SKTF -- Fall 1980

A Quarterly Publication of The Smith-Kettlewell Eye Research Institute’s Rehabilitation Engineering Research Center

William Gerrey, Editor

Issue: SKTF -- Fall 1980

Original support provided by: The Smith-Kettlewell Eye Research Institute and the National Institute on Disability and Rehabilitation Research

Note: This archive is provided as a historical resource. Details regarding products, suppliers, and other contact information are original and may be outdated.

Questions about this archive can be sent to sktf@ski.org

TABLE OF CONTENTS

Dedication

Preface

Foreword

Checking Diodes for Maximum Peak Inverse Voltage

Soldering (Part I)

3 Terminal Monolithic Voltage Regulators

The LM 386 High-Gain Audio Amplifier

Point-To-Point Wiring on Perforated Board

Editor's Corner

DEDICATION

Good teachers are those who invest in the quality of their students. Good teachers, by example, seek to impart to their students a respect for the power of independent study.

An investment having been made in them, those students come away with two blessings -- dignity and the joy of fulfillment. Dignity is theirs when self esteem is built on the foundation of independent study; it is communicated when they become examples to their successors. And ever present is the joy and fulfillment one receives from crafting with the tools of his age.

Mr. Gunderson, for your investment -- dedicating your life to teaching -- this first issue, edited by a student of your writings, is dedicated to you.

PREFACE

Although many textbooks have been translated into braille and onto recorded media, blind technicians and scientists are faced with the continual frustration of obtaining current supplemental materials.

We propose to publish a newsletter quarterly in braille, large print, and recorded form, which will serve as a guide to the current technology as applied to the needs of blind and low vision people. We anticipate this will narrow the existing gaps, and allow technically-minded visually impaired people to pursue their interests.

This journal will include electronics information such as lists of materials already transcribed, catalog abstracts, data on integrated circuits, manufacturers' application notes, and when possible, construction details on devices for the blind using this current technology. Mechanical information will be similar: manufacturers' data, plus articles describing blueprints and techniques for the construction of the devices.

General interest "do-it-yourself" descriptions of such processes as soldering, project layout and the use of power tools will focus directly on techniques used by the blind. Articles will be solicited from readers, training facility personnel, and other interested professionals and non-professionals.

FOREWORD

With the passage of the Rehabilitation Act of 1973, Rehabilitation Engineering Centers were set up to utilize current research and technology for the creation of devices to benefit disabled people. The Rehabilitation Engineering Center here in San Francisco at Smith-Kettlewell Institute has a federal grant to apply current technology to solve problems of blind and visually impaired persons. I, a blind person, work as a rehabilitation engineer at the Center.

The questions that make a conscientious engineer squirm in his sleep are as follows:

Who are the people we serve? How can we properly assess their needs? Given identifiable needs, what is good engineering? How can we expedite the "trial and error" process of optimizing the solutions? Can one engineering staff adequately take on this responsibility?

Obviously, only a reciprocal interactive relationship between rehabilitation engineers and the disabled consumers will permit these questions to be properly addressed. It is my experience that the most "to-the-point" engineering takes place when we, the consumers, fully participate in the needs-assessment and problem-solving processes. We must all become the engineers.

According to my American Heritage Dictionary, the word "engineering" and "ingenious" have a common Latin root, "ingenium," which means talent or skill. In medieval Latin, "ingeniare" was to contrive. In the definition of "engineer," the dictionary quotes the writing of Robert Graves, "Claudius's murder was engineered by his wife, Agreppinella." How about that! She didn't even have a Bachelor's degree. Engineering is contriving solutions to problems in a creative way.

Visually impaired persons have adapted to their impairment by "engineering" solutions to their difficulties. Whether a solution is an adaptive technique, re-allocating responsibilities in a job situation, or a modified monkey wrench, it is still engineering.

The blind and visually impaired community is very much the loser for not sharing individual rehabilitation engineering solutions amongst ourselves. As an example, if I hadn't heard about carrying a little rubber sheet in my wallet to serve as a portable signature guide, I might frantically look around for a suitable straight edge every time I sign my name.

In 1950, a radio and electronics magazine was brought into being through the concentrated efforts of one man, Robert W. Gunderson. This magazine, the Braille Technical Press (BTP), changed the face of rehabilitation for blind people interested in science. It was the only source of current information about electronics components and modern circuit designs to which blind people has independent access. More importantly, the BTP became a forum for the exchange of ideas about assistive devices applicable to the technological fields. Dozens of electrical measuring instruments were developed and redesigned by readers who were eager to solve their own problems. Many of these instruments are commercially available today from agencies who became aware of their market through the forum of BTP.

Of significance to me was that this magazine exposed its readers to an expanding peer group which would have been unknown to me if it had not been for BTP. Articles contributed by hobbyists, engineers, mathematicians, and computer scientists gave me a sense of identity with blind people in all levels of technology. I chose engineering as my career, armed with the knowledge that I would not be the only blind engineer; others had solved related problems before me.

In 1980, just seven short years after the BTP is no longer being issued, the lack of knowledge among the blind about their contemporaries is discouraging. Blind people ask me questions like, "Is there a way that we can set the recording level on a tape recorder?", or "How can we solder?" I hear blind people making statements like, "Of course, blind people cannot work with complicated electronic circuits because of the diagrams."

We cannot blame society's lack of understanding for our ignorance; we must blame our careless isolation which we have allowed to happen. We have isolated ourselves by not sharing our experiences ad by not working toward a collective understanding and identity. In the words of Walt Kelly's delightful cartoon character, Pogo, "We have met the enemy, and he is us."

Smith-Kettlewell as a forward-looking organization, and I as a blind person in a technical profession, offer this publication as your forum. Delight in these pages, reciprocate into them, and meet your strongest ally, for he is also us.

CHECKING DIODES FOR MAXIMUM PEAK INVERSE VOLTAGE

by Robert Gunderson W2JIO

How often have you purchased a handful of surplus diodes, advertised as 1,000 MPIV (maximum peak inverse voltage), and found to your dismay that they quit at a much lower voltage value? A number of these "bargain counter" experiences convinced me that I ought to have an instrument for sorting diodes before putting them to work. This simple-to-build device does this job very well; if you have need for such an instrument, read on.

The forward voltage drop of a germanium or silicon diode is very low, about 0.3V for germanium and 0.7V or so for silicon. However, when the diode isn't conducting (when its anode goes negative on the opposite alternation of the input voltage source), the maximum voltage of the generator, plus the voltage standing across the charged filter capacitors appears across the diode. Suppose the voltage of the generator is 120 volts r.m.s. Its peak value is some 1.414 times this value or 169.68 or 170 volts. This means that the voltage appearing across the filter capacitor in the filter at the output of the diode will also be 170 volts, and these two voltages are in series to produce 340 volts across the open rectifier, open because its anode has gone negative and it isn't conducting. Thus, the peak inverse voltage is nearly 3 times the root mean square voltage of the line. To be on the safe side, a diode having an MPIV of 400 volts would be a good choice for a line-operated supply of this type.

The same reasoning applies to capacitors having an a.c. voltage rating, although it really doesn't belong in an article of this type. The d.c. voltage rating of an oil-impregnated capacitor is about 3 times the a.c. voltage rating. Here again, the charged capacitor's voltage is in series with the generator voltage when the generator polarity reverses, and the peak voltage of the capacitor and the peak voltage of the generator gives 2.828 or 3 times the r.m.s. generator voltage. The diode checker to be described consists of a variable-voltage d.c. supply having two ranges: 0-500 volts and 0-10,000 volts. Further, an aural indicator consisting of a small transistor audio oscillator is included in series with the leads connecting the diode to the test voltage, so that when the diode begins to break down, the oscillator sounds the alarm. You may include a d.c. voltmeter or the input voltage control (a small variac) may be fitted with a braille scale. Here at W2JI0 I used a surplus 'scope power transformer which provides an output secondary voltage of 7,500 volts with 117 volts applied to its primary. The secondary voltage is rectified by a high-voltage diode and fed to a 0.1 mf. capacitor which provides a peak value of 10,600 volts. The same transformer also provides the low-voltage ranges up to about 500 volts. This is done by means of a 6.3 volt filament transformer interposed between the line and the high voltage transformer primary. A 6 pin plug on the rear chassis apron connects an a.c. relay so that this low voltage a.c. source is switched in when the plug is inserted; and the variac is connected directly to the primary when the plug is removed.

A high resistance bleeder is connected across the capacitor to protect the operator, and a large value of series resistance is also connected between the test terminals and the high voltage d.c. source for the same reason. In addition a low voltage transformer-rectifier combination is included in the chassis for checking zener diodes. It makes use of a wire-wound potentiometer connected across the 100 volt d.c. output, with the arm connected to the output terminal for checking zeners. The 6 volt winding on this low voltage transformer for checking the zener diodes furnishes power for the small audio oscillator used to indicate diode avalanching or breakdown.

The bleeder is a 60 megohm high voltage resistor obtained from surplus, and the series isolation resistor is a 3 megohm spun carbon type. There are three terminals associated with the high voltage measuring circuit; one for the hot positive lead, for connection to the diode under test. The second post is a negative terminal about 100,000 ohms above the actual negative ground for connection to the other side of the test circuit, and the third is a negative chassis ground for connection to the voltmeter, should you desire to check the voltage applied to the circuit. Connection of the voltmeter to the terminal off ground would cause the meter current to flow through the resistance connected from ground to the negative terminal and this would give a false indication on the aural indicator; the voltmeter current would operate the alarm rather than the breakdown current of the diode under test. Incidentally, this instrument is also very useful for checking capacitor leakage resistance -- in fact, it makes an ideal insulation tester.

One final note before describing the circuit; the 6-pin plug should be left in the rear socket at all times, so that the unit will operate in the low voltage mode. Use the high voltage connection only when necessary, and be careful. Remember, too, that the 3 megohm resistor in series with the hot input limits the current to about 3 milliamperes. Don't place yourself in series with this circuit; and if you do, use just one hand. The 60 megohm resistor across the capacitor provides for a bleeder current of about 166 microamperes, and this means that it takes time for the capacitor to discharge. I use a clip lead to short circuit the supply when I have finished using it. The capacitor is discharged when the tone disappears. As the current flows into a test capacitor, should you wish to use the instrument as a leakage tester, you will note that the tone continues until the capacitor is fully charged. Do not exceed the working voltage of the capacitor under test when checking it. If the tone remains high pitched, there is considerable leakage -- or even a short- circuited capacitor. On the other hand, a slow motorboating signal means high resistance leakage; no tone at all means a good capacitor. Remember to discharge the test capacitor before using it; or leave it across the instrument with the supply turned off for a minute or two. Better still, connect a clip lead conveniently to short circuit the supply and its test capacitor after the unit has been turned off.

While a separate supply for checking zener diodes is included in the design, I have found that the high voltage low current supply works very well for this purpose. However, it is perhaps wise to apply greater current densities to these diodes to obtain more accurate results. To date, I have found that both systems give the same answers. Therefore, you may include the 100 volt zener supply or leave it out. Actually, this supply is fitted with a potentiometer so that voltages from zero to about 120 volts can be applied to the zeners under test.

In describing this instrument, I feel it might be easier to simply give the circuit and list those parts which are currently available. The high voltage transformer came from World War II surplus and it is quite unlikely that you will be able to obtain a similar unit. In fact, you may build a checker which delivers any maximum voltage value you choose. If you don't anticipate checking 10 Kilovolt PIV diodes, it would be wise to make a lower voltage unit. Of course, it should be obvious to the builder that the maximum peak inverse voltage (MPIV) of the diode in the power supply should be sufficient for any tests you wish to make on your diodes for future project.

The checker is built into a 5 x 10 x 3 inch chassis and enclosed in an amplifier foundation cover and bottom plate. Actually, this chassis is a bit small for the job, inasmuch as one doesn't cram components up close to a 10 KV circuit. Two of the three binding posts -- hot positive and the negative terminal for checking diodes are mounted on a piece of plexiglass, with adequate cut-out made on the front chassis apron for clearance. The common ground and negative terminal for a voltmeter is mounted through the plexiglass and into the metal chassis. The small 1 ampere variac is mounted on one of the end aprons and fitted with a knob for adjusting the applied voltage. The on-off switch is at the right on the front apron, with the three 5-way binding posts for high voltage measurements at the center. Two additional binding posts are at the left and a small 1 1/2 inch speaker for the aural indicator is located below these terminals. The 10K ohm wire-wound potentiometer for adjusting the zener diode measuring supply voltage is mounted on the left end apron. A 6-prong socket on the rear apron is fitted with an appropriate plug. With the plug in place, the unit is set for the low voltage measurements up to about 500 volts; when removed the primary of the high voltage transformer is switched directly to the variac rather than to the secondary of the 6 volt transformer, used in the low voltage measuring circuit. This plug-in arrangement is an added safety feature. The extractor- type fuse holder on the rear apron accommodates the 1 ampere fuse. The variac is wired for the 135 volt connection although this is optional, depending upon the voltage levels you wish to use.

Circuit description

Side A of the 117 volt a.c. line goes to the common end of the variac, with side B going through the on-off switch (SPST) thence through the 1 ampere fuse to the 117 volt end of the variac. These ends of the variac (connected to the a.c. line) also connect to the primary of transformer T2, (two secondaries, 125 volts at 25 ma., and 6.3 volts at 1 ampere). The coil of the relay K1 also connects to the ends of the variac through the 6-prong socket and its plug which act as the interlock. Pins 1 and 6 are connected to the a.c. line at the variac, with pins 3 and 4 connected to the relay coil (pins 2 and 7 on the octal socket for the relay). The plug is suitably jumpered, with one jumper from pin 1 to 3 and the other from 4 to 6. Thus, with the plug inserted, the relay will close when the switch is operated. Removing the plug disables the relay.

The swingers of the relay are connected to the primary of the high voltage transformer T3, and the normally closed contacts (closed with the coil de-energized) are connected to the slider and to the common side of the variac. The normally open contact (closed when the relay coil is energized), are connected to the secondary of the 6.3 volt transformer T1. Thus, with the safety plug inserted, the primary of the high voltage transformer T3 is connected ot the low voltage secondary, whose primary is also controlled by the variac. Both sides of the high voltage transformer have been switched as an additional safety measure, should the relay happen to hang up on either side.

One end of the high voltage transformer secondary is grounded, while the other side goes to the anode of the high voltage diode D1 (see parts list) with the D1 cathode going to the capacitor C1 (0.1 uf, 15 KV), with the other side of C1 grounded. Capacitor C1 is shunted by its bleeder resistor R1 (60 megohms, see parts list) and the hot side of C1 goes through resistor R1 (3 megohms), to the hot positive input terminal. The negative voltmeter terminal is also grounded. The negative test terminal (the terminal to which a diode or capacitor under test is connected, goes through resistor R3 (100K ohms) to ground. This resistor is by-passed by capacitor C2 (1uF), and this test terminal will connect to the aural indicator, which consists of a simple audio oscillator. If an electrolytic capacitor is used for C2, the positive should be connected to the test terminal with its negative grounded.

Indicator circuit

This simple arrangement provides an output signal when the diode conducts. If you test a capacitor for leakage the pitch of the oscillations in the output will decrease as the capacitor charges, and if these oscillations persist after the capacitor under test is fully charged, the capacitor has leakage. A general-purpose NPN transistor is used in a conventional blocking oscillator, feeding a small speaker. If you prefer, you may omit this portion of the circuit and connect a voltmeter or an auditory tuning across the 100K ohm resistor R3. However, the indicator makes for a built-in convenience.

Indicator circuit

The bottom of the T4 primary goes to the minus 9 volts d.c. The center tap goes to the Q1 emitter, with the top end going through capacitor C3 (0.1 uf) to the Q1 base. The T4 primary is shunted by capacitor C4 (.05 uf). The Q1 collector goes to plus 9 volts. The minus 9 volts is connected to the bottom of the primary. The base goes through resistor R4 (47K ohms) to the junction point at the test terminal, mentioned previously. The minus 9 volt connection and the bottom of the primary goes to the chassis ground. The T4 secondary feeds the speaker.

Power supply for the indicator

This is simply the 6.3 volt winding of the transformer T2, connected in a simple half-wave circuit, using a 1N4001 diode D2, with its anode going to one end of the 6 volt secondary, and the other end of the secondary connected to ground, or chassis. The diode D2 cathode goes to the positive side of capacitor C5 (25 uf), with the negative side of this capacitor grounded. This is the plus 9 volt point for the indicator. There is no switching in this portion of the circuit, inasmuch as the oscillator is turned on only when current flows through the resistor R3 when a device is under test.

Checking zener diodes

This portion of the tester consists of a similar half-wave power supply, although its voltage is about 120 volts maximum. The 125 volt secondary on transformer T2 has one end connected to chassis ground and to the negative zener test terminal. The other end goes to the cathode of diode D3 (400 PIV at 1 amp) with this diode cathode connected to the positive side of capacitor C6 (40 uF), and the C6 negative grounded. This positive terminal also goes to the top of potentiometer R5 (10K ohms, wire-wound) with the bottom of this control grounded to chassis ground. The arm of this control goes to the positive zener test terminal.

When checking zener diodes, remember that the diode cathode under test connects to the positive test terminal, with the anode grounded. Connect the zener to the test terminals and shunt your voltmeter across the diode. Set potentiometer R5 to give zero volts. Advance this control along with the voltmeter setting until you reach a voltage point at which the zener breaks down, and a further increase in the control setting produces no further voltage change. This is the voltage rating of the diode, and the control should not be advanced beyond this point, or the diode may burn out.

Remember, this instrument is capable of producing dangerously high voltages. If you charge a capacitor to some high voltage, remember that it is ever ready to deal instant death to the person who comes in contact with its terminals. Because of the 3 megohm resistor in series with the voltage source and the circuit under test, it will take a finite time for a capacitor to become fully charged. Again, remember that while it takes time for the capacitor to charge, it stores energy equal to the product of the capacitance in farads times the applied voltage.

Some readers might suggest that all of the tests outlined in this paper can be carried out with already available instruments; while this is true, I find that the convenience of such a device as this simple checker, bundled up in a single box is well worth the effort put into its construction.

One final suggestion might be a good ground connection to this gadget; better still, use a 3-wire cord with a metallic ground as I did, but which I neglected to mention in the text.

Parts List

• C1 0.1 uf., 15 KV. May be any capacitance value even as low as 0.05
• C2 1 uf., 400 VDC or higher. If the device under test is a short circuit, the voltage across the capacitor approaches 350 volts.
• C3 0.1 uf. 25 volts.
• C4 0.05 uf. 25 volts.
• C5 25 uf. 25 volts.
• C6 40 uf. 150 volts.
• D1 High voltage diode, 30K PIV. I used a replacement type silicon unit which replaces the type 1B3 television rectifier; its type number is 3A3. Also International Rectifier type V-110-HG. This type mounts in large fuse clips, the 1B3 replacement mounts in a standard octal socket. If you use a socket, it would be well to mount it on ceramic insulating pillars.
• D2 1N4001.
• D3 400 PIV, at 1 ampere; obtained from surplus.
• K1 Relay, double-pole, double-throw, 120 volts a.c.; Potter and Brumfield, KRP-11-AG. Pin connections: 2 and 7, relay coil; 1 and 8, swingers; 4 and 5, normally closed; 3 and 6, normally open.
• Q1 General-purpose NPN silicon, small signal; RCA 2N3241A used here.
• R1 60 megohms, high voltage. Special Western Electric type from surplus. Could be made up of high voltage resistors of the 1 to 10 megohm types used in TV high voltage supplies if necessary.
• R2 3 megohms, high voltage type, used in TV high voltage units.
• R3 100K ohms, 0.5 watt, 10 per cent.
• R4 47K ohms, 0.5 watt, 10 per cent.
• R5 10K ohms, wire-wound potentiometer, with knob.
• Switch- SPST toggle.
• Sockets-Two octals, one for relay, the other properly insulated for high voltage diode.
• Speaker-3.2 ohms, small, 1-1/2 inch unit used here.
• T1 6.3 volt filament transformer, 1 ampere.
• T2 6.3 volt at 1 ampere; and 125 volt at 25 Ma.
• T3 Small transistor type, 500 ohm primary, center tapped; 3.2 ohm secondary.
• Variac-O to 135 volts, and connected for 117 volts input. A 1 ampere General Radio unit used here. The knob may be fitted with a pointer and a braille scale if desired.
• Misc.- Chassis, top cover and bottom plate, binding posts, sockets, hardware, etc.

SOLDERING

by Bill Gerrey

Editor's Note -- The rumor is not true! This discussion is not intended to promote the talking book edition over braille by way of injury to the readers.

For the last three years, I have intended to write a book on soldering and the techniques used by blind technicians. There are as many different soldering systems as there are fabrication processes and materials. The compilation of such material would be a monumental undertaking.

I justify this preliminary discussion by proposing that "we have to start somewhere." Arbitrarily, that "somewhere" will be with my own experience in soldering. You can rest assured that this is just the beginning of a series of articles regarding soldering as it is being done by blind and visually impaired people. (At some point, we will be including information on soldering under a closed circuit TV system). If you don't like this article, wait for the next one.

I will not burden you with a long bibliography; there are very few discussions of soldering that are high on science and low on popular myth. One such reference book is Soldering, Its Fundamentals and Usage by Kester Solder Co., Copyright 1961.

This 84 page booklet describes the fundamental scientific principles and the practical applications in a readable style.

Solder does not just "stick" or adhere to the metals being soldered; it alloys with them. Liquid solder is actually a solvent which forms a solution with the surface metals being bonded. With the solder acting as a solvent, the surface metal need not melt for this chemical interaction to occur.

The solder of interest to us is a tin-lead alloy which melts at a much lower temperature than do tin or lead individually. If this alloy is composed of 63 per cent tin and 37 per cent lead, (said to be the "eutectic alloy"), the melting point (183oC, 361oF) is the lowest possible with any tin-lead combination. The eutectic alloy is unique in yet another way -- there is no "plastic" phase. The eutectic alloy turns directly into a liquid when brought to the eutectic temperature, and reverts back to a solid when allowed to cool below the eutectic temperature. (Eutectic solder can be purchased; it has the advantage of having the lowest melting point possible for a tin-lead composition solder. I find in using it that it is extremely sensitive to minor vibration while solidifying, and it can actually "run away" from the heated part of the work to solidify elsewhere.)

The most often used compromise solder alloy has 60 per cent tin and 40 percent lead. This composition, like any other composition that deviates from the 63/37 eutectic alloy, does not abruptly change from a solid to a liquid; it goes through a "plastic" phase which is best described as gummy. We can understand how this can happen by considering that 60/40 solder is eutectic solder with extra lead added to it. When this composition cools to about 374oF, 189oC, crystals rich in lead (about 84 per cent lead) solidify and form little "dendrites" (branch-like structures). High-lead crystals continue to form until all that is left is eutectic alloy. At 361oF, the eutectic temperature, the solder changes from the "plastic" phase (containing the lead-rich precipitant) to the solid phase. The plastic phase comes in handy, the solder doesn't run away from the connection before it cools, and minor vibration, such as the natural tremor in your hands, will not weaken the connection as often as it does with eutectic solder.

In order for "soldering" to take place, the connection must be hot! hot! hot!, about 550oF (288oC). The solder must be completely liquefied, and the alloying temperature must be reached where the solder "dissolves" or forms an alloy with the metals being bonded. The formation of this alloy on the surface of the metals is called "wetting" of the surfaces by the molten solder. The joint which has been "wetted" properly does not have globs, balls, or other rounded sculptures of solder attached to it. A properly soldered joint over which good wetting has occurred is shaped just about like it was before it was soldered; it is covered with solder, but the contours of the joint are still present. The solder will not just terminate its edges in lumps, but will nicely "feather" down to the bare metal at its extreme edges. The joint will look as if a very elastic sheet has been draped over it and been tucked in nice and cozily.

We have an understanding of the metallurgic principles of soldering, but unless the cleaning and soldering operations can be done in an atmosphere of helium, we can't just mix tin, lead, and the surface metals in the face of a blow torch and expect to get passable results. We will proceed to the discussion of "flux."

Metals that are electrically active (in the molecular sense) love to combine with oxygen to form oxides. No matter how well you clean the surface of metals such as copper, oxides will form as soon as you stop cleaning. Furthermore, the formation of oxides is tremendously enhanced with the application of heat to the metal. These oxides prevent the metallurgical process of soldering; the molten solder cannot contact the bare metal surface to alloy with it (no wetting occurs). The oxides which rapidly build up at the point of heat transfer (between the hot iron and the work) insulate the iron from the work so that poor heat transfer occurs.

There are agents which remove the oxides as the work is being heated and soldered. For those of you who are chemists, these compounds (which are called "fluxes") perform a reducing action which separates the oxides from the surface metal. The flux residue, containing the oxides, floats to the top of the boiling solder and allows the metal-to-metal alloy to form underneath.

Rosin, as used in soldering, is an inert polymer (complex chain of molecules) which resists forming chemical bonds when at temperatures below those require for soldering. When the resin flux is heated, the polymers break up, leaving active molecules to serve as the reducing agent. Upon cooling, the resin and oxide compounds are once again inert, and harmlessly remain on the surface of the joint.

Resin is about the only flux which is inert, the others are conductive corrosive nasty substances which work wonders during soldering, but which must be washed away with water or other polar solvents after use. Why have to bathe your projects? Use resin flux. There are a few other resin based fluxes whose residues are inert, but they are not as chemically sable as resin at high temperatures. I use resin flux myself.

In hand soldering operations, thin tubular solder is used, in the center of which the flux has been premeasured and stored. In electronics work which is on the scale of the projects described in this issue, solder of about 0.050 inches in diameter is suitably sized (Ersin Multi-core solder comes by the gauge number; No 20 through 22 are suitable). I prefer this size of solder because it is rigid enough to be used as a "feeler," and it is small enough in diameter to permit accurate gauging of the volume applied.

The metals being soldered must be in contact with each other, so that they will heat up together. In order for efficient soldering to be accomplished the iron should be applied to the point of maximum heat capacity, i.e., the largest piece of metal permitting the largest contact area. (All the surfaces being bonded should be in contact with the iron if possible). Solder which is applied to the iron alone will not bond to the connection; its flux will be used up cleaning the iron and not the work. Solder should be applied at the point of contact of the work and the soldering iron. Once solder has melted at this point of contact the flux will attack the oxides cooking away between the iron and the work, and at the same time, molten metal will flow between the iron and the work, causing very efficient heat transfer. After that, solder applied just about anywhere on the connection will be involved in the formation of the surface alloys, because the connection itself is hot enough to melt the solder. When the connection is hot enough to melt the solder, you are just about guaranteed to have made a "good" solder connection.

The essential properties of the soldering iron warrant discussion here. The tip of the soldering iron must be made of a metal whose thermal conductivity is high. In addition, the tip of the iron must be able to accept and to retain a surface alloy of tin or solder (known as "tinning") so as to afford the formation of a completely continuous metallic path between the iron and the work, which is necessary for efficient heat transfer. Finally, the iron must be powerful and efficient enough to heat up the localized area of interest faster than heat can be dissipated or transferred away from the iron.

Because of its high thermal conductivity, copper is widely used as a base metal in soldering tips. The use of bare copper tips is a long-standing tradition. Bare copper is highly soluble in solder, however, so that these tips actually wear away and require frequent reshaping and/or replacing. Because of the high degree of maintenance required on bare copper tips, another type of soldering iron tip has become very popular; it is relatively maintenance free. This type uses copper as a base metal, to take advantage of the thermal conductivity of copper, and has a plating or cladding of ferrous metal (iron or steel).

"Tinning" the iron specifically refers to applying a coat of solder or tin to the tip. This process retards the buildup of oxides on the exposed tip metal. It also assures that fresh solder, when applied to the point of heat transfer on the joint, will be able to wet the soldering iron tip and establish a complete continuity of metals from the iron to the work. Bare copper tips can be tinned by applying fresh solder to them on a regular basis. (Periodically, the copper tip must be filed down to a smooth new finish and tinned again.) Tips which are clad with ferrous metals are tinned by the manufacturer, because the cladding does not readily dissolve in solder. After a coating of tin is alloyed with the tip by the manufacturer, solder can alloy with the surface tin. Keeping fresh solder on the tip will help prevent oxides from separating the tin from the cladding. Once the factory tinning has become flawed, the tip must be replaced.

A common fallacy is that the iron must be small enough not to damage the work. (I made this mistake for years.) Used inappropriately, low-powered irons do more damage than soldering. Their intended application is in cases where a larger iron cannot be maneuvered into position for soldering. In general purpose soldering, low-powered irons do not heat up the localized area of the connection quickly, allowing considerable heat to be transferred to and absorbed by components of the work. In most cases, an iron of 50 watts or more can heat up the connection quickly and efficiently, allowing soldering to be accomplished without overheating the entire collection of circuit elements. The iron must be "big enough," not "small enough."

You ask, "Is he, the author, ever going to get down to business?" Yes, yes, let us descend from the ivory tower of conceptual cognition and get our fingers warm.

The remainder of "Soldering" will focus on soldering with "instant heat" soldering irons or guns, which have proven to be well suited for use by blind technicians. The next article will discuss the techniques that I use when soldering with a continuous-heat soldering iron, along with a survey of readily available tools and accessories used in soldering.

A variety of soldering irons and soldering guns are available whose features include "instant heat" capability; they warm up to soldering temperature within a few seconds of being turned on. However, only a particular type of these "instant heat" irons is of interest to us. We are interested in the irons and guns whose tips have very low mass, and consequently have low heat storage capacity. These not only heat up quickly, they cool down quickly after being turned off.

Unlike conventional soldering iron designs in which a large heating element heats the tip by thermal conduction, the irons to be described here use high current electricity to heat a small low-mass tip. The tip itself is the heating element. Within 60 seconds after the current through the tip has been turned off, the tip is cool enough to be touched.

These instant-heat, fast-cooling irons have two major disadvantages in comparison to conventional irons. The first is that the small surface area of the tip, which is the heating element, does not permit efficient heat dissipation in free air. If the current through the tip is left on while the tip is not in good thermal contact with the metals being soldered, the tip can reach an extremely high temperature which will quickly oxidize its surface and ruin the tinning job. The second disadvantage is that because very high current is necessary to heat the tip, a physically heavy high-current power source must be incorporated into the handle of the iron or gun. These disadvantages are something that can be lived with, and they are more than offset by the convenience offered to the user who wants to guide the iron into position with his fingers while the tip is cool to the touch. For the hobbyist who occasionally rolls up his sleeves and solders a project together, these irons eliminate the need for constant practice necessary to safely use a continuously hot iron. Two kinds of these "instant heat" fast-cooling irons are commercially available. Actually, they differ only in size. The first is the transformer type soldering gun, and the second is a battery operated "cordless" soldering iron.

The tip of the transformer type soldering gun is a simple elongated loop of wire. The body of the gun contains a power transformer to match the high current load of the tip to the a.c. power line. These guns are available from a couple of manufacturers; however, in my opinion, the best is the Weller Model 8200 dual-heat soldering gun (also marketed by Radio Shack as the Archer Professional Dual-Heat Gun, stock No 64-2190, for $16.00). Its bare copper tip is cheap and simple; a piece of 12 gauge can be fashioned into a tip in a pinch. The binding posts for the tip are good solid construction so that the ends of the tip can be secured firmly (these binding posts need to be tightened occasionally.) The two position trigger switch provides two power ratings, 100 and 140 Watts; one for small jobs and one for larger work. The main limitation of these guns is that the tip is too large to be used on integrated circuits and other crowded assemblies.

With the advent of sealed rechargeable batteries, a junior member of this family of instant heat irons has become available. These units are said to be "cordless." Instead of using a power transformer to supply current to the tip, a nickel-cadmium battery is incorporated into the handle. For practical reasons (not the least of which is that the high tip current must be controlled directly by a push-button switch), these cordless irons are very low power and are only good for very small work. In my opinion, the best of the cordless irons is available from Radio Shack as the Archer Cordless Iron, stock No. 64-2075, for $20.00.

They work fine for circuit board work (either printed circuits or point-to-point wiring on perforated boards), but they are not husky enough for soldering terminal lugs such as those found on plugs and jacks.

With a complement of two irons, a cordless iron and a transformer type gun, those who wish to use instant-heat soldering irons can cover the full range of electronic assembly work by using the appropriate tool.

Tinning these irons and preparing them for use is a simple matter if you carefully monitor what is happening by holding on to the solder. Spool off a couple of feet of solder and wrap about three inches of it around the tip of the iron or gun. Hold on to the solder about an inch away from the tip. Turn on the iron and wait for the solder to melt, which will disconnect the iron from the solder in your hand. Immediately and simultaneously release the button or trigger and give the iron a quick little shake in a direction away from you. After a minute, feel the tip and inspect your tinning job. The tip should feel smooth and perhaps a little gummy from the flux. If the tip feels rough in spots or if it has a glob or an "icicle" of solder that might get in your way during the first connection, repeat the process.

Some thought should be given to the preparation of the work area. Since splattering and dripping of the solder is inevitable, choose a work surface on which marring is of no concern.

Because the instant heat guns and irons are characteristically heavy, a collection of blocks, books, or heavy transformers against which your hands can be braced may help slipping of the iron off of the connection.

Your paraphernalia should include good solid holding devices which can rigidly support the work. A small table vise with a swivel is good for most applications. (More on this in Part II.) In designing the layout for your projects, try to arrange for the soldering to be done in accessible places. When stringing connecting wires around your project, make them long enough so that they can be gotten out of the way during soldering.

When fashioning the connection to be soldered, make sure that all the metals being joined are in contact with each other so that they will all heat up together. It has been said, in fact overstated that the connection should be mechanically self-supporting before soldering. This philosophy can get you into trouble; if leads are wrapped around and around terminals to make them mechanically rigid, the solder may not flow around all the surfaces leaving portions of the connection unsoldered. Maximizing the area of metal to-metal contact will minimize the susceptibility of the solder bond to shear stress, but the small weight and size of modern electronic components has outdated the wrapping practice. Some specific examples of good practice are:

When soldering a wire to a round terminal post, bend the wire three quarters of the way around the post.

When soldering a wire to a flat terminal lug which has a hole through it, pass the wire through the hole and bend it over so that it lays against the flat surface of the lug.

When components are installed on a printed circuit board, lean the wire over at an angle before soldering it. In the military code, the wire must be bent right down against the printing on the board, but this makes component replacement difficult.

When attaching components to terminals, always leave at least a distance of one-eighth inch between the terminal and the component. This practice will permit the attachment of a clip-on heat sink when appropriate, and will prevent direct heat from the hot iron on the body of the component.

Identify the item of largest heat capacity on the connection so that the iron can be put in good contact with it. In the ideal situation, the iron should be in contact with all the metals being soldered; however, this is not always possible. To be specific, when a wire is being soldered to a terminal lug, you must at least apply heat to the lug. When a comparatively heavy component lead is being soldered to a very small terminal on a socket for an IC (integrated circuit), you must at least heat the wire, because it is massive compared to the socket terminal and can absorb much more heat energy.

Using the above guide lines, clamp a project in your vise and fashion a connection to be soldered. Place your iron or gun against the work and brace your hands as necessary so that you are in a comfortable holding position. Check to see that the "barrel" of your iron does not run close to wires or components that could be damaged by the heat. With your free hand, hold the solder at a point about three-quarters of an inch back from the end, and place the end of the solder up against the connection and the tip of the iron never against the iron alone. Make sure that the three-quarter inch piece of solder between your fingers and the connection is straight, so that you know which direction to move in feeding the solder once it melts. Turn on the iron and wait for the solder to melt.

When the melting occurs, the solder will "disappear" off the end of the piece in your hand. Depending on the size of the connection, you must supply one-eighth to one-half inch of solder onto the joint. When feeding the solder after the initial melting, apply it to the connection and not to the iron. After the desired quantity of solder has been applied, leave the iron against the joint for one or two more seconds. Then, slide the iron off the connection with a smooth deliberate motion. Sliding the iron off the connection rather than jumping straight away from it will help to break the surface tension of the liquid solder without leaving sharp "icicles" in the direction of your departure.

Any of all of three indications can be monitored to affirm that good wetting and soldering has taken place. First will be if you can get the solder to melt when applied to the connection at a point that is not in direct contact with the iron; this means that the entire connection has reached soldering temperature. The second indication is that very small motion of the iron or of the components will feel "squeaky;" the flux has done its cleaning job and the solder-wet surfaces are "squeaky clean" like dishes in soapy water. (My thanks to Dennis Bernier, Vice President of Research and Development at Kester Solder Co. for explaining this effect to me.) The third sign is that heat transfer to all the component leads becomes very efficient; the temperature of the connecting leads will rise sharply.

After the connection has been allowed to cool (and don't allow components to move in relation to each other until the solder has completely solidified), a systematic inspection of the joint is in order. Five indications of a faulty joint should be looked for: If no solder has made it on to the connection, the terminals and leads will feel "rusty" i.e., they will have corroded from being heated without the presence of flux and solder to protect them.

If good wetting has occurred, the connection will not have lost its characteristic shape, but will be smoothly covered with solder. If, however, solder has melted but not wetted onto the connection, round pieces can be picked off, because they have not formed any surface alloy with the metals being joined.

Gently wiggling each of the connected components should cause no rattling or relative motion with respect to the others. If they are soldered together, the connection is one piece of metal.

If too much solder has been applied to the connection, heavy droplets may be hanging down from the underside of the joint which could short something out or break off and rattle around in the project until they bridge two other connections. Turning the connection upside down and reheating it can often take care of this problem.

Finally, it is often difficult to keep connections from "bridging," i.e., adjacent connections may become soldered together accidentally. This can happen if too much solder is applied, or if adjacent connections are inadvertently heated simultaneously by the iron. A small probe, such as a braille stylus or small screwdriver, should be passed between closely spaced terminals to check for bridges. If a bridge is found, reheat the connections separately and clear the bridge with your probe as the solder melts.

3-TERMINAL MONOLITHIC VOLTAGE REGULATORS

By Bill Gerrey

With the advent of monolithic integrated circuit voltage regulators, the design of regulated power supplies has become a breeze. With these units, you can make your own hum-free battery eliminators, logic power supplies, 12 to 16 volt converters, etc. No more ripple or saggy voltage sources need be tolerated.

Equivalent devices are available from several manufacturers, and a summary of specifications is given here. If a specific parameter is needed in detail, consult the manufacturer's data sheet.

The Fairchild and Motorola numbering systems are similar. (The Fairchild prefix is uA, and the Motorola prefix is MC). The 7800 series is for positive voltages, and the 7900 series is used in negative voltage supplies. For the 7800 series, 5, 6, 8, 12, 15, and 24 volt units have the designation numbers 7805, 7806, 7808, 7812, 7815, 7818, and 7824, respectively. In like manner, a 7915 will deliver minus 15 volts. Those units with a C suffix are "commercial grade;" they are limited to operating temperatures from 0 to 150 degrees C.

National Semiconductor designates their positive regulators as LM340, and their negative ones as LM320. Two grades are available from National. Using the TO220 package as an example, LM340T units have somewhat relaxed specifications as compared to LM340AT and 7800 units. Finally, the voltage of National's regulators is tacked on the end of the number, LM340T-5 is a 5 volt regulator.

Two case styles are available, TO3 and TO220. The TO220 package is easily mounted with a single No. 6 screw, or it can be soldered into the circuit and left to hang in free space. The TO3 style, on the other hand, is a diamond shaped package requiring two mounting holes and two clearance holes for the input and output terminals.

These devices have short circuit current limiting (less than 1 amp), and they have thermal shut down. They will deliver up to 2 amps peak and 1 amp continuous (if mounted to the chassis.) The input voltage can be a maximum of 35 volts (40 volts for the 7824.) In general, the output voltage will be regulated to within 5 per cent of the voltage specified. The ripple rejection is better than 50 dB, and can be as much as 70 dB for some. The quiescent current drain is typically less than 5 mA.

Remember, you can't get out what you don't put in. The input voltage must be 2 volts higher than the output voltage for the regulator to operate. The power transformer should be chosen so as to allow a comfortable margin over this "drop out" voltage of 2 volts. For a 6 volt regulator, the transformer should deliver more than 7 volts and not higher than 24 volts RMS. The input to the regulator will be about 1.4 times the RMS output of the transformer, i.e., 7 times 1.4 equals 9.8 volts. The drop out voltage for a 6 volt regulator is 6 plus 2 equals 8 volts. Full- wave rectification is recommended, so that the input filter capacitor can keep the input above this drop out point.

With a circuit connection known as "boot strapping," output voltages other than those specifically available can be obtained. Since these devices are series-pass regulators, the common terminal can be lifted off ground and connected to a voltage divider from the output to ground. Of course, regulation will suffer, since the output will be dependent in some ratio to the input voltage. In calculating the resistances in the voltage divider, you cannot neglect the quiescent current drawn by the regulator itself (about 5 mA.) Since this quiescent current varies considerably from unit to unit, I choose the resistances experimentally. A good starting rule of thumb is to assure that the current in the top resistor is more than 3 times the quiescent current, i.e., the regulator voltage divided by the top resistance should be greater than 15 mA. The bottom resistor can then be chosen experimentally.

Simple Circuit for a 6 volt Power Supply

The primary winding of a 117 to 7 volt filament transformer (good for 1 amp) is connected to the a.c. line. Each end of the secondary goes to the cathode of a rectifier diode (such as 1N4003), with the anodes of these two diodes going to the common terminal of the regulator (7806). Each end of the secondary also goes to the anode of a rectifier diode, with the cathodes of these two diodes going to the input terminal of the regulator. An electrolytic filter capacitor of perhaps 500 to 1000uF has its negative end connected to common, and its positive end going to the input terminal. The output terminal is bypassed to common by 0.01uF to avoid oscillations in the regulator chip. Output is taken between the output terminal and common terminal of the regulator.

Pin Connections

TO220 Package

hold the package with the leads pointing upward and with the mounting surface toward you.

For the 7800 series (LM340), the three leads from left to right are input, common, output.

For the 7900 series (LM320), the three leads from left to right are common, input, output. Caution! The case is connected to the input, and not to common, as with the 7800's.

TO3 Package

hold the package with the mounting holes at the extreme left and the extreme right, and with the terminal pins d6 2 Scloser to the left hand mounting hole (bottom view, pins facing you).

For the 7800 series, the top pin is output, the bottom pin is input, and the case is common.

For the 7900 series, the top pin is output, the bottom pin is common, and the case is input.

Also, on the sound advice of Bob Gunderson, I shall describe the connection of a volume control to the LM386.

The bottom of a 10K audio-taper pot goes to Pins 2 and 4 of the LM386 and to ground. The arm of the pot can go directly to Pin 3, with the top of the pot going through a coupling capacitor (0.1uF or larger) to the signal source. (Sorry, Professor Gunderson, for this oversight. Can I at least have a B in the course?)

THE LM 386 HIGH-GAIN AUDIO AMPLIFIER

For those of us who remember the "good ol' days," an audio amplifier of this capability took at least an entire day of punching holes in a metal chassis and wiring the sockets of two or three vacuum tubes. With this integrated circuit (IC), even someone as slow as your editor can design the layout and wire up this equivalent amplifier in an hour.

The National Semiconductor LM386 amplifier chip is used extensively here at Smith-Kettlewell Institute in our auditory test equipment designs. I use this circuit as a general purpose test amplifier in my lab. This amplifier would make an excellent intercom, etc.

The LM386 comes in an 8 pin dual in-line package. A very small indentation on the corner of the top marks pin 1. It will operate on supply voltages from 4 to 12 volts. At 6 volts, the quiescent current drain is less than 8 mA. Also, at 6 volts this little device will deliver a distortion free signal (0.2 per cent total harmonic distortion) of 125 mW. The voltage gain is internally set to 20, but external positive feedback can be connected between pins 1 and 8 to achieve a voltage gain as high as 200. This means that with pins 1 and 8 left open, 50 mV will drive the amplifier at full undistorted power (125 mW.) With the addition of a capacitor connected between pins 1 and 8, a 5 mV input signal will give you full power. The addition of 1.2 kohms in series with the capacitor will give you a gain of 50, requiring a 20mV signal to drive the amplifier to full undistorted output. The input impedance is 50k ohms and the above power ratings are based on an output load of 8 ohms.

Like most high-gain circuits, this amplifier tends to oscillate, thus causing raucous distortion in the audio output. The manufacturer suggests putting the series combination of 10 ohms and .05 uF across the output to eliminate this problem. This solution never worked for me, but connecting a .22 uF disc capacitor with very short leads between pins 5 and 4 (the output and ground, respectively) works every time.

Circuit

Pin 4, the negative supply terminal, goes to the negative end of the battery and to ground. Pin 2, the inverting input (this is actually an operational amplifier), is also grounded. Pin 6, the positive supply terminal, goes through a switch to the positive side of the battery. Pin 7, used for decoupling the input stages, goes through a 25 uF electrolytic capacitor to ground (positive side toward pin 7). Pin 5, the output, goes through a 250 uF capacitor to one side of the loudspeaker (positive end of this capacitor going to pin 5). The other side of the speaker is grounded. Pin 5 also goes through .22 uF to pin 4 and to ground. Pin 3, the non-inverting input terminal, goes through 0.1 uF to the signal source or to an input jack.

If a voltage gain higher than 20 is desired, connect pin 1 through a 10 uF electrolytic capacitor in series with a resistor to pin 8 (positive side toward pin 1). If the capacitor is connected directly between pins 1 and 8, the highest possible gain of 200 will be the result.

POINT-TO-POINT WIRING ON PERFORATED BOARD

The obvious economic advantages of printed circuit technology over older hand wiring schemes have led to a universal change over on the part of industry. Virtually all components are now available in a form which allows them to be "PC" mounted (on a printed circuit board.) In some cases, as with small potentiometers, PC-mount units are the only style readily available. Hand wiring has become obsolete because of the change in style of electronic components. How does the hobbyist and experimenter cope with this change over? Must he laboriously design a printed circuit for every one-of-a-kind project? This would necessitate all of us to become draftsmen. What is an economic boon to industry in a large scale production scheme would slow down the experimenter immensely.

Naturally, since all development work requires a system of prototyping the designs, various schemes have been devised for building "one-off" circuit boards without having to commit these prototype designs to the rigidity of a printed board.

This article will discuss the process of "point-to-point" wiring as a method of building a circuit on a board without much preparation. I propose to illustrate this process by example; I will describe what I would do in building an audio amplifier using a National Semiconductor LM386, powered using a Fairchild 7808 3-terminal monolithic voltage regulator.

In a future issue, a rundown on products of the Vector Electronic Company will be printed, in which various push-in terminals and other prototyping accessories will be described. This project will use nothing but hookup wire and a perforated board. The term "pint-to-point" wiring specifically refers to connecting the components together without terminals, and doing so almost exclusively using the wire leads on the components themselves. (I cheat, I run a bare wire along opposing edges of the board for supplying power to the circuit.)

The terminals on dual in-line integrated circuits (such as the LM386) are spaced at 1/10 inch intervals. Perforated boards are available with various hole patterns and spacings, but we want the board with holes spaced at 1/10 inch centers. In general, larger pieces of board come cheaper. Radio Shack sells a fair sized piece of this board which is 6 x 8 inches, and which costs about $2.20. It is stock number 276 1396. I usually buy a piece of Vector board which is 4-3/4 by 17 inches (Vector catalog number 169P47EP.) To get the board into smaller pieces, you can either use a coping saw to saw along a row of holes, or the board can be broken off along a string of holes. My favorite size of circuit board is 1-5/8 inches wide and 4-3/4 inches long. (This 4-3/4 inch length comes from the fact that the board I buy starts out being a very long piece that is 4-3/4 inches wide. Actually, one of the advantages of point-to-point wiring is that you don't have to care how long the board is, just knock off the extra length when your project is finished.

To break the board off at a string of holes, clamp the board in a vise with the holes of interest just peeking up above the jaws (in this case, the top edge of the jaws will bisect the 16th row of holes, making the width of the broken-off piece to be slightly over 1-1/2 inches).

You can position a desired row of holes along the top of the vise by feeling them with the point of a braille stylus or small probe. After the board has been positioned and clamped firmly, use a scribe or knife point to scribe a line across the holes, using the top of the vise as a guide. Scribe this line on both sides of the board. Then, take the board in both hands with your fingers very close to the top of the vise and bend the board slowly until it breaks off.

If you broke the board on the 16th row of holes, you have a piece of board that is several inches long and slightly wider than 1-1/2 inches. Insert the ends of a piece of bare hookup wire (18 or 20 gauge) into the corner holes so that the wire runs along one of the long edges of the board. Do the same on the opposite edge, making the board look like a short piece of railroad track with a rail along each long edge of the board. These rails are commonly called "bus bars" and will be used to distribute power to the amplifier circuit. These bus bars run along the bottom of the board; the components are inserted from the top side so that their leads protrude through the bottom side on which the bus bars run.

Facing the bottom side of the board, we will arbitrarily call the bottom bus bar ground and minus 8V. The top bus bar will become plus 8V.

In order to shortcut the discussion of the components' position on the board, let's adopt the Cartesian coordinate system and number the holes. (For those of you who are just reading this discussion to learn techniques and are not actually building the amplifier, relax and read on. I promise not to over-do it with the numbers).

In numbering the holes, we will list the number away from the left end first, and the number up from the bottom second. IMPORTANT -- these hole positions are given with component side down.

The bottom left hole will be (1,1), If we count 10 to the right and 7 up from the bottom, we arrive at number (10,7). Leaving 1/2 inch of room at the left end so the board can be mounted later on, we will build the power supply on the left end. The 4 diodes of the bridge rectifier will be arranged in a diamond pattern as follows:

• The anodes of two of the diodes (1N4005) will be inserted into the 6th column near the vertical center of the board (6,7) and 6,8). The cathodes will be inserted near the bottom and top edges of the board in column 10, (10,4) and (10,11). Pull the leads through as far as they will go, causing the diodes to lay flat on the top side of the board. At this point, bend the anodes straight down and solder them to the minus 8V bus bar.
• Insert the anodes of two more diodes in the 11th column just to the right of the first two diode's cathodes, (11,4) and 11,11). The cathodes of the latter diodes are inserted next to each other in column 15, (15,7) and 15,8). Pull the diodes up against the board and bend the pair of cathodes straight down; this time, cut them off just before they touch the minus 8V bus bar. We have a cathode and an anode in row 4, and another of each in row 11. In each case, cut these leads so that they protrude 1/8 inch through the board, and bend them down against the board so that they cross each other. These are the inputs of the bridge rectifier and will be soldered to the ends of the secondary on the power transformer (12V filament transformer). Do this after the rest of the amplifier is built, so that your board won't be tethered to a heavy anchor while being wired.

Insert the 7808 voltage regulator with its input lead in column 16 next to the pair of cathodes (16,3). The three legs of the voltage regulator can go in columns 16, 17, and 18, just above the bus bar in row 3. Bend the input leads of the regulator over against the two cathodes and solder it to both of them. Bend the common lead of the regulator (17,3) down against the minus 8V bus bar and solder it there. The installation of two more components will finish the power supply. An electrolytic capacitor (250uF with this low current drain) has its negative end inserted into (5,2), just to the left of the anodes and their junction with the minus 8V bus bar. This lead is then bent down against the bus bar and soldered. The positive lead can go into hole (14,3). It can be bent over against the two diode cathodes and soldered. A 0.01 uF disc ceramic capacitor will be added so that it shunts the output of the regulator. In column 19, insert this capacitor to the right of the regulator with its leads in (19,2) and (19,3). As you would expect, (19,2) gets bent down and grounded to the bus bar, (19,3) goes to the output of the regulator which was in (18,3). A jumper of insulated hookup wire is inserted from the component side of the board and connects the output of the regulator to the top or plus 8V bus bar.

No complicated graphical analysis leads me to choose the stunning array of hole numbers listed in the above example; I had the parts in my hand and I just stuck them in different places until I was satisfied. Don't think about numbers, just plug the stuff in and solder it together.

We can now wire up the amplifier in the same fashion. Plug in an 8 pin socket for the LM386 into the board with pins 1 and 8 in row 12 near the top of the board. (These IC's are numbered counting the pins in a clockwise direction. Therefore, pin 1 is the top pin of the right-hand column of four, and pin 8 is the top pin of the left-hand column of four.)

We will put pin 1 in (30,12) and pin 8 in (27,12). Bend the pins down against the board to hold the socket up against the component side. If you want more room for soldering, bend pins 2 and 3, and 6 and 7 in toward the center line of the socket, and bend pins 1,4,5, and 8 outward away from the socket's center line.) Jumpers of insulated wire can be put on the component side of the board and bent down against the socket terminals to connect pin 4 to the minus 8V line and pin 6 to the plus 8V line. A little hair-pin loop can be plugged into the 31st column and bent down against the socket terminals to jumper pins 2 and 4 together.

An electrolytic capacitor (25 uF) has its negative end plugged in down by the ground bar, (25,2). Run the positive end of the capacitor through some spaghetti tubing and plug it into the hole adjacent to pin 7, (26,11), bend it over against the pin and solder it. A 0.22 uF capacitor is inserted immediately below the socket and has its leads bent down against pins 4 and 5. A 10 uF electrolytic capacitor has its positive end going through a hole nearby pin 1 to which it is soldered. The negative end of the capacitor goes through a hole nearby pin 8 to which it is connected.

The 250 uF electrolytic output capacitor comes in a variety of shapes and sizes; use your own judgment in mounting it somewhere to the left of the IC, that is, between the amplifier and the power supply circuit. If long leads must be used, run them through pieces of spaghetti tubing to prevent them from shorting out to other items in the circuit. The positive end of this 250 uF capacitor goes to pin 5, and the negative end goes through the speaker voice coil to ground. Finally, a 0.1 uF disc or tubular is located to the right of the IC, with one end of it being soldered to pin 3, the input, and the other end bent to the right so that it runs off the end of the board.

If this happens to be your first circuit board project, plug in the power transformer, connect a microphone from the far end of the input capacitor to ground, and sing praises to yourself through a large loudspeaker.

Editor's Corner

At last! Technical material in a readable form is at hand!

We plan to order from the various printers based on the orders we get from you, so please respond as soon as you can if you are interested in receiving the next four issues. To save time in preparing your request, we suggest that you send us your particulars (in either braille or typed form) in the following format.

1. Name, call sign (if any).
2. Address, complete with zip code.
3. Braille, large print, or talking book (flexible disc).
4. Comments and suggestions.

Example:

1. Bill Gerrey, WA6NPC
2. 18962 Sandy Road, Castro Valley, CA 94546
3. Braille
4. How come my copy is all wet and stepped on?

We would appreciate everyone responding to this simple "questionnaire" because of the initial confusion in setting up the mailing list. Any redundancy will save us considerable work in cross referencing names and call letters.

For those of you who wish to subscribe, the fee, which will cover the next four issues is $12 for braille, $12 for large print, and $6 for the flexible disc talking book edition. checks and money orders should be made out to Smith-Kettlewell Institute of Visual Sciences, 2232 Webster Street, San Francisco, CA 94115. Please hurry, we must base our order from the printers on the interest shown within two months after the initial mailing of this issue.

Would those of you who use large print, please advise us as to the readability and convenience of this issue. Since we are generating "camera ready" copy on an IBM Selectric typewriter, we can be flexible as to print size, line spacing, format, etc.

We hope to print a preliminary list of text materials in the next issue form which the beginners amongst us can learn basic theory of electronics. If you have used any such material and would like to recommend it, please pass your suggestions along to us.

Look forward next time to some information on solderless protoboards. Until then, 73's and good experimenting.

THE END