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-1989

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 Details regarding products, suppliers, and other contact information are original and may be outdated.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







by Bernie Vinther Edited by Jay Williams


In "Bernie's Little Blue Box" (SKTF, Winter 1987), I described a pocket-sized instrument which serves several functions. It not only serves as a relative indicator for voltage, current, and resistance, but also can be used as a battery tester, and as a continuity tester of extraordinary sensitivity. The enhanced version presented in this article includes two more functions: it can now indicate relative AC voltages, and it can be used as a signal tracer. Thus, you can now look for pulse trains in digital circuits, the presence of audio signals, and so forth. Also, the new design allows the metering of a much higher range of voltages--up to 400 volts, if appropriate caution is exercised. In spite of its having many functions, the new version is much easier to operate and requires one (not two), input jacks. This device is the perfect add-on to a talking multimeter.

[Editor's warning! The wiring of the switching circuits is not for the timid soul. Keep your wits about you at all times, and check and recheck. I assure you, you do not want ever to have to troubleshoot this circuit.]

(Various forms of print and diskette documentation are available from the author/designer. If you wish to buy one pre-assembled, this circuit-analyzer unit is also available from him. The author's address will be listed at the end of this article.

Description and Features

The CMOS ICM7555 is a chip whose binary output responds to analog voltages on its so-called "Trigger" and "Threshold" pins (see the article "Inside the 555," SKTF, Winter 1987). If you tie these pins together, the 555--and its CMOS equivalent--becomes a "Schmitt trigger." It takes a voltage of a certain "threshold" to trip the circuit (bringing the output low); once tripped, the input has to be brought below a "trigger" level, half the threshold voltage, to reset the circuit (bringing the output high again). The changes of output state can be heard through a small audio transducer (a piezo-electric device, in this case).

Logic voltage levels, or even audio signals, can trigger this Schmitt trigger to make output transitions that we can hear. By controlling the charge current in a "timing capacitor" (charging it from the outside and using a "Discharge" pin that the chip provides to "retrigger" the circuit), you create a current-controlled oscillator; the frequency will be proportional to the current drawn from an input voltage being tested, or the frequency can depend on the current supplied through a resistor being tested.

As in the original Blue Box, the CMOS version of the 555 timer chip, the Intersil ICM7555, is required. An ordinary bipolar 555 will not work because it does not allow the trigger and threshold points to be brought very near ground. Even so-called "low-power" versions won't work. The CMOS 7555, on the other hand, operates even when pin 5 (a junction on an internal voltage divider that defines the trigger and threshold points) is forced to within a few hundred millivolts above ground, and this makes the resultant Schmitt trigger a very sensitive voltage detector.

One improvement over the old Blue Box circuit (see SKTF, Winter 1987) is that there are no threshold voltages that must be overcome before the Box will oscillate. (The original design had a set of thresholds such as 100mV, 500mV, 1.5V, etc.) Now, a bias pot is provided to subtract out the threshold altogether, and the three ranges of input voltage are taken care of by resistors in the input network. [Bill Gerrey's note: Gee, I rather liked those thresholds for testing batteries. You could set them up to be go/no-go indicators of good vs. bad. Oh well, I'll build both kinds of Blue Boxes.]

The voltages that can now be measured are from 0.25 volts to about 400 volts. The lowest range can be used to measure directly across meter movements; in this service, the Blue Box is a "relative meter reader."

One interesting new feature is the "Audio Monitoring Mode." The selector switch, S1, has a mid-position in which the free-running capability of the 7555 is disabled (by breaking the connection to pin 7, its discharge pin).

This gives you direct control of the Schmitt trigger; the output (pin 3) remains high until a small positive voltage is applied to the input jack, at which time the 7555's output will switch low--causing a click in the transducer. When the voltage at the input jack falls to zero or goes negative, the 7555's output will switch high again--causing another click in the transducer. If you apply an AC signal of an audible frequency to the input jack, the 7555's output will switch high and low at the same frequency as the input signal, thus making it possible to listen to audio frequency signals.

Of course, the 7555 turns nice sounding music and voice audio into raspy sounding square waves, but at least you can hear if there is something there or not.

The Radio Shack blue box can still be used to house the instrument, but I prefer using boxes made by the Bopla company, or UniBox, because of their superior impact resistance. Also, the UniBox has a very nice built-in 9-volt battery compartment.

Using a good quality glass-epoxy perforated board (such as the ones made by Vector rather than the Radio Shack ones) will give you slightly better performance. This is true because of the extremely high leakage resistances and low stray capacitances offered by glass-epoxy.

The circuit has an extremely high input impedance, over 44 megohms in the medium-range position. It won't put a load on a circuit unless you want it to; you can provide a load by selecting from a wide range of built-in loading resistors, a necessary feature for battery testers. These load resistors also make it so that you can measure small currents with the Blue Box.

A 9-volt alkaline battery powers the circuit and will last for up to a year or so. It will also operate quite well down to 3 volts or up to about 15 volts in case you should want to use another type of battery. At 9 volts, the battery drain is only about 200uA. Anyway, I love equipment that will fit neatly into my shirt pocket and won't keep me broke buying batteries for it, and I don't like fumbling for a wall outlet to plug it in, cluttering up my work space with bulky size and power cords.

Survey of Controls, Jacks, and Physical Features

The box contains: two jacks, a set of function switches, a set of "load-resistance" switches, and a set of "sampling" resistors. These resistors serve as references with which an unknown resistance can be compared.

Function Switches

Although this circuit draws a minute current in most cases, it could be left in a mode where accidental shorting of the test probes would load the circuit heavily. Therefore, an on-off switch is good insurance. I'll call it S0 since its uncomplicated nature needn't concern us any further in this discussion.

S1 is a 3-position toggle switch. In position 1, the box functions as a volt-current meter; in position 2, signal tracer; and in position 3, relative resistance indicator and continuity tester.

S2 is another 3-position toggle switch which provides three ranges of sensitivity for the voltmeter function. When using the continuity tester mode to measure resistance, S2 should be in its middle position because the high- and low-range positions introduce complexities in the input voltage divider which would flaw the performance.

S3 is a 2-position toggle switch which selects DC or AC metering. Keep in mind that the DC setting is the optimal setting when using the signal tracer function.

Selectable Load Resistors

A selection of "load resistors" can be switched in; these provide multi-step shunts to ground for the input signal. When S1 is in the volts-current-meter position, these resistances provide a multitude of shunts for measuring small current values. For the "ohms/continuity" function, these load resistors determine the short-circuit current in the probes, and thus different ranges of test resistances can be selected. A 3-position toggle, S4, selects two different arrangements of resistors, and its middle position eliminates this resistance network from the circuit.

I suggest three switching schemes for the load resistors: 1) a make-before-break 11-position rotary switch in combination with a DPDT toggle for S4; 2) a set of 11 DIP SPST switches with the DPDT toggle for S4; and 3) a set of separate mini-toggle SPST switches with the DPDT toggle for S4. For reliability and ease of operation, I would recommend the mini-toggle arrangement. However, the DIP switch configuration promotes economy of space, ease of replacement and of assembly. The rotary switch will have to be set between positions to take advantage of all possible resistance combinations which, although an awkward procedure, may be acceptable if a knob-and-pointer method of selecting the resistors attracts you.

Sample Resistances

Sample values of resistance are provided for comparison in the "ohms/continuity" function. These project from the box in groups of three; the middle unit of each group is shorter than the outer ones so that the probe can be made to touch both outer ones (putting those two in parallel), or to touch all three (paralleling all three) as desired. If you wish to compare one of these to an unknown resistor, leave the positive probe connected to the unknown; then, alternately touch the negative probe to the unknown, then to a selected sampling resistor until a similar pitch is heard for both connections. Armed with some knowledge of how to calculate parallel combinations of the sample resistances, you can determine the value of your unknown resistance to a pretty fine point.


So that they can never be confused, the jacks are of different sizes. The earphone jack (a 1/16-inch sub-mini) is of the closed-circuit type; the transducer (speaker) is disconnected when the earphone plug is inserted. (A high-impedance earphone is required.) The test-probe jack is a 1/8-inch open-circuit type. Note that the continuity tester mode reverses the polarity of the input leads. Remember this when checking the polarity of electrolytic and tantalum capacitors, as well as diodes.

Circuit Operation

The switching arrangements in the Blue Box are where most of the "tricks" are performed. The arrangement of the 7555 is fairly standard. One difference from the usual circuit connection is that pin 5, the "Voltage Control" pin, is shunted to ground so that the voltage swing over which the RC timing circuit operates is greatly reduced (by a factor of about 15).

A small variation is that the timing capacitor off pins 2 and 6 is returned to plus 9 volts instead of ground. This prevents the capacitor's leakage from loading the continuity tester when in its most sensitive mode. Remember, this circuit is very sensitive to such things. This arrangement also lets you hear any indication of excessive leakage which would necessitate the replacement of this capacitor. A silver mica or other good quality capacitor is essential here.

In the "voltage-sensing modes" (whether direct-coupled or through a coupling capacitor), the sleeve of the input jack is referenced to a bias point, not to circuit ground. This "bias point" is created by a voltage divider whose bottom element is a 50K trim pot connected as a rheostat. This trimmer is adjusted so as to get rid of the threshold voltage; it puts the input jack above circuit common. In the DC mode, the test probe going to the sleeve is the negative input lead.

For DC operation, the tip contact of the input jack is coupled through two 22-megohm resistors in series. To achieve the lower voltage range, a 6.8-megohm resistor is shunted across the 44 megohms of input resistance. For the highest range, a 2.2-megohm resistor is connected from pins 2 and 6 of the 7555 to the wiper of the threshold trim pot, thus shunting most of the input signal to ground. The high-range position also shunts the input resistance with a 0.47uF 1,000-volt capacitor to compensate for the shorter time constant due to the shunting action of the 2.2-megohm resistor.

Switching the circuit to the AC mode "unshorts" a coupling capacitor on the positive input lead and places a rectifier diode across the input circuit (at the far end of the coupling cap). This diode is shunted by 22 megohms (which lowers the impedance of the input to 14.7 megohms on AC).

In the signal-tracing mode, Pin 7 of the 7555 is disconnected from its charging resistor; the 7555 will no longer "free-run," but will respond to a change in direction of the input voltage, provided it crosses the threshold as determined by the trim pot. Also, the cold input lead is AC coupled to input ground by a 0.47uF- 500-volt capacitor to prevent DC components from returning through the input.

Load resistances are introduced by throwing S4, a 3-position toggle switch, to either side of "center off." This places either of two arrangements of shunt resistors between the tip of the input jack and ground. The input circuit beyond these shunts is protected by the 470K resistor associated with pole C of S1.

In the continuity-tester mode, the sleeve of the input jack is taken to plus 9 volts. The circuit, or resistor, under test provides a current path for charging the timing capacitor (a path which includes the 680K discharge resistor between pin 7 and the pin 2-and-6 junction).

Note that since the sleeve of the input jack now goes to plus 9 volts, the polarity of the test leads is opposite from when voltages are being measured.

For continuity measurement, leave the voltage range switch in the middle position and the AC/DC switch in DC. When the load-resistance switch, S4, is in the center-off position, no shunt to circuit common is present and the continuity tester is at its most sensitive.

In position A, S4 presents a shunt made from 11 switchable resistors, one resistor per switch. In Position B, the shunt has additional resistors in parallel.

Circuit Description

Note: There are many junctions containing more than one solder connection, so hang on to your pigtails. The point is, read the circuit through a couple of times before blazing away; this will alert you as to where to solder right away and where to wait for additional parts.

The Easy Part

The positive of the 9-volt battery goes through an on-off switch, S0, to the plus 9V line. The negative side of the battery goes to circuit ground. From +9 volts to ground is the parallel combination of 47uF and 0.1uF capacitors; the negative lead of the 47uF unit is at ground. Pin 1 of the 7555 is grounded; pins 4 and 8 are tied together and go to the plus 9V line. Pins 1 and 8 are shunted by a 10uF capacitor whose negative lead is at Pin 1.

Pins 2 and 6 are tied together and go through a 100pF capacitor (silver-mica or glass) to plus 9 volts. Pin 5 goes through a 4.7K resistor to ground.

Pin 3, the output, goes through a 0.1uF 25V capacitor to the top of a 25K volume pot whose bottom is grounded. The wiper goes to the tip of the sub-miniphone output jack whose sleeve is grounded. The transducer used as the loudspeaker has one lead grounded and the other going to the switched contact on this jack.

A 50K PC-mount pot has one end grounded, and its other end is tied to its wiper; this junction goes through a 470K resistor to plus 9V. Hereafter, the junction of this rheostat and the 470K resistor will be called the "bias point."

Input Switching

Now for the fun part. In order to maintain orientation, the toggle switches and their terminals are designated as follows: S1--mode switch, 3-pole double-throw with center-off; S2--voltage range, 2-pole double-throw with center-off; S3--AC/DC, 2-pole double-throw 2-position; and S4, load-resistance group selector, 2-pole double-throw with center-off. Each of these has more than one pole; the poles will be lettered "A," "B," and "C" (just "A" and "B" for the double-pole switches). The "Positions" will be labeled as they are: the AC position of pole A on S3 will be "the AC position of S3A"; the Low-range position of pole B on S2 will be "the low-range position of S2B."

The positive input lead goes through a 0.1uF 1,000-volt capacitor, then through two 22-megohm resistors to the arms of S1A, S2A and S2B. The junction of the capacitor and the 22-meg resistor goes to the DC position of S3A. The junction of the tip of the input jack and the capacitor goes to both arms of S3. (Unused are the AC position of S3A and the DC position of S3B.)

The AC position of S3B goes to the cathode of a 1,000 PIV rectifier diode; the anode of the diode goes to the bias point. This diode is shunted by a 22-megohm resistor.

The tip of the input jack also goes through a 0.47uF 1,000-volt disc capacitor to the high-range position of S2b. S2A puts a 6.8meg resistor in parallel with the 44meg; the 6.8meg resistor has one end going to the junction of the input cap and the first 22meg resistor, while the low-range position of S2A goes to the free end of the 6.8meg unit. The high-range position of S2A goes through 2.2 megohms to the bias point.

The sleeve of the input jack goes to the arm of S1B. The sleeve also goes through a 0.47uF 500-volt capacitor to the bias point. The "voltage" position of S1B goes to the bias point. The "Continuity" position of S1B goes to the plus 9V line.

Both "voltage" and "continuity" positions of S1A are tied together and go to pin 7 of the 7555. The arm of S1A (which also goes to the arms of other switches, see above) goes through 680K to pins 2 and 6 of the 7555.

S1C controls the "ground return" of the load resistors. The voltage position of S1C goes to the bias point. The continuity position of S1C is grounded ("circuit ground" inside the Blue Box). The arm of S1C goes through a 470K resistor to the bias point; this arm also goes to the arm of S4B, the load group selector switch to be seen shortly.

Bringing On The Load Resistors (the DIP- or Toggle-Switch Approach)--

One side of all eleven SPST toggle switches are tied together and go to the arm of S4A. The two "on" positions of S4A are tied together and go to the tip of the input jack.

The free terminal of each SPST switch goes to its assigned resistor in the following respective order: 330 ohms, 1K, 3.3K, 10K, 33K, 100K, 330K, 1meg, 3.3meg, 10meg, and 33meg (this last being made of a 22meg in series with 11meg). The first five resistors are 2-watt units, the 100K is a 1/2-watt unit, and the five higher-value ones can be 1/4-watt.

The free ends of these resistors are tied together and go to the arm of S4B.

The even-numbered switches have additional resistors; i.e., the junction of SPST switch No. 2 and its 1K resistor also goes to another 1K unit, No. 4 has an added 10K unit, 6 has an added 100K resistor, 8 has an added 1 megohm, and No. 10 goes to an additional 10 megohms. The first three are 1-watt resistors, while the two highest ones are 1/4-watt. The free ends of these resistors are tied together and go to the downward position of S4B. (The upward position of pole B is left open.)

Rotary Switch Approach

A single-pole 11-position rotary switch of the shorting type is used. Its arm goes to the arm of S4A; the two "on" positions of S4A are jumpered together and go to the tip of the input jack.

Positions 1 through 11 go to resistors in the following respective order: 330 ohm, 1K, 3.3K, 10K, 33K, 100K, 330K, 1meg, 3.3meg, 10meg, and 33meg (made of 22meg in series with 11meg). The first five resistors are 2-watt units; the 100K unit is 1/2-watt, and the five higher-value ones are 1/4-watt.

The free ends of those resistors are tied together and go to the arm of S4B.

The even-numbered positions, 2, 4, 6, 8 and 10, have the following resistors added to them in this respective order: 1K, 10K, 100K, 1meg and 10meg. The first three are 1-watt resistors; the two high-value ones are 1/4-watt.

The free ends of those additional resistors are tied together and go to the downward position of S4B (the upward position of pole B is left open).

Sampling Resistors

All of these resistors have one end tied to plus 9 volts. Their free ends stand up and protrude through the lid of the box in order to be "sampled" for comparison.

They are arranged in groups of three; each group differs from its neighbors by a factor of 10.

  • Group 1: 10 ohms, 33 ohms, and 51 ohms.
  • Group 2: 100 ohms, 330 ohms, and 510 ohms.
  • Group 3: 1K, 3.3K, and 5.1K.
  • Group 4: 10K, 33K, and 51K.
  • Group 5: 100K, 330K, and 510K.
  • Group 6: 1meg, 3.3meg, and 5.1meg.
  • Group 7: 10meg, 33meg, and 51meg.

(The 33 meg is comprised of 22 meg in series with 11 meg; the 51 meg is made of two 22 meg units in series with 6.8meg.)

Construction Tips

I cut the circuit board so that it fits snugly between all four edges of the underside of the box's cover, with the corners of the board rounded out just enough to clear the corner posts. The board is mounted to the top cover with four bolts near the corners. Spacers under the top assure that there is room for the chip and socket, etc.

I chose to let the board carry all the components--jacks and switches. Corresponding holes in the top were made to gain access to these switches and jacks, as well as the sampling resistors.

To make sure that the holes in the top would match the locations of the controls, I then drilled 1/16-inch holes both through the board and the cover of the box every place I wanted to have a jack or switch. Then, the holes in the circuit board were drilled to the appropriate size for mounting the components. Holes in the cover can then be drilled and/or shaped so that switches can be moved properly and so that the bodies of the plugs can fit down through the top. (The required clearance hole for the input plug is 3/8 inch, while the hole for the earphone plug is 9/32 inch; moreover, this arrangement helps to protect the jacks and plugs should the box get dropped off the bench or out of a pocket while something is plugged into it.)

I found that mounting S1 and S2 along one side was best. I mounted the rotary switch on the midline near one end; S4 was then mounted in line with this. Near the end is also a good place to mount the toggle-type DIP switches instead of the rotary switch. Don't forget to leave enough room for the bunches of load resistors. The input jack, earphone jack and volume control are in line with each other across the opposite end of the box; the on-off switch is nearby.

I mounted the IC socket near the middle of the circuit board. I mounted the 50K trim pot near pin 5 of the IC socket, with its adjusting screw facing one side of the box.

If you are going to use the toggle-like DIP switches, there is a right way of doing things: Use high-profile sockets for the DIP switches and a low-profile one for the chip. Instead of just plain perforated board, choose the kind with individual pads around each hole, so that the switch sockets can be soldered in place. All circuit components must not stick up more than 1/16 of an inch above the switch sockets; otherwise the bodies of the DIP switches will not be able to stick up flush with the outside surface of the box's cover. Because of the foil pads, you will probably want to mount the switches and jacks on the top cover--either that, or the pads near each item should be removed.

In one of the Blue Boxes, I put the DIP switches into two side-by-side columns: One column had the 330 ohm, 3.3K, 33K, 330K, 3.3 meg and 33 meg resistors in it; the other column had pairs of the 1K, 10K, 100K, 1 meg and 10 meg resistors in it. With this arrangement, I could quickly select any single or parallel combination of load resistance that I wanted.

All along one side of the cover, the ends of the sampling resistors stick up through small holes. As mentioned earlier, these are in groups of three. I cut the middle one of each group of three slightly shorter so that I could short together any combination of adjacent long and short pins for the desired sample resistance that I might want.

Rather than depending on just the resistor leads themselves, I used some unusual and strong pins called "headers." Wire-wrap headers--or even individual wire-wrap pins--may be appealing because of their rigidity.


The biasing circuit must be adjusted so that any DC voltage on the input jack exceeds the "Threshold" required by comparators in the 7555. Make sure that the volume control is turned all the way up, and turn the Blue Box on.

Set S1 to the "Volts" position and S3 to the DC position. Short the probe leads together. The 50K trim pot is set to the point at which any oscillation just ceases. If it wasn't already singing or making some clicking sounds, turn the pot until it starts and then back again until the clicking just stops. On the other hand, you may find it preferable to leave the trim pot set so that there is always a clicking sound coming from the box. Set thusly, clicking will stop if you connect the probe leads to a voltage of reversed polarity. Also, the clicking will remind you that you have left the box turned on.

The box is now ready to go to work for you.

Calculating Resistor Combinations

For combinations of the load resistors, or for combining sampling resistors (by touching the probe to two or three at once), you need to calculate parallel resistances. It's a snap if you parallel two equal resistances; the result is one-half the value of the individual. It's still fairly easy if only two are involved; the result is the product of their individual resistances divided by their sum. Consider the following simple cases:

Suppose, with the load selector (either rotary or DIP-switch version) you have selected an "even-numbered" position--say position 2 (or switch 2 of the DIP's). With S4 in the upward position, this gives you only one resistor--1K (1000 ohms). With S4 in the downward position, you get two of these resistors in parallel, and since they are equal, the result is 500 ohms.

For this second case, throw S4 to the upward position. Now, if you have the shorting-type rotary switch set between positions 2 and 3, or if you close DIP switches 2 and 3, you are combining 1K and 3.3K. The result will be 3.3 over 4.3 (3.3 times 10 to the 6th ohms over 4300 ohms). (You can do it in ohms, or you can do it in K-ohms, but don't dare mix the two.) The answer is about 0.77K (about 770 ohms).

If you throw S4 to its downward position, you have three resistors in parallel: two 1K and one 3.3K. However, you know that combining the two 1K units amounts to 500 ohms. Therefore, you can treat the problem as if you had a 3.3K resistor in parallel with a 0.5K one. Taking the product over the sum, you have 1.65K over 3.8K, or 0.434K (about 430 ohms).

With your test probe on the sampling resistors, or if you throw more than two DIP switches on the load switch, you can have more than two resistors in parallel. In that case, you have to calculate the reciprocal of the sum of their conductances. However, what you immediately know is that the resultant combination is less than the smallest value in the group.

With S1 in the volts position, you can apply a load to any circuit from 200 ohms to over 44 megohms in any one of 35 different steps.

CAUTION: You must be careful when you switch in lower and lower values so as not to overheat the load resistors inside the Blue Box, or so as not to overload the circuit under test.

To select a 200-ohm load, set the rotary switch in between positions 1 and 2 with S4 in the down position; this places three resistors, two 1K and one 0.33K, in parallel, thus applying a combined resistance of about 0.2K to the input jack. If you switch S4 to its middle position, you disconnect all of the load resistors entirely.

All you have to remember is that the odd-numbered positions have "3"s in them; they start at 330 ohms at position 1, go to 3.3k at position 3, and so on until you get up to 33 megs at position 11.

Note that the sampling resistors are in groups of 3. This makes them easier to keep track of. These resistance values repeat themselves from one resistor group to another, with only the decimal point changing. Of course, if you don't want to bother remembering which combination gives you what resistance, you can build more samples into your box.

Using the Blue Box

Voltage Measurement

For safety's sake, check to see that you start each measurement with S4, the load-group selector, in its center-off position. Throw S1 to the "volts" position. The range you choose can be picked experimentally; start with the middle position, since the input impedance is so high here that the Blue Box won't be damaged. DC or AC is selected by S3 as desired.

When in this mode, the test lead which is connected to the sleeve of the input plug is negative, and the one going to the tip is positive.

The greater the voltage, the higher the pitch. It's as simple as that.

When I'm checking a circuit that seems dead and I'm wondering if there is power getting through to any part of it, I put the box into the volts mode and switch S4 to its center-off position. I then connect the negative probe to the ground of the circuit under test, hold the positive probe in one hand and touch different parts of the circuit with a free finger. Because of the box's extremely high input impedance, almost any voltage present will couple through my fingers into the Blue Box. Of course, this is only good if there aren't any voltages high enough to get a shock from.

A great enhancement to talking meters is to put the box into the volts function and connect its probes to the leads of your multimeter. You can now hear an instant indication of just what is going on. This is one of the most useful things you can do with the Blue Box, regardless of what kind of multimeter you have.

Signal Tracing

To trace audible frequencies or pulses in a circuit, switch S1 to its middle position. So that the audio won't be rectified, S3 should be thrown to the DC-volts position.

If you connect the probes to an audio output jack--like the one on a tape recorder, for example--you will notice that the audio coming out of the box is very distorted. This is because the 7555 takes all of those nice sounding audio wave forms and turns them into little square waves. Nevertheless, this audio-monitoring feature of the Blue Box is a good way of finding out if there is any audio at a given point.

Continuity and Resistance Measurement

This is done with S1 in the "continuity" or "ohms" position; you should also have S2 in the center or "medium-range" position and S3 in the DC-volts position.

When in this mode, the positive test lead is the one connected to the sleeve of the input plug, while the negative test lead is the one going to the tip. This is opposite polarity from that in the voltage mode; old technicians used to early VTVMs will remember this and have no trouble.

If you short the probe leads together, you will get a fairly high pitch tone from the box. If you switch S4 to its middle position and touch the probe leads with your fingers, you will see how little it takes to get the box to start sounding off.

Now start adding in load resistors. If you switch S4 to the up position and select a 10meg load, you will notice that you have to hang on to the leads a little harder to get the pitch high again. If you select the 330k load, you will have to squeeze the leads rather hard to get much of a tone at all. Now, with 330K still switched in, take the probe tip and run it down the row of sampling resistors. Notice how the pitch quickly changes from high to low as you sample from low to high resistance values.

To make better use of the samples, you have to remember that if you short any two of them together with the probe tip, you will have a sample resistance that is smaller than the smallest one you are touching. With a load of perhaps 100K switched in, let's look at the group with 10K, 33K and 51K in it. If you short the 10K and 51K together with the probe, you get 8.36K. Notice the tone you get. Now, if you just touch the 10K, notice that the tone is a little lower, and if you touch the 51K the tone is lower still.

To know pretty closely just how much resistance you have in a component or test circuit, use the sampling resistors until you get the same--or nearly the same--pitch from touching one or more of those as you do from the circuit under test. Use the load resistors and S4 to keep the box's tone somewhere in between nothing and the pitch you get from a dead short.

If you select 200 ohms as the load (between positions 1 and 2 with S4 down), you should be able to tell the difference between a 4-ohm speaker and a dead short.

CAUTION: When checking for resistance in a circuit with stiff loads (the load selector and S4 in their low-value positions), up to 45mA will flow through the circuit under test, so be careful. I recommend using higher resistance settings most of the time when checking for continuity so as not to hurt anything. (This also puts up to 45mA of load on the little 9-volt battery in the Blue Box, so don't leave the probes connected to something for long.)

In addition to measuring resistances, the continuity feature of the Blue Box is handy for testing other components. The capacitor and diode tests to follow are done with the tester in this mode.

Checking Small Capacitors

Set S4 to its middle position. Connect the probe leads to the capacitor and listen for a descending tone. Small caps of about 0.001uF should allow the pitch to fall to zero rather quickly. Thereafter, they produce a click of only about once a second or less; leaky caps will click several times a second or more, and you should consider throwing them out. Disc caps are almost always somewhat leaky, but are usually not bad. Larger caps, such as 0.1uF, will take a long time to charge; you may have to switch S4 up or down until the cap is charged. Then, switch S4 to center position again and listen for leakage.

Testing Electrolytics

First of all, the polarity of the probes has to be right or the capacitor will not take a full charge. In fact, a very handy feature of this tester is the ease with which the polarity of electrolytics can be identified. Note that the test probe going to the sleeve of the input plug is now positive.

Select a relatively high load resistance to start with and throw S4 to the up position. If the box's tone does not come down in pitch fast enough to suit you, lower the load resistance until you hear the tone coming down and going to zero in a few seconds. If the pitch of the tone drops fairly rapidly at first, but then stops dropping or starts to come back up again, you have the electrolytic hooked up backwards. Don't leave it hooked up backwards; in fact, do it the immediate favor of charging up the right way, since leaving a reverse charge on it could damage it or shorten its life.

If it doesn't accept a full charge in either direction, make sure that its rated voltage is at least 9V. Charging time of all electrolytics should take about 4 to 8 seconds to check for proper polarity; this time is set by experimenting with load resistances.

About the same time as you hear the tone get to zero, switch S4 to the middle position and listen for the clicks to start increasing as the electrolytic leaks off. All electrolytics are more or less leaky, but if the clicks increase to a low pitch tone within a few seconds, it is very leaky and should be tossed out.

Checking Diodes

Of course, you can check the polarity of any PN junction in solid-state devices. If S4 is in the center-off position, the leakage of any diode can be detected. The box will not stop clicking until the resistance across the leads exceeds several thousand megohms.

If you connect a small diode that is enveloped in a clear material in reverse and shine a bright light on it, you will be able to hear the diode's leakage increase as you shine the light on it.

Resistance Substitution Box

Whether the Blue Box is turned on or off, the load resistors appear across the input when the "mode switch" is in the volts position. S3 has to be in the DC position as well. Finally, the charging network of the 7555 will be the least invasive when S2 is in the mid-position ("medium-range"). Once you get below 3.3megs or so, the influence of the Blue Box circuitry will be insignificant.

The precaution mentioned earlier has to be followed: Carbon-composition resistors are good for half of their power rating. At best, a combination of the even-numbered resistances is rated at 3 watts; you should limit the power dissipation in this combination to 1-1/2 watts. The lower-value individual resistors are rated at 2 watts; you should limit dissipation in these to 1 watt. The formulas for power are: the square of the current times the resistance; or, the square of the voltage divided by the resistance.

Parts List


  • 1--47uF 10V electrolytic
  • 1--10uF 10V electrolytic
  • 1--0.47uF 500V disc ceramic
  • 1--0.47uF 1,000V disc ceramic
  • 1--0.1uF 1,000V disc ceramic
  • 1--0.1uF 25V disc
  • 1--0.1uF 16V disc
  • 1--100pf low-voltage silvered-mica or glass

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

  • 1--10 ohm
  • 1--33 ohm
  • 1--51 ohm
  • 1--100 ohm
  • 1--330 ohm
  • 1--330 ohm (2-watt)
  • 1--510 ohm
  • 1--1K
  • 2--1K (2-watt)
  • 1--3.3K
  • 1--3.3K (2-watt)
  • 1--4.7K
  • 1--5.1K
  • 1--10K
  • 2--10K (2-watt)
  • 1--33K
  • 1--33K (2-watt)
  • 1--51K
  • 3--100K
  • 2--100K (1/2-watt)
  • 1--330K
  • 1--330K (1/2-watt)
  • 2--470K
  • 1--510K
  • 1--680K
  • 3--1 megohm
  • 1--2.2 megohm
  • 2--3.3 megohm
  • 2--6.8 megohm
  • 3--10 megohm
  • 1--11 megohm (12 megohm will do)
  • 6--22 megohm (available only as 1/2-watt units)


  • 1--50K PC-mount trim pot
  • 1--25K chassis-mount volume control
  • Switches:
  • 1--SPST on-off switch
  • 1--Load-Resistor Switch, 11 mini-toggles, 11 DIP switches, or an 11-position rotary
  • 1--DPDT without center-off
  • 2--DPDT with center-off
  • 1--3-Pole double-throw toggle with center-off


  • 1--1,000 PIV low-current rectifier diode
  • 1--Intersil ICM7555

Earphone and Speaker:

  • 1--high-impedance earphone with sub-mini phone plug
  • 1--Piezo-electric Transducer, Radio Shack 273-073 (Small PM speakers won't do. Moreover, if a substitute piezo-electric unit is found, make sure that it is of the 2-lead type and has no internal electronic driver of its own.)

Miscellaneous Hardware:

  • 1--8-pin low-profile DIP socket for the 7555
  • 2--high-profile sockets for DIP switches (if you choose to use these for the load resistors)
  • 1--Open-circuit mini-phone jack
  • 1--Closed-circuit sub-mini phone jack
  • 1--set of test leads attached to a 1/8-inch mini phone plug
  • 1--9V battery connector (perhaps built into a holder, such as the Bopla DE30)
  • 1--9-volt alkaline battery
  • 4--bolts, nuts, and appropriate spacers
  • 1--Bluer-than-blue box, such as the Bopla "Project Box" E430, the Unibox 120 or P/N130, or the Radio Shack 270-222.

Note: the Bopla P/N130 has a battery compartment, but it is too thin to accommodate a rotary switch.

Documentation from the Author

A standard size print copy of this article, which includes a print diagram, is available for $2.50. (A print diagram and parts list only is $1.) An IBM 360K floppy disk copy in ASCII format is $1; the 1.2 megabyte floppy disk version is $1.50.

If you wish to buy one pre-assembled, this circuit-analyzer unit is also available for about $60 for the rotary switch version and about $70 for the one that uses mini toggle switches in place of the rotary switch.

[Editor's Note: Those items are available from Bernie Vinther, not from Smith-Kettlewell. Make those checks payable to him.]

Well, I guess this is more than enough for now.

As always, Bernie Vinther, 915 West Grand Ronde, Kennewick, WA 99336, Phone (509) 586-8060.


by Tom and Susan Fowle


This is a "doorbell" which produces a soft note that gets softer the longer the button is pressed. Means of connecting it to existing doorbell circuits without damage to rented houses are discussed.

When we moved into our rented duplex a number of years ago, the landlord had included a new electric doorbell with a recent remodeling effort. This consisted of the usual button outside, hidden transformer and wiring, and a small round innocent-looking buzzer mounted high on the wall near the door. The first press of this new high-tech feature created a noise reminiscent of the warning used at a nuclear weapons test in the vastnesses of the Nevada desert, not a mere call to the door. After being blasted into near hysteria several times (only to find that the mail had arrived), we decided that this acoustic atrocity just wouldn't do, and came up with this little circuit.

The new "bell" was mounted in a small plywood box and hung by a loop of wire from a drapery rod near the offending buzzer. A switch on the side of the box allows the power to be moved between the new circuit, where it almost always remains, and the old buzzer, in case we ever go deaf. There is a center off position for total isolation too--how very convenient on Sunday mornings.

Once the top of the buzzer is removed, one wire is unfastened from its screw terminal. This wire is then twisted with the end of a lead that comes from the arm of the SPDT switch in our circuit, and the connection is then wrapped with tape. Another lead runs from that screw terminal over to the "buzzer" position of our switch. The other terminal of the buzzer gets a wire added to it for the "common side of the power."

These circuits are almost always AC powered, and usually run from 10, 16 or 24 volts RMS. It is a good idea to check this with a voltmeter before proceeding. (I have never heard of a 110-volt doorbell, but who knows.) However. (Remember that little fingers around unknown circuits might get bit when the button is pressed because of inductive spikes from the buzzer or bell.)

Circuit Operation

Because a wide range of voltages may be encountered, a 15-volt regulator has been included as a "limiter" for VCC. A 10-volt doorbell system will not power this regulator correctly, but the circuit still works. At the other extreme, as long as the filtered output of the bridge rectifier is under 40 volts, the regulator will take care of the circuit.

The circuit uses an NE555 oscillator set for a nice low tone of 250 Hertz. The trick is: The output of the 555 goes through a 47-ohm resistor in series with the speaker to the anode of a 1N4001 diode; the cathode of this diode goes to the positive end of a fairly large electrolytic capacitor whose negative is grounded. Since the diode permits current to flow in only one direction, a charge accumulates on the capacitor which decreases the voltage swing seen by the series combination of the loudspeaker and resistor. Eventually, the voltage at the positive end of the capacitor approaches that of the positive swing of the oscillator's output and the sound fades away.

Another diode is connected with its anode at the positive of the output cap and its cathode going through a "bleeder resistor" to VCC. This diode will become forward-biased when the power is removed; this allows the output cap to be refreshed. The time taken to refresh the Pest-Bell will depend on the value of the bleeder resistor.

Thus, the longer the circuit is on, the softer it is; even if the bill collector keeps pushing the button, you won't be disturbed more. In our unit, the button must be released for fifteen seconds before the capacitor discharges and the tone returns to its original volume.


The wire removed from one terminal of the original buzzer goes to the arm of a single-pole double-throw switch (which may have a center-off position). The "buzzer" position of the switch is returned to the recently vacated screw terminal of the buzzer.

The "Pest-Bell" position of the switch goes to one input of a bridge rectifier whose other input is connected to the other terminal of the buzzer, the "common side of the power.

The negative output of the bridge is our circuit ground. The positive bridge output is bypassed to ground by 1000uF (negative at ground).

A 15-volt regulator (such as the Fairchild uA7815 or the National LM340-15T) has its "Common" terminal grounded (the "ground" which is the negative side of the bridge rectifier). The regulator's "Input" terminal goes to the positive output of the bridge. The 15V "Output" goes to the VCC line, and is bypassed by 220uF (negative of this cap at ground).

The NE555 has pin 1 grounded, while pins 4 and 8 go to VCC. Pins 2 and 6 are tied together and go through 0.01uF to ground. Pins 2 and 6 also go through 270K to pin 7. Pin 7 goes through 10K to VCC.

Pin 3, output, goes through a 47-ohm 1/2 watt resistor, then through the speaker to the anode of a 1N4001 diode. The cathode of this diode goes to the positive of a 2200uF electrolytic whose negative is grounded. The junction of the positive of this cap and the diode goes to the anode of a second 1N4001 diode; the cathode of this goes through a resistor (the bleeder, perhaps 4.7K) to VCC.

Good luck with this one. It sure makes our house more livable, and can be removed in minutes should we ever move.


by Jay Williams


There isn't a modern convenience which doesn't also provide its own brand of static. The extension telephone is no exception. The device described in this article minimizes the 20th-Century chaos by providing an audible indication that other householders have finally gotten off the phone. The Grabophone eliminates the need to inform other users that the line is free. It also "dings"--only once--when the phone rings. A third bonus with this device is that you can get some idea of the frequency and length of the other users' phone activities without eavesdropping. It does all this by electrically sensing activity on the phone line.

How often do you pick up the phone and, upon hearing an ongoing conversation, realize that you must gracefully replace the telephone receiver. (This can be accomplished easily with the "standard" black units which have been around for years. It is quite another matter to gracefully replace the receiver on the newer "designer" phones which fit together like the two halves of a puzzle.)

Design Considerations

I wanted an auditory indication which was both distinct and unobtrusive. Tom Fowle's "doorbell" circuit solves the problem nicely by providing an NE555 timer chip with an automatic gain reduction system and a pleasant "clarinet-like" tone. To conserve battery drain, the activity of the phone line activates an N channel VMOS power FET (see "Save Your Batteries," SKTF, Fall, 1983) which, in conjunction with the RC network on the input, provides power to the oscillator circuit only for the amount of time required for the tone to fade into inaudibility.

The builder's main concern is to account for the voltage changes on the telephone line. When the phone is in use, the line voltage is around 5 to 6 volts. When the phone is disengaged the line voltage rises to 48 volts. I chose to couple the input to the device capacitively to avoid undesirable DC paths and signal deterioration. The input capacitor should have a voltage rating of at least 150 volts in order to withstand the 80-volt AC ringer signal.

The capacitor feeds a 4-to-1 voltage divider so that the power FET and its clamping diodes never see any of those brutal ringing transients. Incidentally, if you wish to experiment with different values in the input network while the circuit is on-line, you would do well to keep your fingers out of the circuit while the phone is disengaged. That ringer voltage provides quite a kick.

The pitch of the "ding" can be customized by choosing the timing capacitor off pins 2 and 6 of the 555. The value of this cap should be determined by the user's taste. I found that a 0.005uF unit gave a comfortably resonant tone without being too piercing.

Circuit Description

The positive side of the telephone line goes to the plus lead of a 10uF to 22uF 150-volt electrolytic capacitor. The negative capacitor lead goes through 680K, then through 160K to the negative side of the phone line and to the source of the N-channel power FET (Siliconix VN0300M or similar unit). The source of the FET also goes to the negative lead of a 9-volt battery whose positive lead goes to the VCC line of the oscillator circuit. The circuit's minus 9-volt line, "Circuit common," goes to the drain of the FET. The VCC line is bypassed to circuit common through a 1000uF electrolytic capacitor, its negative lead at circuit common.

The junction of the 680K and the 160K resistors goes to the gate of the FET. This gate goes to the anode of a diode whose cathode goes to the VCC line. The gate also goes to the cathode of another diode whose anode goes to the source of the FET.

Pin 1 of the 555 goes to circuit common; pins 4 and 8 are tied together and go to VCC. Pins 2 and 6 are tied together and go through 270K in series with 10K to VCC. The junction of these two resistors goes to pin 7. Pins 2 and 6 also go through a capacitor to circuit common (0.0047uF or 0.01uF).

The oscillator's output, pin 3, goes through 47 ohms (1/2 watt), then through the speaker to the anode of a 1N4001 diode. The cathode of this diode goes to the positive lead of a 2200uF capacitor whose negative lead is grounded. The diode-cap junction goes to the anode of another 1N4001 diode; the cathode goes through 4.7K to plus 9 volts.

Installation Tips

I cut the plug off one end of a store-bought telephone connecting cord and soldered its bared leads into the Grabophone circuit. A Y connector was then used which accepts the cord connecting the phone and the cord connecting to the Grabophone. If you choose this method, you will have to grapple with the infamous "tinsel" leads in these cords. I took Bill Gerrey's advice: After gingerly stripping tinsel wire so as not to pulverize it, wrap the stripped end with a strand borrowed from a more conventional stranded wire. Tin this mixture quickly. Strip all four wires within the cord and look at them with a voltmeter until you encounter the expected voltage; then remove the two unwanted leads.

If you choose to install a 2-wire cord on terminals within the phone, remember that the network box inside is not easily memorized. In addition, there is likely to be more than one lug on any one of the screw terminals, so, when you have ascertained which terminals are used for connection to the outside world, work carefully. You really don't need the frustration of trying to make the phone work right by holding an errant lug to this or that terminal.

Parts List

Resistors (1/4-watt 5%):

  • 1--680K
  • 1--270K
  • 1--160K
  • 1--10K
  • 1--4.7K
  • 1--47 ohm 1/2-watt


  • 1--2200uF 16V electrolytic
  • 1--1000uF 16V electrolytic
  • 1--10uF or 22uF 150V electrolytic
  • 1--0.0047uF to 0.01uF low-voltage Mylar or disc


  • 4--1N4001 diodes
  • 1--Siliconix VN0300M or VN10KM N-channel power FET (Radio Shack 276-2073)
  • 1--555 timer chip


  • 1--8-ohm loudspeaker
  • 1--piece of perforated board, 2 inches by 3 inches
  • 1--9-volt battery connector
  • Cabinet and mounting hardware


This is from a letter dated November 20, 1986 from Allen Ward, K0BDD, of Cannon Falls, Minnesota.

"I wonder if you have read the book entitled 'The Chip -- How Two Americans Invented the Microchip and Launched a Revolution.' I found it to be a very interesting, non-technical account of the development of modern electronics, and a book which I believe a great many of your Smith-Kettlewell readers would find interesting. It is Library of Congress Cassette No. RC-23393."

* * *

From Bernie Vinther:

Mouser Electronics now has a toll-free number: (800) 346-6873. They also sell a high-impedance earphone (10K ohms) which works well with the Blue Box. They can be gotten with either mini or sub-mini phone plugs on them.

My favorite place to buy parts by mail is a place called Round-Up Electronics in Pendleton, Oregon. Their number is (503) 276-3152.

And still another good place with nationwide distribution is Active Electronics. They have super-good prices on CMOS 555 timers and carry just about everything that the Vector company makes. Their number is (800) 343-0874.

I will sell parts at cost and ship them in padded envelopes. (The editor suggests you call Bernie for prices.)

Finally, unedited versions of my articles are available from me on IBM diskettes; $1 for 360K floppies and $1.50 for 1.2 megabyte ones.

Bernie Vinther, 915 West Grand Ronde, Kennewick, WA 99336,Phone (509) 586-8060