SKTF -- Spring 1986

The Smith-Kettlewell Technical File

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

William Gerrey, Editor

Issue: SKTF -- Spring 1986

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 Details regarding products, suppliers, and other contact information are original and may be outdated.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

From Paper to Project, Part IV

The Smith-Kettlewell Timing-Tones Generator

An Audible Multi-Microfarad Meter

The Tweedle Dump

The "Kings Central Catalogue" of Books

FROM PAPER TO PROJECT Part IV

Abstract

This installment describes two major pieces of test equipment, a resistance bridge and a capacitance bridge. Obviously, these can be used to test resistors and capacitors. They have a more important use for the blind technician, however; with these instruments, such components can be
identified, thus avoiding the reading of color-coded and printed labels. Given the background of the last installment, it is possible to cover both devices here, since their solid-state circuitry is identically that of the amplifier and oscillator presented in Part III.

Introduction

[Note: These bridges have been in SKTF before--"The Simplest Auditory Capacitance Bridge" by Albert Yeo (Winter 1984), and "The Simplest Auditory Resistance Bridge," merely copied from Yeo's work by the Editor (Spring 1984).]

Well, have you forgiven me for the last one, Part III? I beg you to give me another chance; completion of these animals only involves gross manipulations and some mechanical work on the boxes. You will be loafing, compared to soldering chip sockets.

In order for this "easy work" to begin, however, you have to have built two spare circuit boards containing the amplifier and
oscillator of that famous Part III. What's more, they should be tested, troubleshot, and working before you go on. For these boards you do not need: the input and output miniphone jacks of Part III's completed instrument, SPST on-off switches (you will see why later), and the 10uF coupling capacitor at the output of the oscillator (although if this has been installed, you can leave it there; it won't hurt anything). What you should have on the board are: a speaker connected via hookup wire, and the arm and bottom terminals of the volume control connected with hookup wire leads.

Procure some alligator clip cords with which you can temporarily connect a battery to the boards. On the battery, the large snap is its negative terminal; this should be clipped to the ground bus. The small snap should be clipped to the VCC bus. Use your nicely packaged amplifier/oscillator: listen to the boards' oscillators (via a 1 meg attenuation resistor), and drive the boards' amplifiers (via a 1 meg attenuation resistor). Or, you can follow the procedures of Part III whereby the companion circuits on the boards can be used to test each other.

Know Your Parts

[Note: There are several components that are common to both instruments. Since it is assumed that you will build both--although, if you have a commercial multimeter, you will not need the resistance bridge--the quantity
of parts common to both is doubled (two 20K trim pots, four binding posts, etc.). The only parts not common to both are the "standards" (precision resistors and capacitors). It is also assumed that you built two extra circuit boards already; those parts will not be re-listed here. If you have not built two spare boards, go back to Part III and order two of everything--excluding obvious items such as the cabinet (which, for these instruments, will be larger) and the jacks.]

Parts Common to Both Bridges

  • 2--39K 1/4-watt 5% resistors
  • 2--2eK 15, or 20-turn PC-mount trim pots, Jameco 43P-20K, Mouser 3NA402
  • 2--DPST or DPDT toggle switches
  • 2--single- or double-pole 6-position switches, Jameco MRS26 e, Mouser 10XY206
  • 2--precision linear pots, Clarostat 58C1-10K
  • 2--Braille dials for the above precision pots (have been enclosed in the magazine for your convenience)
  • 2--long pointer knobs for the above Braille scale, G.C. Electronics 37-582
  • 4--short pointer knobs for "range" and "volume" controls
  • 4--binding posts; may want two red and two black, Radio Shack 274-662
  • 2--plastic cabinets; 7-3/4 by 4-1/4 by 2-3/8, Radio Shack 270-232

Since Mouser- Electronics figures elsewhere in this parts list, you should know of my favorite items:

  • Square 1-1/2-inch speakers with mounting holes--Mouser 25SP016
  • 9-volt battery connectors with 6-inch leads--Mouser 12BC095

Capacitance Standards

[Note: It's hard to find "precision capacitors." However, there are ones around in the "fairly good" class. The "dipped silver-mica" units listed here will probably, be within 5% of their rated value. Garden-variety Mylar units are usually good to a tolerance of plus/minus 10%. Fortune smiles upon those who can order from Mouser, since the Mylar units listed here are good to within 1%. The tantalum electrolytic will come out to be "heaven-knows-what." One could solve these problems by getting a hatful of each, then taking them over to the nearest electronics school to be tested on a visual bridge. However, since the main use for this instrument is identifying capacitors of less quality than those listed here, I recommend just installing a set and hoping for the best. You will want an extra 0.01uF silver-mica unit to calibrate your instrument; this explains why there are two of them listed.]

  • 1--100pF silver-mica, Mouser ME232-1500-100
  • 1--1000pF (0.001uF) silver-mica, Mouser ME232-1500-1000 or Mylar, Mouser 23PF210
  • 2--0.01uF Mylar, Mouser 23PF310
  • 1--0.1uF Mylar, Mouser 23PF410
  • 1--1uF 10V tantalum urn, Mouser 551-1.aM25
  • 1--10uF 10V tantalum, Mouser 551-10M25

Resistance Standards

[Note: These are surprisingly common (although not at Radio Shack). Mouser numbers are listed here--they're only $.08 each. This means that, unless you avoid it by buying $20 worth of stuff, they'll get you for a $5 "small-order" charge. You could order lots of these resistors; they figure heavily in other projects (the "Multi-Farad Meter" of this issue, for example). Among the many things they have that I like, I really make use of their small loudspeakers and battery clips also listed here.]

  • 1--100 ohm 1%, Mouser 29MF250-100
  • 1--1K 1%, Mouser 29MF250-1.00K
  • 1--10K 1%, Mouser 29MF250-10.0K
  • 2--100K 1%, Mouser 29MF250-100K (the extra for calibration purposes)
  • 1--1 megohm 1%, Mouser 29MF250-1.00M
  • 1--10 megohm 1%, Mouser 29MF250-10.0M

Experimenting with the Parts--Purposefully mix up the piles of resistors. Now, sort them out with your continuity tester. The lowest resistance will give you the highest pitch. The 10 meg unit will give you such a low "pitch" as to merely "tick" like a clock. If you included the two 39K resistors in the jumble, you will know them by the fact that there are two similar pitches (they may differ slightly, since the 39K units have a 5% tolerance). Now that you have them sorted, you can jab them into a block of Styrofoam, or arrange them in order in a muffin tin.

Until you get this bridge built, sorting the capacitors will have to be done by size and shape. If necessary, get sighted help, knowing that this method will soon become obsolete for reading capacitors.

The high-value-units can be found with your continuity tester; the lug unit will make it emit a quick chirp, and the 10uF unit will bring forth a descending tone. Also, one lead of these may be longer than the other (the long one being positive). Within any other type--Mylar or silver-mica-the size will increase as the capacitance goes up (assuming the compared units are of the same "working voltage").

If you stray between types, you will notice wide variations in size. For example, your very low capacitance silver-mica unit may be as large as one of the high-capacitance Mylar ones. The leads of your mica ones will probably be farther apart; mica cannot be rolled up as can Mylar, so these are made in flat assemblies with leads corning off the corners.

The switches are the next thing to ferret out. The term "pole," as often used when describing switches, refers to an isolated switch within the unit. A switch may have two or more "poles"; for example, the on-off toggles of this project need to be "double-pole" (DP) because two separate circuits are being switched. With a double-pole switch, you could turn on a light that runs off 110V with one pole, and ring the doorbell (which runs off 10V) with the other pole; throwing the switch would do both.

The term "throw" is synonymous with "position." Thus, it is correct to call your 6-position switch a 6-throw switch. The double-pole on-off switch you buy may be of "single-throw" (ST) or "double-throw" (DT).
Both poles need only be of "single-throw," since all we are doing is closing (turning on) and opening (turning off) two circuits; this would be a double-pole single-throw (DPST) switch. DPDT (double pole double-
throw) switches may be all you can get, since producers have recognized that you can always simply not use the contact of the unwanted position. (With a DPDT switch, you could ring the doorbell in, one position and turn on the light in the other, just by choosing the appropriate contacts of the appropriate pole.)

[There is another common-variety DPDT switch that has three distinct. positions--one being a "center off." Sometimes these are described as being "on-off-on." Their use would be, for example, to allow you to turn on your light, ring your doorbell, or do
neither (in the "off" position). The true 2-position DPDT switch (which is the kind required for this project) would give you no option of doing neither; you get the doorbell or you get the light.]

The 6-position switch may come in any one of several varieties. The 2-pole unit listed here was chosen mainly for its ready availability from Jameco. (I have a tricky embellishment to accomplish with the second pole, although this is not a necessity.) Whatever kind you get, you should cultivate the habit of ferreting it out with your continuity tester.

Sometimes in looking at one, a rotary switch will make physical sense. The "swinger" or "arm"--the mechanical finger that contacts the "position terminals" as it wipes across them--has a solder lug dedicated to it; this could be anywhere in the mass of terminals, but it is sometimes off by itself.

In the case of the MRS260 listed here, the "swinger" terminals for the two poles are inside the circle of other lugs, where they can be easily spotted. Sometimes, the position number progresses as you move farther away from the swinger terminal, with position 1 being closest to it. Surprise! On the MRS260, the swinger resides between positions 3 and 4. (I have seen a 2-pole switch whose swingers were assigned to the lugs on the opposite edge of the wafer; the positions near me were addressed by the swinger terminal that was farthest away.) The lug for the swinger terminal may not be "off by itself"; it may be in the main circle with the rest of the guys. Where this is the
case, gentle examination of the tiny contact fingers may reveal one which is ever-so-slightly longer. This is probably the swinger, and it very likely resides near position 1.

Put a knob on your switch and do a little detective work to find these things out before wiring up the switch. Once you have found two lugs that make the tester beep, change positions with the knob, fishing around with the tester until you have picked up the trend. Then, if you want to, mark position 1 with a small loop of bare wire, or carefully thread a small taper clip through it.

A popular arrangement is to sell switches whose number of positions can be altered mechanically. Centralab, for example, makes so-called "2- to 11-position universal switches"; these are an old favorite of mine. You can add more poles to them as well, just by stacking wafers. The Centralab units require that you take them apart, then break off little tabs which protrude from a "stop plate." Other makes sometimes have "stop fingers" which are secured in place by the same mounting nut that you use for installing the switches.

The screwdriver-adjustable trimmer potentiometer is another item to be explored. The type recommended here has an adjustment screw at one end of a plastic capsule with PC-board pins coming out the bottom. The screw operates a tiny leadscrew inside the unit which moves a threaded wiper assembly from end to end; the wiper (or arm) traces along a resistance element. Depending on the pitch of the leadscrew, 10 to 25 turns may be required to carry the wiper from one end to the other; the number of turns is specified when you buy one. The more turns required, the better--it makes adjustment easier.

Most often, the wiper is the middle terminal, with the ends of the resistance
element being at the ends of the package. Just to be sure, though, find the ones which do not affect the pitch of your continuity tester as the adjustment screw is manipulated. (Note that if the screw has been turned too far in one direction, it will take a few turns of the adjustment screw before the wiper assembly is picked up again by the leadscrew; no damage will be done, however.)

The last items to be examined are the binding posts. The ones recommended here are
of the so-called "5-way" type. What are the five ways? Beats us. A three-day engineering seminar was held at Smith-Kettlewell, and
the following tentative conclusion was reached:

The top portion has a plastic sleeve which can be unscrewed part way to expose a screw underneath. First, a wire lead can be wrapped around the screw, then secured by tightening down on the plastic knob. This screw also has a cross-wise hole drilled through it; this hole can accept a test probe or a piece of solid wire; the knob bears down on what's in the hole as well. The top of the screw is hollow, so as to accept a banana plug. You invent the rest of the five ways.

Made to be insulated from metal panels, the mounting bushing is a threaded plastic portion that carries a mounting nut. At the very bottom is a solder post; after securing it in the panel, connection to the circuit is made by wrapping a wire around the small post and soldering there.

If binding posts are mounted in holes whose centers are exactly 0.75 inches apart (which can only be obtained by skilled use of the AFB "Rotomatic Rule" or their "Click Rule"), standard 2-prong banana fixtures can be plugged in. However, since most new capacitors have fairly long leads, and since many tubular capacitor bodies are too long to afford comfortable use of this 0.75 inch spacing, you may wish to forego the luxury of providing for these banana assemblies.

Bridge Circuits

The Wheatstone bridge, and the principles that make it interesting, is standard subject
matter in any electronics course. I am leaving it up to you, if your curiosity burns brightly, to investigate it on your own. If you wish, get a book from Recordings for the Blind or through one of the private libraries that store school materials (contact the National Library Service for lists of appropriate transcribing outfits who may already have done such work). In the meantime, I'll tell you a secret that is rarely given away by electronics teachers--you can picture the layout of parts and build circuits without understanding everything about them. Thus, you have my permission to carry on.

Suffice it to say that the "Wheatstone Bridge" part consists of two branches, in parallel, being driven by the oscillator. One of these branches is the series connection of two rheostats--one adjusted by a screwdriver, and the other graduated in Braille. The other branch is the series combination of a selectable standard (the standards mounted on your 6-position switch), and the "unknown" (a test part being put across the binding posts). The amplifier then listens to the signal gotten from the "bridge centers"--the junctions in the middle of these series branches. Savvy? No? Well, carry on anyway.

I am taking the trouble to present the whole circuit, even restating that of the boards. Minor modifications will have to be made to your boards; as you read the circuit, try to anticipate what recommendations I'll make later in the step-by-step section.

Oscillator Circuit

Pin 1 of a 555 goes to the negative side of its own 9-volt battery. The positive of the battery goes through one pole of the DPST on-off switch to the 555's VCC line. Pins 4 and 8 of the 555 are tied together and go to this VCC line. A 0.1uF bypass capacitor goes between pins 1 and 8 (located close to the chip).

Also on the 555, pins 2 and 6 are tied together and go through 0.01uF to the negative side of this supply. Pins 2 and 6 also go through 100K to the 555's output, pin 3. Pin 3 goes through 100 ohms to what we will call "Point 0," the hot output of the oscillator system.

Amplifier Circuit

Pins 2 and 4 of an LM386 are tied together and go to the negative side of another 9-volt battery. The positive side of this battery goes through the other pole on the on-off switch to the amplifier's VCC line. This VCC line is bypassed by 220uF (negative of the capacitor at battery negative). Pin 6 of the LM386 goes to this VCC line. Pin 7 is bypassed by 22uF (negative end of the capacitor at battery negative).

Pin 8 goes to the negative of a 10uF capacitor; the positive end of this cap goes through 470 ohms to pin 1. Because I have an extra pole on my 6-position switch, I connected the swinger of the free pole to the junction of this capacitor and resistor; positions 5 and 6 are tied together and go to pin 1 of the 386. (When you select either of these positions, the gain of the amplifier goes up to make the measurement of high resistance and high reactance a little easier to hear.)

Pin 5 of the 386 is bypassed to pin 4 by 0.1uF. Pin 5 also goes to the positive end of a 100uF capacitor; the negative end of this cap goes through the speaker to the negative of the 386's battery.

Pin 3 of the 386 goes through 0.1uF to the arm of the 10K volume control. The bottom of this control is grounded, while the top end goes through 39K to what we shall call "Point A," the hot input of the amplifier system.

Resistance Bridge Circuit

Point 0 goes to one end of the 20K trimmer pot (it doesn't much matter which end). The arm of this trimmer goes to the arm of the 10K precision Braille-calibrated pot. The bottom of this pot goes to the 555's battery negative.

Point 0 also goes to the arm of the 6-position range switch. Each position goes to the end of its assigned precision "standard" resistor; the far ends of all the standard resistors are tied together and go to the positive (red) binding post. The negative binding post (black) goes to the 555's battery negative.

As stated, each switch position goes through a resistor to the positive binding post as follows:

  • Position 1--100 ohms
  • Position 2--1K
  • Position 3--10K
  • Position 4--100K
  • Position 5--1 meg
  • Position 6--10 meg

Point A, the amplifier's hot input, goes to the positive binding post. The negative side of the amplifier's battery goes to the arm of the 20K trimmer.

Capacitance Bridge Circuit

Point 0 goes to the bottom of the Braille-calibrated pot. The arm of this pot goes to the arm of the 20K trimmer} pot. One end of this trimmer (it doesn't matter which) goes to the 555's battery negative.

Point 0 also goes to the arm, of the 6-position range switch. Each position goes to the end of its assigned "standard" capacitor; the far ends of all these capacitors are tied together and go to the positive binding post (red). The negative binding post (black) goes to the 555's battery negative.

As stated, each switch position goes through a capacitor to the positive binding post as follows:

  • Position 1--the positive end of 10uF
  • Position 2--the positive end of lug
  • Position 3--0.1uF
  • Position 4--0.01uF
  • Position 5--0.001uF (1000pF)
  • Position 6--100pFt

Point A, the amplifier's hot input, goes to the positive binding post. The negative side of the amplifier's battery goes to the arm of the 20K trimmer.

Step-by-Step Construction

Modifying the Boards

Immediately from the circuit, you can see that something must be done to create two power supply systems. The answer lies in use of your wire cutters; cut a section out of each bus wire on the oscillator side of the 220uF bypass capacitor. Make the first cuts right up against the connections of the bypass (and do so on the oscillator's side); then make the other cuts so as to create 1/4-inch gaps in the bus lines.

The negative ends of two battery clips can then be put through holes from the component side and soldered to their respective "battery negative" buses. Just where--along the bottom edge--you poke these leads through makes very little difference. Attach 6- or 8-inch stranded-wire leads to the upper segments in the same way. After stripping and tinning the far ends of these "plus-supply wires," you are able to install the on-off switch.

With a continuity tester, find out which pairs of terminals "turn on" when the switch is thrown in one direction. The oscillator's plus lead and the red lead of its battery clip are turned on by one pair of switch contacts (termed a "pole," as you remember); the amplifier's plus lead and its battery's red lead are operated by the other half of the switch.

Do try to remember how to tell the "red" from the "black" lead on the battery connector. My memory trick is to note that the small button on a flashlight cell is its positive; coincidentally, the small snap of a 9-volt battery is also positive. Therefore, on the mating connectors you are installing, these are reversed; the red lead will be the one which shows continuity to the large snap.

Two additional components are now installed on the board. The first of these is a 39K resistor, the second is the 20K trim post.

The wire off the top end of the volume control, instead of going to an input jack, now goes to one end of the 39K resistor. This resistor can go in the very edge holes at the amplifier's input end; center it between the two mounting holes. Insert the lead from the volume control near one of these leads, bend them over so as to cross on the wiring side, and solder them--cutting off excess when you're done.

The trimmer can go right up against the 220uF bypass, on the oscillator side of it. There exists the possibility that, if you installed it, the 10uF oscillator output capacitor may be physically in the way of your trimmer. If so, heat its. connections and pull it out.

On my boards, I pointed the adjustment screw toward the negative buses, just now declaring that this would be the edge I see when the cover plate of the cabinet is removed. More than once, I have embarrassed myself by facing a trimmer in an inaccessible direction. If this happens to you, calibrate the thing with the board unmounted, then bury it for the rest of time.

When mounting components such as trimmers, first make sure that the pins
are straight, and then take careful note
of their pattern. When they go into the board, they should come through in the
same configuration. If they do not, pressing the component down on the board
will put stress on any terminal which is out of place; this can lead to breakage.

Once through, bend the terminals out to the
sides, making. sure that they touch nothing else, however. At this time, a long jumper wire can connect the free end of the 39K resistor to the arm of the trimmer (probably the middle terminal). Using solid wire, install this jumper--whose overall length may be 4 inches long--running it along one edge of the board for its main distance. Where you plug it in--beside the free resistor lead and beside the pot's center terminal--should be old hat by now; use your painfully acquired good judgment.

As I've described it, the center terminal of the trimmer will probably reach under the bypass, if you've bent its leads outward as I suggest. If the trimmer is snugly placed against the bypass capacitor, you will not be able to put this jumper in exactly adjacent to the desired terminal; you will have to approach it from the other side of the trimmer. That's all right; it does mean, though, that the stripped end of the jumper will have to be long enough to reach under the pot and cross that terminal. Therefore, strip this end bare for about 5/8 of an inch.

For now, the "end" terminals of the pot
will be left alone, since where they go depends on which bridge you are wiring. Two 8-inch stranded lengths of wire are all that is left to do. One of these comes from the oscillator's negative line, and the other from the amplifier's negative battery line. Basically, the modifications to the boards are now complete.

The Resistance Switch

When you first look at a new component, always give some thought to the "target practice" that you, as a blind person, will have to do; this is just a process of planning ahead. The 6-position switches recommended here have two terminals inside an outer circle of terminals--these are the swingers. If you were to solder to the outer ones first, the swingers would be much harder to reach. Therefore, take a look at your circuits to see where these things go. In each of these bridges, the swingers go off to other items; this means that they must need long pieces of wire. The first thing, then, is to solder 8-inch pieces of stranded wire to the swingers, subsequently being careful not to bend these too often and break them off their terminals.

Now tin the far ends of your swinger wires. You will use these for making connection to your continuity tester; with the other tester lead, you will be able to find "position 1" and "position 6" with certainty. For the resistance bridge, the precision resistors have been sorted already (see "Experimenting with the Parts"). On one pole of the switch (on the only pole, if that's the kind you have), the lowest value resistor --that gives the highest pitch--goes on position 1, with the highest value ending up on position 6. Cut one end of each resistor to 1/2 inch. With needle-nose pliers, make a hook in each short leg, hook it through its intended switch terminal and squeeze the hook closed with the pliers. There should be some space between the switch terminal and the body of the resistor--perhaps 1/4 inch. With the shaft of the switch held in a vise, arrange the assembly for stability and solder each connection in turn.

There is a tradition that says not to solder to a switch with its terminals pointing straight upward--even toggle switches. Depending upon the construction of the switch, solder can sometimes "wick" down the terminal to ruin its internal assembly. Rotary switches are most susceptible to this, so it is always wise to grab the shaft in the side of the vise jaws, having the terminals come off straight to the right or left. Being right-handed, I usually have them pointing to the left, so that I can--with the iron--first find the vise, the carriage of the switch, and then the terminal. I want to avoid the resistors and wires off to my left.

Holding each resistor so that it comes straight off to the left, which is how I've done my switch, can be done by hooking its far end to some object on the table via a length of wire, a string, or a weak rubber band. Another way is to wrap solder around its lead at the terminal (so that you no longer have to feed solder) and hold the far end of the resistor yourself. In using this latter method, it is important to make sure that the wrapping of solder does not slide away from the terminal toward the resistor body. Also, leave this wrapping connected back to the spool; you will know the solder has melted when the length to the spool drops away from the connection.

The far ends of the resistors can now be connected together. You could just bundle them together and solder them in a clump. This lacks a certain elegance, and besides, it is hard to solder that many pieces of metal without an inner one escaping involvement with the solder. I separately hook each resistor around a curved piece of solid bare wire, thus having six connections to solder along this wire. I eventually give the segment of wire the same curvature as the
circle of terminals. The result is a nice little skirt of resistors that extends off the switch; people come from miles around to see it. To accomplish this, cut the resistors' ends to 1/2 inch and put hooks in them; position them at 3/16-inch intervals along the bare wire and squeeze the hooks closed. After soldering each hook, attach an 8-inch stranded wire lead anywhere along this assembly.

You are not quite done if you have a 2-pole switch and you want to include my gain-boosting trick. Assuming you have connected a lead to the swinger of this pole, connect another a-inch stranded lead to both positions 5 and 6. Do this by stripping and tinning a 5/8-inch portion of one end; pass this through the lug of position 6 and hook it in position 5. Solder both lugs.

Test your work with your continuity tester. Temporarily mount a knob on the shaft and put your tester between the swinger and the junction of all the resistors. You should be able to hear all 6 pitches as you turn the knob.

The Capacitance Switch

Put a wire on the swinger (or swingers, if yours is a 2-pole unit). If you want to include my gain-boosting trick, put a wire through position 6, hook it around position 5, and solder both lugs.

Since the capacitors vary in size, the assembly around the switch cannot be as symmetrical; it can, and should, be made neatly. Basically, the same thing is done as with the resistors. If you really want to be neat about it, you can attach the largest unit first (probably the 0.1uF on position 3, or a silver-mica unit, if its size prevails). Then, all smaller units can be given enough extension from their lugs to be "centered" with respect to the largest dancing partner-this is a frill.

For each unit, spread its leads to the sides, cut the one intended for the switch short as before, hook it in its lug and solder it in place. Position 1 goes to the
positive lead of 10uF, position 2 goes to the positive lead of 1uF, ... and position 6 goes to one end of the 10epF silver-mica unit. (Only the first two have polarity concerns.)

At the far ends, cut their leads so that
they all end at about the same distance from the switch (individually, their leads will hence be of different length, depending on size). Hook these ends around a curved piece of bare wire and solder them, just as you did with the resistors. Attach a stranded lead to this grand junction.

Gain-Boost Connections

As described, two wires should emanate from the second pole of your 2-pole 6-position switch; your continuity tester should register that these become shorted when the switch is in position 5 and 6. If you will recall from Part III, there is a "gain enhancement" network around pins 1 and 8 of the LM386 amplifier. As described, a 470-ohm resistor goes from pin 1 to the negative end of a 10uF capacitor. The leads from this pole on your switch can be used to short out that resistor in positions 5 and 6, thus making the amplifier a bit more sensitive. (This truly is a
frill, and should not be considered crucial.)

I will leave it up to you as to where to plug in the switch leads. The idea is to solder one to each end of the 470-ohm resistor. Make sure that the one for pin 1 is clamped securely; you don't want it to flop over onto pin 2 and cause you circuit-board troubles.

Other preparation

As you buy them, the shafts of pots are often made long, expecting you to cut them to size. About 1/2 inch of shaft is usually safe; filing it shorter may later prove desirable to bring your knob close to its Braille scale. Clamp the unwanted portion of the shaft in a vise and make this cut with a hacksaw. Use a coarse file to finish off the top of the shaft.

The main pot only needs two wires, one for the arm and one for the "bottom" (counterclockwise end). Find the two terminals that make your continuity tester's pitch drop as you turn the shaft clockwise, and go back up to indicate a short circuit at the counterclockwise end. Attach 8-inch stranded leads to these two terminals.

Final Wiring of the Resistance Bridge

On your resistance-bridge board, install a jumper between one end of the trimmer (it doesn't matter which) and the free end of the 100-ohm resistor off pin 3 of the 555. The aim of the main 10K pot, its middle terminal, goes to the arm of the trimmer--this point already has a long jumper back to the amplifier. Plug this wire in somewhere near the trimmer, bend it down on top of the amplifier's jumper wire and solder it there. The other wire off the main pot goes to the oscillator's negative battery line.
Also going to the 100-ohm resistor off pin 3 of the 555 is the arm of that pole on the 6-position switch that is involved with the resistors. This can be plugged in in two places--at either end of the jumper that goes from this resistor to the trimmer. The lead from the amplifier's negative supply line goes to the junction of all the precision resistors; hook it around and solder it there.

Final Wiring of the Capacitance Bridge

On your capacitance-bridge board, install a small jumper from one end of the trimmer to the oscillator's negative battery line. The arm of the main 10K pot, its middle terminal, goes to the arm of the trimmer--this already has a long jumper going back to the amplifier. Wherever you find room, insert this pot lead near the trimmer, bend it down on top of the amplifier's jumper and solder it there. The other lead from the main pot goes to the free end of the 100-ohm resistor off pin 3 of the 555.

Also going to the 100-ohm resistor (and the bottom of the main pot) is the arm of that pole on the 6-position switch involved with the capacitors. The lead from the amplifier's negative supply line goes to the junction of all the capacitors; hook it around this junction and solder it there.

Initial Testing

Before mounting them in cabinets, you can test them. If you notice, in the jungle of wires associated with each assembly there are only two which are uncommitted. One is from the junction of standards on the switch; the other is from the oscillator's negative. These will go to the binding posts, after they are mounted. Not only can you test them, but you can make a good shot at calibration in this state (the main source of error being that the scales I sent you may not exactly match your pots).

Connect batteries to the resistance bridge and turn it on. With the volume control turned up, it should make noise; turning the main 10K pot should change the level of its audible tone, but it should never go away. Now, short out the two "test wires." With these connected together, turning the main pot to zero (counterclockwise) should cause the tone in the loudspeaker to "null" (or nearly so).

Connect batteries to the capacitance bridge and turn it on. With the amplifier turned up, turn the main pot and listen to the tone; it should increase in volume as you go clockwise, and it should "null" as you approach zero. Now, without shorting them, twist the "test wires" together for as long a distance as you can (creating what is known as a "twisted pair"). Make sure that their bare ends are not touching. Now move the range switch to position 6--the 100pF position.
The "null" in the tone will not be found at zero, but some small ways up from zero. What you are testing is sometimes called a "gimmick capacitor;" what's more, it is variable--the more wire you twist together, the higher will be the capacitance.

Boxing the Project

Now that projects are getting complicated, there are things you have to watch out for. In the specific case of these bridges, these things are: Mounting the board against a panel, then mounting a switch or pot so close to it that you can no longer screw it down; mounting the main post so close to another control that there is no room for the Braille dial; mounting the range switch of the capacitance bridge so close to a wall of the box that there is no room for its standards; and drilling speaker holes adjacent to where a switch will be located, only to find that there is not now room for the speaker. To prevent some of this, I'll tell you where I put things.

I considered the aluminum cover plate to be the bottom of the cabinet; only the batteries are mounted on this plate. With its open side down and the long side of the box toward you, drill a 3/8-inch hole 2-1/8 inches from
the right end, and 2-1/8 inches from the side nearest you; this is for the main post. Three-eighths inch holes are drilled for the volume control and range switch at 1-1/4 inch from the left end--1-1/4 inches away from you for the volume control, and 1-1/4 inches down from the farthest edge for the range switch. The on-off switch is vertically centered (2-1/8 inches away from you), and is 2-1/2 inches to the right of the left end. Because I prefer to reach over the box and manipulate
them, the binding posts are mounted each side of center on the back panel; these cry for 5/16-inch holes. All of this leaves room for the board on the right-hand two-thirds of the front panel (nearest you). The speaker is on the right end of the cabinet.

As good as I am, my projects are not without quirks. The Radio Shack 2-1/4 inch loudspeaker is just large enough to prevent installation of the bottom plate, once it has been glued in place. Therefore, I used both thumbs to bend the dickens out of one end of this pleat, making a lip that accommodates the lower edge of the speaker. Filing a notch for the speaker in the end of the plate would be another way of handling this.

For the main pot, and possibly for the range switch, you will want to make use of the "locating lug" so as to prevent any chance of twisting. As mentioned in Part III, put the control through from the bottom and twist it in the hole, causing its locating lug to score an arc on the back side of the panel. Then, somewhere on this are, drill a hole whose diameter is the width of that locating lug.

The Braille scale can be attached with "double-sided Cellophane tape"--a standard item in stationery stores. Handling this tape is no picnic, but it works very well if you can avoid putting creases in it as you
place it along the four edges of the dial. Before applying the tape, wipe both the dial and the top of the box with isopropyl alcohol to clean them.

The Braille dials take a bit of preparation. You will notice that I have provided four of them which must be cut apart. Next, they have been Thermoformed with a washer in the middle of the master so that you can accurately find the middle. The raised portion made by this washer can be placed over the 3/8-inch hole and pushed down into it; the result is a dimple on the back side of the dial that will help you center it in the hole, once it has been loaded with double-sided sticky tape.

By the way, if you don't like these dials,
you can copy them in thin sheet metal. Cut some pieces of sheet metal the same size and put 3/8-inch holes in the middle. (To get these 3/8-inch holes, the only safe way is to drill smaller pilot holes in them first, then use a gadget called a "tapered reamer" to enlarge them.) Lay the sheet metal on a piece of hard wood and place the plastic dial on its back over the metal. Then, with a center punch and a hammer, bash the dots into the metal, using the plastic dots as pilot holes for the center punch.

Once you have taped the plastic blank in place, you can "pare" around in the 3/8-inch hole to clear away the unwanted dial plastic. Use a small knife blade for this and be careful; you don't want to take some of the box with you.

Mount everything in the cabinet and prepare to solder the binding posts. Before you tighten the nuts on the binding posts, make sure that they are turned so that the hole through their screws (under the plastic sleeve) is oriented vertically; nothing is so frustrating as to have to find out which way to put a wire through. After securing everything, and with the box lying on its back, fold a lip on a file card and hang it over the edge of the box so that when--not if, in my case--you hit the box with your soldering iron, you won't contaminate the tip of the iron and make a mess of the box.

Solder the wire from the "standards" to the red binding post, and the one from the oscillator's bus line to the black binding post. Label these in Braille so that, if I come to your house, I can tell the difference. (In Nemeth Code and Computer Braille, the plus
sign is the "ING" sign, and minus is a hyphen.)

Mounting the pointer knob on the main pot deserves a little discussion. The mechanical rotation of a pot is almost never the same as its significant electrical rotation. This is true because most pots have crimp fasteners that hold on to the ends of their resistance element; the "wiper" or "arm" can usually rest on these for a couple of degrees of rotation. When installing the knob, it is necessary to assure that "zero" marks the beginning of the resistance element, not the
counterclockwise stop.

With the terminals of the resistance bridge shorted and with its range switch set to position 3 or 4, set the knob so that the tone just begins to increase as the pointer advances from zero. With the terminals of the capacitance bridge left open and the range switch set to position 1, set the knob so that the tone just begins to increase as
the pointer advances from zero.

If you can't find the pointer knob I recommend, you can always make your own. This is commonly done in two ways. If you find a knob with a fairly large skirt--perhaps 7/8-inch in diameter--you can cement a pointer on the bottom of it with Pliobond. I have gotten very practiced at cutting rectangles of plastic the width of the skirt, then filing the corners and cutting one end down to a point; the result is a disc with a triangular pointer emanating from one side. Another scheme is, if you can find a good-quality knob (which has a metal insert for two setscrews), using one setscrew hole in the knob to carry a long bolt which has been bent down and filed to a point. To keep this bolt from turning, put a nut against the outside of the knob which can be tightened to secure its position.

Once again, I must leave it up to you as to
how to mount 9-volt batteries. If you discover a good way, will you please let me know?

Calibration

Both instruments are calibrated similarly. Affix a standard to the binding posts--0.01uF in the case of the capacitance bridge, and 100K for the resistance bridge. In both cases, turn their range switch to position 4. Set the knob to exactly full scale, then turn the trimmer until a null occurs in this position.

If you were to connect, for example, a 2.2K resistor to your resistance bridge, no null could be gotten in position 1 (0 to 100
ohms), no null could be gotten in position 2
(0 to 1K ohms), and a null would finally occur a little past the second double-dot mark with the switch in position 3 (0 to 10K ohms, hence 2.2K). Advancing to higher positions would cause the null to be so close to the zero mark as to be meaningless.

If you were to connect, for example, 0.047uF to your capacitance bridge, a null would be gotten at zero in position 1 (0 to 10uF), a similar null would be gotten in position 2 (0 to 1uF), and a significant reading would be possible in position 3 (0 to 0.1uF, where the null would occur just before mid-scale). No null would be possible on higher ranges.

You will notice that in positions 5 and 6 of your capacitance bridge, the null is not at zero, even when a test unit is not connected. What you are seeing is the so-called "distributed capacitance" of the instrument-from perhaps 3 to 5 picoFarads. This is unavoidable, and you should add this figure onto any capacitors measured in position 6.

You will notice, when measuring very high resistances (especially in position 6), that the "null" will be very imperfect (not very deep). This is because of the distributed capacitance of the instrument, and is unavoidable. You should consider the accuracy of measurement in position 6 to be less than that of other ranges.

If you get an imperfect null with your capacitance bridge, it means that the "Q" (quality factor) of the capacitor is less than that of your standards. If this effect is extreme, you should discard the capacitor.

Reflections

Except for chips, which are very difficult to identify electrically and which you are better off putting in labeled envelopes, you can now identify all common components as to their type, and can measure the value of those most common--capacitors and resistors.
Besides these instruments, I would like to advance a couple of other attributes you have gained.

If you look closely at the language in the "Step-by-Step" section of this Part IV, you will notice that it began to strongly resemble that of the "Circuit Descriptions." This is no accident. As you become more familiar with the language of this science, you will care more that a magazine article tell you where things connect; familiarity with the physical materials will help you dictate where they are mounted. Furthermore, your building techniques may not be compatible with some other guy's who wants to commit everything to a printed board and who puts his chips at odd angles.

Finally, you have some good cabinet work under your belt. You have seen five electronics projects go from the abstraction of a collection of parts through to finished articles with real functions. You are in a powerful position now; you can help your nieces and nephews through the difficult stage of "realizing" their project merit badges in the Scouts; you, their aunt or uncle, know how to solder parts and drill holes, and can give them the meaningful head start that they missed in Part I.

Address List

[Note: The hardest thing to track down was a 10K precision pot. Even if you locate the pot of your dreams, you may have the Devil's own time trying to find a place that will sell it to you in small quantities. The pot decided upon is the Clarostat 58C1-10K, a $4 item with good linearity. Furthermore, I stumbled across a very nice operator at Brill Electronics in Oakland; Pamela at Ext. 26 tells me that she will try selling to you.
If this cannot be arranged, call Clarostat and get the name of a store near you; make sure that they are of such a size as to sell to individuals. I was not so lucky as to
find a supplier for the G.C. pointer knob; call G.C. and see what they can do to help you.]

  • Brill Electronics P.O. Box 1378 Oakland, CA 94606
  • (415) 834-5888, ask for Pam at Ext. 26.

  • Clarostat: (603) 742-2183 for the name of your local distributor.
  • Jameco Electronics 1355 Shoreway Road Belmont, CA 94002 (415) 592-0097
  • Mouser Electronics 11433 Woodside Avenue Santee, CA 92071 (619) 449-2222
  • G.C. Electronics 400 South Wyman street Rockford, IL 61101 (815) 968-9661

The SMITH-KETTLEWELL TIMING-TONES GENERATOR

by Jay Williams

Abstract

Designed for applications in making taped radio productions, this audible "timer" contains a high-pitched beeper which goes off each second and a low-pitched beeper which goes off every ten seconds. The two beep rates are derived from a common generator, the XR2242; the trigger and reset buttons are common to both beepers, so synchrony is assured. If the XR2242 (made by Exar) proves hard to get, the Fairchild UA2240 or the Intersil ICM7240 can be used as well, since the features used here are common to all such chips (See "Timer Chips," SKTF, Winter 1986).

Introduction

Necessity, that well-known mother of invention, drove our engineering lab to devise this instrument on behalf of those who must monitor seconds and groupings of seconds for their work or hobbies. Specifically, Mr. Roland Blount of Durant, Florida, needed something to assist in timing introductory announcements in relation to theme music in taped radio production work.

The unit has the following controls: an on-off switch, a trigger switch that starts both beepers running, a reset button that stops them and resets their count to zero, and a selector switch that allows you to hear the seconds beeper by itself, the ten-second beeper by itself, or both. As described here, the circuit is run from a 9-volt battery; this demands that you remember to turn the device off during long intervals of disuse. Like the "Volume Level Indicator" (SKTF, Winter 1981), it eats a battery quickly and quietly.

Circuit Operation

An XR2242 timer is adjusted so that the output of its last divider, pin 3, is a 1Hz squarewave. When so adjusted, a 256 signal appears on pin 8, the output of the internal clock. This signal is useful for calibration. The 1Hz signal drives one-half of an NE556, this half being the "seconds beeper." The 1Hz signal also feeds a CD4017 so-called "one-of-ten decoder;" pin 3 of this chip provides a divide-by-ten output. This output drives the second half of the 556, which comprises the "ten-second beeper."

Since the output of the XR2242 goes from logic high to logic low when triggered, the CD4017 is wired for negative-edge triggering. This is done by tying its clock pin, pin 14, high, and clocking the chip with the "Enable pin," pin 13. In addition, the outputs of these dividers change logic states each half-cycle. If these outputs were to drive the 556 directly, the beeps would be intolerably long. Therefore, these outputs each go through an RC network and a PNP transistor to the charging resistors of the respective halves of the 556. As a result, the beeps can be made crisp; the lower-pitched one was given a somewhat longer duration than the seconds one.

Circuit

The negative battery terminal goes to circuit ground. The positive battery terminal goes through an SPST switch to the plus-V line. The plus-V line goes through a 100uF capacitor to ground (its negative lead at ground). Pin 4 of the XF2242 is grounded; pin 1 goes to plus v. Pin 8 of the CD4017 is grounded; pin 16 goes to plus V. Pin 7 of the 556 dual timer is grounded. On the 556, pins 14, 10 and 4 go to plus V.

On the XR2242, pin 5 (reset) and pin 6 (trigger) each go through their own 47K resistor, then through a normally open pushbutton switch to plus V. For the 256Hz "internal time-base," pin 7 goes through a 0.047uF Mylar capacitor to ground, and also through a 100K calibration rheostat to plus V. Pin 8, the time-base output, goes through a 22K pull-up resistor to plus V. Pin 3, the 1Hz output, goes through 22K to plus V, and to pin 13 of the CD4017 (formally called the 4017's "Enable terminal"). As mentioned in the text, the CD4017's "clock terminal" now becomes its "enable;" pin 14 goes to plus V. Pin 15 of the 4017 (reset) goes through 10K to ground; pin 15 also goes to the junction of the reset button and the aforementioned 47K resistor leading to pin 5 of the 2242.

The 1Hz output, pin 3 of the 2242, goes through 22K, then through 0.47uF to the base of a 2N2907. The 0.1Hz output, pin 3 of the 4017, goes through 22K, then through 1uF to the base of another 2N2907(positive of the cap a t the base). Each base goes through 22K to plus V, and the emitters both go to plus v.

The "1Hz collector" goes through 22K to pin 1 of the 556; pin 1 goes through 22K to pins 2 and 6, which are tied together. Pins 2 and 6 also go through 0.01uF to ground. The "0.1Hz collector" goes through 68K to pin 13 of the 556; pin 13 goes through 68K to pins 8 and 12, which are tied together. Pins 8 and 12 also go through 0.01uF to ground. The arm of a 3-position rotary switch is grounded. Position 1 is left blank, position 2 goes to the junction of pins 2 and 6, and position 3 goes to the junction of pins 8 and 12.

On the 556, pins 5 and 9 (the beeper outputs) each go through 47 ohms to the positive leads of 10uF capacitors. The negative ends of these capacitors are tied together and go through the speaker to ground.

Calibration

Unless you have access to a frequency counter (or have elected to clock the unit with a crystal oscillator and frequency divider system), you must employ
patience and ingenuity in order to obtain a reasonable facsimile of seconds. If you have a shortwave radio, you can use WWV-or another such service which broadcasts time information. Barring that, you can achieve virtually the same accuracy with the help of one of the ubiquitous quartz-controlled clocks or watches, listening for stray RF signals on a standard AM radio. These timepieces produce all sorts of hash which often includes 1Hz pulses, or a simple multiple thereof. Hold the quartz clock near your radio's antenna; fish around at the low end of the band until you encounter the "hash." Then jockey the clock around until the desired pulsations emerge through your radio. Now, the trick is to somehow anchor that mess so you can attend to the adjustment of the
timing-tone generator.

If you're far enough out that quartz clocks are not prevalent, remember that a very large number of good-ol' mechanical watches tick five times per second. If your pitch recognition is good enough to be considered "perfect pitch," monitor the 256Hz signal on pin 8 of the 2242. Standard middle C is 261.626Hz; you won't end up completely out of the ballpark by using this as a guideline.

No matter what method you use, you will preserve your sanity if you do your calibration based on actual requirements, rather than on some absolute standard of accuracy. Remember, the timebase oscillator in this project is a form of "relaxation oscillator," and is likely to drift with temperature and changing battery voltage. Good luck, and good timing.

Pin Diagram for the CD4017

[Editor's Note: The Exar timer chip was covered in the Winter 1986 issue; besides
which, all of its pins are used here. The 556 timer is a recurring theme--even in the capacitance meter of this issue. For an example of how this "one-of-ten" decoder can be used, see "A Counter and Other Improvements for the Thermoform Copier," SKTF, Spring 1984.]

Introduction

The CD4017 has ten outputs, 0 through 9, which go high--one at a time--as its clock terminal is pulsed. Besides these outputs, a "carry-out" pin is provided for cascading these chips. Like many "counter" and "one-shot" chips, the "Clock" and "Enable" pins are interchangeable. Both operate a 2-input NAND gate inside the chip. However, the "Enable" pin first goes through an inverter before feeding this NAND gate. The purpose of this is so that the device can be set up for either positive-edge triggering or negative-edge triggering. With the true Enable, pin 13, grounded, triggering will occur on the positive transition of the Clock pin. With the Clock, pin 14, tied high, triggering will occur on the negative transition of the Enable pin.

  • Pin 8--Ground
  • Pin 16--Plus supply
  • Pin 3--0 Out
  • Pin 2---1 Out
  • Pin 4--2 Out
  • Pin 7--3 Out
  • Pin 10--4 Out
  • Pin 1--5 Out
  • Pin 5--6 Out
  • Pin 6--7 Out
  • Pin 9--8 Out
  • Pin 11--9 Out
  • Pin 12--Carry Out
  • Pin 13--Enable (low for enable)
  • Pin 14--Clock positive-edge triggered)
  • Pin 15--Reset

Parts List

Capacitors:

  • 2--0.01uF Mylar or disc ceramic
  • 1--0.047uF Mylar
  • 1--0.47uF Mylar or 10V electrolytic
  • 1--1uF 10V electrolytic
  • 2--10uV 10V electrolytic
  • 1--100uF 10V electrolytic

Resistors

(1/4-watt 5%, unless otherwise stated):

  • 2--47 ohm 1/2-watt.
  • 1--10K
  • 8--22K
  • 2-47K
  • 2--68K
  • 1--100K PC-mount pot, connected as a rheostat

Semiconductors:

  • 1--Signetics NE556 dual timer, or 556 of any make
  • 1--Exar 2242 timer chip
  • 1--RCA CD4017 one-of-ten decoder, or 4017 of any make
  • 2--2N2907 or 2N2907A

Switches:

  • 1--SPST on-off switch
  • 2--Normally open pushbuttons
  • 1--3-position single-pole rotary

Miscellaneous: small speaker, 9-volt battery connector, small cabinet of your choice.

AN AUDIBLE MULTI-MICROFARAD METER

[Editor's Note: This circuit came to us in translated form from The Federal Republic of Germany. Authored by Mr. K. Britz, this design first appeared in Funk und Elektronik, published by Deutsche Blindenstudienanstalt of Marburg. We have tampered with the design somewhat--combining two chips into one, and so forth. However, the initial work constitutes a brilliant invention; its originator deserves full credit. The Editor thanks Mr. Bob Trottman for getting us the circuit, and thanks, Jay Williams, for authoring this resultant article.]

Abstract

This instrument uses an audible timer to time the duration of a "one-shot" whose timing, in turn; is determined by an unknown capacitor. It is designed to measure large-value capacitors (electrolytics, for example), and can practically cover the range from 0.05 to 50,000 microfarads. The user counts beeps of the audible timer until the one-shot is timed out. A switch permits different charging rates to be selected; the timer's "beeps" may thus represent different capacitance values--from 0.1uF per beep to 1000uF per beep.

Introduction

With the advent of solid-state electronics, we now encounter capacitors ranging in tens, hundreds, or thousands of microfarads.
(Those measuring tens of thousands of microfarads are getting to be more common. I recently held a little beastie in my fist whose value was 15 farads. Its internal resistance was too high for filtering; it was designed for keeping computer memory alive.) Identifying the capacitance value of units lying around is always an issue for those of us who cannot read the labels, and this instrument serves this purpose splendidly.

In the article "Continuity Tester Uses" (SKTF, Fall 1982), it was described how a continuity tester can be used to test for
leakage, a shorted or open condition, and to test for polarity. It was also mentioned that one could get a "qualitative" idea of the value by noting the rate of descending pitch of the tester. While the unit to be described cannot be effectively used to test for leakage, shorts, or polarity, you can, by counting beeps, get a reasonably accurate "quantitative" value of capacitance--accurate to perhaps one-and-a-half decimal places. It is an adjunct to, not a replacement for, your other instruments.

The output of this new instrument is
digital. When the "on button" is pressed, a number of beeps are emitted by the instrument, after which it falls silent. A selector switch sets the order of magnitude at which charging of the unknown capacitance occurs. In this way, each beep of the unit takes on a weighted value of: 0.1uF, 1uF, 10uF, 100uF, or 1000uF. If the meter is set to the 1uF position, a 5uF capacitor should cause it to emit 5 beeps. Should you desire to take the time, putting the meter on the 0.1uF range will allow you to measure to another decimal place--47 beeps would mean 4.7uF.

If the capacitor is leaky, the number of beeps will increase, if not to infinity. If it is connected in reverse polarity, or is shorted, it will beep unto death. If it is on its way to being open, you will hear too few beeps--if any. Your suspicions may be aroused by an erroneous reading, whereupon you should test the suspected component with your continuity tester to see just what is going on.

As you use the device, you will notice wide variations in readings gotten from even "good" capacitors having identical ratings. Electrolytic capacitors are noted for being imprecise. (With this meter, the Editor has witnessed variations up to 30% as being rather common.) It has even been shown that the value of many electrolytics goes up when they are used at voltages significantly below their rated "working voltage;" this has been ascribed to the idea that their dielectric "re-forms" and gets physically thinner, thus causing the capacitance to increase.

Here at Smith-Kettlewell, Tom Fowle devised a scheme which gives resolution to the half-unit. He reasoned that if you could hear just where, within the cycle, the timer was interrupted, you could make a rough estimate as to the fraction of the next place. The output of this modified device sounds something like those two-tone sirens found on the other side of the pond. The circuit for both systems will be given.

Circuit Operation

An NE555 timer is wired as a one-shot whose timing capacitor is the one to be measured. The charging resistor is selected by a switch, determining the units of capacitance that are to be articulated. This one-shot is automatically triggered when power is applied; pin 2 (trigger) is provided with an RC network that brings it below 1/3 VCC long enough to ensure triggering. (This clever feature eliminates the need for a separate switch to start the cycle.)

Pin 5, the "Control pin," is provided with an adjustable voltage divider (a 10K pot across the power supply) so that the charging time may be altered for calibration.

The one-shot output, pin 3, activates the Enable terminal, pin 4, of the first half of an NE556 dual version of the 555. The first half of this 556 serves as the "beep rate oscillator"; it is the "timer" whose pulses are counted by the user. The output of this timer, pin 5, governs the Enable terminal of the second half of the 556, pin 10. This second half generates beeps that are heard by the user.

Note that the output of the "beeper" is capacitively coupled to the loudspeaker; this is done to avoid any current flow during silent periods between beeps. (In various other circuits, you will see the speaker connected, through a current-limiting resistor, between output and the plus supply. This is not possible here; because of the operation of the "Control terminal," this arrangement would put the output low between beeps, thus drawing current when the beeper is silent. We chose to do as the author of the circuit did and include the capacitor.)

In the Tom Fowle "siren-like" circuit, the Enable pin of the timer is tied high; the one-shot's output operates upon the Enable pin of the beeper instead. The timer's output goes through a resistor to the "Control pin" of the beeper so as to modulate its frequency. In this way, the silence between beeps now becomes an upward jump in pitch. (The "Control pin," sometimes called the "Voltage-Control-pin," gives the designer access to the voltage divider inside the chip that determines the point at which the "charge" and "discharge" processes occur. If left alone, this pin will rest at 2/3 VCC; if externally forced to be at some other voltage, the timing that governs the discharge point will be affected, resulting in a frequency shift.)

In this latter system, a time cycle is comprised of first a low pitch, then a higher one. The user counts each completed high beep, and then makes a guess as to how much of the final cycle was completed. I shall present both configurations as part of the circuit description; the Britz system will be labeled "(A)", and the Fowle configuration will be labeled "(B)."

Circuit

The negative battery terminal is grounded; its positive terminal goes through a normally open pushbutton switch to the VCC line. VCC is bypassed by a 100uF capacitor whose negative lead is at ground. Pin 1 of a 555 is grounded; pins 4 and 8 go to VCC. On the 556, pin 7 is grounded; pin 14 goes to
VCC.

Pin 2 of the 555 goes through 0.1uF to ground; pin 2 also goes through 470K to VCC. Pin 5 goes to the arm of a 10K ten-turn PC-mount pot (calibration); one end of this pot is grounded, while its other end goes to VCC.

Pin 7 of the 555 goes through 1K to pin 6. Pin 6 also goes to the positive test terminal, while the negative test terminal is grounded. (Be sure to mark their polarity clearly.) Across these terminals is the series combination of 22 ohms and a normally closed pushbutton; this discharges the capacitor for a second try at measurement. (This switch may be combined with the on-off button--a DPDT switch, one side used to turn the instrument on, and the other side used to short out the test capacitor when the button is released.)

Pin 6 also goes to the arm of a single-pole 5-position rotary switch (range switch). Each position goes through its respective 1% precision resistor to the VCC line as follows:

  • Position 1 (0.1uF per beep) goes through 10 megohms to VCC.
  • Position 2 (1uF per beep) goes through 1 megohm to VCC.
  • Position 3 (10uF per beep) goes through 100K to VCC.
  • Position 4 (100UF per beep) goes through 10K to VCC.
  • Position 5 (1000uF per beep) goes through 1K to VCC.
(A) Britz Configuration

Pin 3 of the 555 (the one-shot output) goes to pin 4 of the 556 (Enable of the timer). On the 556, Pin 5 (timer output) goes to pin 10 (the Enable of the beeper). Pins 3 and 11 (both Control pins) are not used.

(B) Fowle Configuration

Pin 3 of the 555 (one-shot output) goes to pin 10 of the 556 (beeper Enable). The timer's Enable, pin 4 of the 556, is tied to VCC. The timer's output, pin 5 of the 556, goes through 5.6K to pin 11 (the beeper's control pin).

Pins 2 and 6 of the 556 are tied together and go through 0.1uF to ground (being a key element in the timer, this capacitor should be of Mylar or other good quality); 2 and 6 also go through 4.7 megohms to pin 1. Pin 1 goes through 10K to VCC.

Pins 8 and 12 of the 556 are tied together and go through 0.01uF to ground. Pins 8 and 12 also go through 150K to pin 13; pin 13 also goes through 10K to VCC. Pin 9, the beeper output, goes through 47 ohms to the positive end of a 10uF capacitor; the negative end of this cap goes through the speaker to ground.

Testing and Calibration

First, set the 10K calibration pot somewhere in the middle of its range, otherwise the instrument will appear to malfunction. Also, calibration will be facilitated if this pot is at least a ten-turn unit.

Set the selector switch to position 1 (0.1uF per beep). Connect a capacitor of greater than 0.047uF across the terminals (it can be as large as 1000uF for this initial test). Carefully observe its polarity if it is electrolytic! Press the discharge button (if it is a separate switch) to ensure that the capacitor has no initial charge.

Pushing the test button should now make the tester emanate a series of beeps. Now, set the range switch over to 1000uF per beep; press the discharge button, and then the test button. It may not emit a beep at all; if it does, it will be a short one (unless you really do have 1000uF connected).

Calibration

You will look far and wide and spend much time and money trying to obtain capacitors whose measured values are consistent with what is stated on their packages. Probably the best you can do is to employ a number of identically labeled capacitors, setting the tester to their average value.

Set the tester to at least the next smaller unit from that you wish to measure. For example, if you have a plethora of 10uF capacitors, calibrate the meter with the range switch set to articulate 1uF intervals. Or, if Job gave you some of that patience of his, use the 0.1uF setting and shoot for an average of 100 beeps.

Some thoughts on Accuracy

While garden-variety aluminum electrolytics are noted for their impreciseness, electrolytics of the tantalum type should be somewhat more predictable--in addition to which, their leakage is a lot lower. Capacitors of Mylar or other plastic dielectric materials should be better yet. Remember what this meter is good for, identifying units you have laid down and forgotten about.

The timer in the first half of the 556 is given to its own impreciseness. The first beep, and hence the first time cycle, will always be about 20% longer than the following ones. This is true, since its timing capacitor starts out with an initial condition of 0 volts when the circuit is first activated, but initiates each successive cycle at about 3 volts--1/3 VCC. This off set in timing will be rendered insignificant aver a large number of articulations.

This circuit has inspired us here at Smith-Kettlewell to explore other counting and output modes for measuring capacitors of large value. Consider a fancy version of this circuit using Exar or Intersil timer chips to get more digits of information. For example, if an ICL7250 timer chip were used (see "Timer Chips," SKTF, Winter 1986), a two-digit BCD number would result from each measurement. We'll let you know of any future designs. Please let us know of any-thing you discover, and thank you, readers, for your valuable assistance so far.

[The Editor was so possessed by playing with this instrument that he changed his pushbutton on-off switch to a toggle. This way, he can listen to 470 1-microfarad beeps if he wants to. The instrument does draw current if left on, however, so don't forget it.]

Parts List

(Note: Use of a DPDT pushbutton, although it is a little harder to find, will mean that
you don't have to remember to "discharge" the unknown capacitor between tests; the normally closed contacts of one pole, through a 22-ohm current-limiting resistor, can take care of this when the button is released.)

Capacitors:

  • 1--0.01uF Mylar or disc ceramic
  • 2--0.1uF Mylar (the one off pin 2 of the 555 can be ceramic)
  • 1--10uF 10V electrolytic
  • 1--100uF 10V electrolytic

Resistors

(1/4-watt 5%, unless otherwise stated):

  • 1--22 ohm 1/2-watt 1--47 ohm 1/2-watt
  • 1--1K
  • 1--5.6K (used only in Fowle configuration) 2--10K
  • 1--150K
  • 1--470K
  • 1--4.7 megohms
  • 1--10K 10-turn PC-mount pot

Precision Charging Resistors

(given the accuracy of the instrument, these could be 5% units, but we had the 1% ones in stock):

  • 1--1K
  • 1--10K
  • 1--100K
  • 1--1 meg
  • 1--10 meg

Semiconductors:

  • 1--Signetics NE555, or 555 of any make
  • 1--Signetics 556, a 556 of any make, or two 555's

Switches:

  • 1--DPDT pushbutton, or two SPST normally open push buttons, or DPDT toggle
  • 1--single-pole 5-position rotary

Miscellaneous:

  • Two binding posts, preferably of different color and marked in Braille.
  • Appropriate 9-volt battery holder and snaps.
  • Small loudspeaker.
  • Small plastic cabinet.

THE TWEEDLE DUMP

The "Transmission-in-Progress Alarm for Modem Users" (SKTF, Winter 1985) has grown up to be a production item. In summary, there is often a need for an audible indicator "in line" with RS232 connections so that the computer user knows when data are being sent. The most obvious application for such a device is when "downloading" files from a data base via a modem; an audible detector will alert you that a file is being sent, and will fall silent when the transmission is finished.

While the prototype we described in Winter 1985 was viable, it was more complex than necessary. Dr. T.V. Cranmer (Tim Cranmer), who heads the Research and Development Committee of the NEB, simplified the "Transmission-in-Progress Alarm" considerably, and has seen to it that the device is commercially available. The following are excerpts from a letter describing the product and how to get it.

"Two garden-variety 25-pin D connectors, one male and one female, are mounted together using a shell assembly from B&B Electronics, Model 232CS (call them at 815-434-0846). The transducer is by Projects Unlimited, Model AI251, obtained from J.C. Hofstetter (513- 296-1010).

"The transducer is positioned so as to speak through a hole in the shell assembly. This 'peep hole' should be placed on top--the side adjacent to the longer row of pins.

"The name has also metamorphosed. Since the thing 'tweedles' all of the time you are dumping data, the ever-whimsical Fred Gissoni dubbed it--what else--the 'Tweedle Dump.' This is so apropos that we will ignore the pole of propriety and call it that-'TD,' if you want to be familiar.

"I have arranged for a local technician to make up a few dozen units. They will be sold to blind guys and gals at little more than the cost of the parts--namely, $16. They can be gotten from Mr. John Monarch, 525 Pawnee Trail, Frankfort, KY 40601.

"Circuit"

The negative side of the transducer is grounded and goes to pin 7 of both D connectors. Pin 2 on both connectors goes to the anode of a 1N34A diode; pin 3 of both connectors goes to the anode of another
IN34A. The cathodes are joined together and go through 10K (1/8- or 1/4-watt) to the positive side of the transducer. (I like 1N34A's because they have a smaller voltage drop than the silicon jobs.) Also, pins 4, 5, and 20 are connected straight through in the connector assembly."

The editor finds the $16 price tag to be quite reasonable; this represents about a $4 markup. If I were to make the necessary phone calls to try and obtain single pieces, I'd be foolish. Anyhow, I'd endure a fair-sized flogging, rather than solder to those connectors, and I'm not embarrassed to hire it done.

We keep ours in the line at all times--running printers and other peripherals.

Thanks, Dr. Tim, and thank you, Mr. Monarch.

THE "KINGS CENTRAL CATALOGUE" OF BOOKS

Organized by a true dynamo of energy, Chris Mackey, the Kings Central Catalogue is a list of recorded books in a bunch of libraries. Seven of the libraries are free lenders. One, the Volunteers of Vacaville, does charge a fee of some sort. I quote from her letter as follows:

"Do you, for example, want to locate the instructions for the Panasonic JE720E Electronic Calculator, or maybe all the books by Earl Stanley Gardner? Before you take your problems to your local transcribing group-and after you have inquired at the National Library Service and Recordings for the Blind--try Kings Central Catalogue (KCC). Once you have inquired through the KCC, you
will be notified if, and where, the book is located; it is then up to you to contact the library and make arrangements for borrowing.

"Inquiries should be accompanied by a self-addressed stamped envelope for a print reply; a return label will be sufficient for a reply in Braille or on tape. Telephone inquiries should not be made, since I may take a bit of time to search the catalogue to find what you're looking for."

Write to Mrs. H.V. Mackey at 202 Grangeville Blvd., Hanford, CA 93230.It is the editor's impression that this "list" is volatile; i.e., it is probably not printed in its entirety for your perusal. So, put a little thought into what you're requesting; say, "computer manuals, programming in 'Fourth,'" or "Computer manuals, Apple." Since she's a friend of mine, please be specific.