and always connect the grounded filament lead to the socket terminal next to the grid socket terminal.
One of the most common causes of ground currents in the chassis is the routing of transformer leads through the chassis so that the center-tap leads are through one hole while the end-taps are through another hole. Thus the entire input current circuit forms a magnetic turn around a portion of the steel chassis. The induced voltages set up circulating currents, all modulated with "hash" frequencies. As an example, the ground currents were eliminated in one receiver by sawing a slot between the two lead holes in the chassis.
One general rule that should always be followed in "hash" elimination work is that any magnetic or electrostatic shield should be closed on all sides, with good electrical and mechanical contact made in all joints. This provides a short-circuited path in all directions of current flow.


Figure 58 illustrates the typical interrupter vibrator circuit diagram, with the minimum amount of interference filtering provided. A RF by-pass capacitor is connected from the transformer center-tap of the primary to ground, and a single RF choke is provided in the "A-hot" or battery lead. For the secondary, or output circuit, a single RF by-pass capacitor is provided, although this is eliminated sometimes by the use of an electrolytic capacitor having low impedance characteristics.

 

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.

More information about formatting options

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.
11 + 4 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.

Who's new

  • JM
  • samsilva77
  • araghube
  • stoll
  • mt
  • orionkw
  • pulper
  • mauriss
  • jbares
  • christiank79

Who's online

There are currently 0 users and 45 guests online.