Silicon ChipVersatile Waveform Generator - February 2026 SILICON CHIP
  1. Contents
  2. Publisher's Letter: Waiting for the AI bubble to burst
  3. Feature: Max’s Cool Beans by Max the Magnificent
  4. Project: Power LCR Meter, part one by Phil Prosser
  5. Feature: Teach-In 2026 by Mike Tooley
  6. Project: Versatile Waveform Generator by Randy Keenan
  7. Feature: Circuit Surgery by Ian Bell
  8. Review: Raspberry Pi Pico 2 microcontroller module by Tim Blythman
  9. Project: The PicoMite 2 by Geoff Graham
  10. Feature: Techno Talk by Max the Magnificent
  11. Feature: Audio Out by Jake Rothman
  12. Review: Mini UPS modul by Jim Rowe
  13. Feature: The Fox Report by Barry Fox
  14. Back Issues
  15. PartShop
  16. Advertising Index
  17. Market Centre
  18. Back Issues

This is only a preview of the February 2026 issue of Practical Electronics.

You can view 0 of the 80 pages in the full issue.

Articles in this series:
  • Max’s Cool Beans (January 2025)
  • Max’s Cool Beans (February 2025)
  • Max’s Cool Beans (March 2025)
  • Max’s Cool Beans (April 2025)
  • Max’s Cool Beans (May 2025)
  • Max’s Cool Beans (June 2025)
  • Max’s Cool Beans (July 2025)
  • Max’s Cool Beans (August 2025)
  • Max’s Cool Beans (September 2025)
  • Max’s Cool Beans: Weird & Wonderful Arduino Projects (October 2025)
  • Max’s Cool Beans (November 2025)
  • Max’s Cool Beans (December 2025)
  • Max’s Cool Beans (January 2026)
  • Max’s Cool Beans (February 2026)
Items relevant to "Power LCR Meter, part one":
  • Power LCR Meter PCB [04103251] (AUD $10.00)
  • PIC32MK0128MCA048 programmed for the Power LCR Meter [0410325A.HEX] (Programmed Microcontroller, AUD $20.00)
  • Software & STL files for the Power LCR Tester (Free)
  • Power LCR Meter PCB pattern (PDF download) [04103251] (Free)
  • Power LCR Meter panel artwork and drilling diagrams (Free)
Articles in this series:
  • Power LCR Tester, Part 1 (March 2025)
  • Power LCR Meter, Part 2 (April 2025)
  • Power LCR Meter, part one (February 2026)
Articles in this series:
  • Teach-In 12.1 (November 2025)
  • Teach-In 2026 (December 2025)
  • Teach-In 2026 (January 2026)
  • Teach-In 2026 (February 2026)
Articles in this series:
  • STEWART OF READING (April 2024)
  • Circuit Surgery (April 2024)
  • Circuit Surgery (May 2024)
  • Circuit Surgery (June 2024)
  • Circuit Surgery (July 2024)
  • Circuit Surgery (August 2024)
  • Circuit Surgery (September 2024)
  • Circuit Surgery (October 2024)
  • Circuit Surgery (November 2024)
  • Circuit Surgery (December 2024)
  • Circuit Surgery (January 2025)
  • Circuit Surgery (February 2025)
  • Circuit Surgery (March 2025)
  • Circuit Surgery (April 2025)
  • Circuit Surgery (May 2025)
  • Circuit Surgery (June 2025)
  • Circuit Surgery (July 2025)
  • Circuit Surgery (August 2025)
  • Circuit Surgery (September 2025)
  • Circuit Surgery (October 2025)
  • Circuit Surgery (November 2025)
  • Circuit Surgery (December 2025)
  • Circuit Surgery (January 2026)
  • Circuit Surgery (February 2026)
Articles in this series:
  • Techno Talk (February 2020)
  • Techno Talk (March 2020)
  • (April 2020)
  • Techno Talk (May 2020)
  • Techno Talk (June 2020)
  • Techno Talk (July 2020)
  • Techno Talk (August 2020)
  • Techno Talk (September 2020)
  • Techno Talk (October 2020)
  • (November 2020)
  • Techno Talk (December 2020)
  • Techno Talk (January 2021)
  • Techno Talk (February 2021)
  • Techno Talk (March 2021)
  • Techno Talk (April 2021)
  • Techno Talk (May 2021)
  • Techno Talk (June 2021)
  • Techno Talk (July 2021)
  • Techno Talk (August 2021)
  • Techno Talk (September 2021)
  • Techno Talk (October 2021)
  • Techno Talk (November 2021)
  • Techno Talk (December 2021)
  • Communing with nature (January 2022)
  • Should we be worried? (February 2022)
  • How resilient is your lifeline? (March 2022)
  • Go eco, get ethical! (April 2022)
  • From nano to bio (May 2022)
  • Positivity follows the gloom (June 2022)
  • Mixed menu (July 2022)
  • Time for a total rethink? (August 2022)
  • What’s in a name? (September 2022)
  • Forget leaves on the line! (October 2022)
  • Giant Boost for Batteries (December 2022)
  • Raudive Voices Revisited (January 2023)
  • A thousand words (February 2023)
  • It’s handover time (March 2023)
  • AI, Robots, Horticulture and Agriculture (April 2023)
  • Prophecy can be perplexing (May 2023)
  • Technology comes in different shapes and sizes (June 2023)
  • AI and robots – what could possibly go wrong? (July 2023)
  • How long until we’re all out of work? (August 2023)
  • We both have truths, are mine the same as yours? (September 2023)
  • Holy Spheres, Batman! (October 2023)
  • Where’s my pneumatic car? (November 2023)
  • Good grief! (December 2023)
  • Cheeky chiplets (January 2024)
  • Cheeky chiplets (February 2024)
  • The Wibbly-Wobbly World of Quantum (March 2024)
  • Techno Talk - Wait! What? Really? (April 2024)
  • Techno Talk - One step closer to a dystopian abyss? (May 2024)
  • Techno Talk - Program that! (June 2024)
  • Techno Talk (July 2024)
  • Techno Talk - That makes so much sense! (August 2024)
  • Techno Talk - I don’t want to be a Norbert... (September 2024)
  • Techno Talk - Sticking the landing (October 2024)
  • Techno Talk (November 2024)
  • Techno Talk (December 2024)
  • Techno Talk (January 2025)
  • Techno Talk (February 2025)
  • Techno Talk (March 2025)
  • Techno Talk (April 2025)
  • Techno Talk (May 2025)
  • Techno Talk (June 2025)
  • Techno Talk (July 2025)
  • Techno Talk (August 2025)
  • Techno Talk (October 2025)
  • Techno Talk (November 2025)
  • Techno Talk (December 2025)
  • Techno Talk (January 2026)
  • Techno Talk (February 2026)
Articles in this series:
  • Audio Out (January 2024)
  • Audio Out (February 2024)
  • AUDIO OUT (April 2024)
  • Audio Out (May 2024)
  • Audio Out (June 2024)
  • Audio Out (July 2024)
  • Audio Out (August 2024)
  • Audio Out (September 2024)
  • Audio Out (October 2024)
  • Audio Out (March 2025)
  • Audio Out (April 2025)
  • Audio Out (May 2025)
  • Audio Out (June 2025)
  • Audio Out (July 2025)
  • Audio Out (August 2025)
  • Audio Out (September 2025)
  • Audio Out (October 2025)
  • Audio Out (November 2025)
  • Audio Out (December 2025)
  • Audio Out (January 2026)
  • Audio Out (February 2026)
Articles in this series:
  • The Fox Report (July 2024)
  • The Fox Report (September 2024)
  • The Fox Report (October 2024)
  • The Fox Report (November 2024)
  • The Fox Report (December 2024)
  • The Fox Report (January 2025)
  • The Fox Report (February 2025)
  • The Fox Report (March 2025)
  • The Fox Report (April 2025)
  • The Fox Report (May 2025)
  • The Fox Report (July 2025)
  • The Fox Report (August 2025)
  • The Fox Report (September 2025)
  • The Fox Report (October 2025)
  • The Fox Report (October 2025)
  • The Fox Report (December 2025)
  • The Fox Report (January 2026)
  • The Fox Report (February 2026)
Constructional Project Project by Randy Keenan Versatile Waveform Generator This versatile waveform generator (also known as a function generator) is handy for a variety of uses, including audio equipment analysis, circuit development, displays and demonstrations and as a pulse source for developing switching and motor controls. It uses three op amps to deliver square, pulse, triangle, ramp and sine waves from 1Hz to 30kHz. W aveform generators are often built around specialised ICs, such as the Exar XR2206, Intersil 8038 or the Maxim MAX038. However, I wanted to make a waveform generator using only generic components, like op amps, with these features: ∎ Output frequencies covering the audio range and more, from 1Hz to 30kHz. ∎ Waveform outputs of: a. square/pulse, variable from 5% to 95% duty cycle, or wider b. triangle/ramp/sawtooth, variable from positive to negative ramps c. sinewave with a total harmonic distortion (THD) of around 1% ∎ Duty cycle/symmetry adjustments do not alter the frequency or amplitude appreciably ∎ Output amplitudes of the three waveforms can be matched, peak or RMS, from 0V to 6V peak-to-peak. ∎ An output impedance less than 200W. ∎ Battery-powered for portability and isolation. ∎ Compact size. The design presented here is the result. It uses three op amps, two voltage regulators, six diodes, plus passive components. If any of the specified ICs become scarce, others of the same or better specifications could be substituted. Operating principle The circuit needs to generate the three basic types of waveform: square/ pulse, triangle/ramp and sine. Since producing triangle/ramps and sin- ewaves from a pulse is complicated, the design begins with an op amp integrator producing a repeating triangle/ramp waveform. Referring to the block diagram, Fig.1, the integrator at left produces the triangle/ramp waveform, with its frequency range set by switching in one of nine different integrator capacitor values. The triangle/ramp waveform is fed to a comparator that turns it into a square/pulse waveform, which is then fed back via the frequency adjustment pot to ensure oscillation. This gives us the triangle and ramp waveforms. The two diodes and symmetry adjustment pot allow the positive and negative ramp rates to be varied to give square/pulse output waveforms. Modifying (shaping) the triangular waveform by a separate circuit section converts it into a sine shape. While the result is not a perfect sinewave, it’s pretty close, as demonstrated by its relatively low distortion/THD figure of about 1%. The waveforms are selected by the middle switch, buffered and level-­ adjusted by IC3, and then fed to the outputs. Circuit details Fig.1: the Waveform Generator is designed around three op amps. IC1 is configured as an integrator and its output feeds into IC2, acting as a comparator, which feeds back into IC1. This feedback loop causes both to oscillate, with IC1 generating a triangular or sawtooth waveform and IC2 producing a square or pulse wave. A triangle-to-sinewave shaper produces the third waveform option. The full circuit is shown in Fig.2. The heart of the circuit is the integrator composed of op amp IC1. It uses capacitors as the timing element and switched frequency range switch S1. 34 Practical Electronics | February | 2026 Versatile Waveform Generator Fig.2: the complete Waveform Generator circuit. S1 selects between nine possible frequency ranges by switching a different amount of capacitance across the integrator (IC1). Switch S2 is used to choose the desired waveform; its level is adjusted using VR5, then it is buffered by IC3 and fed to two pairs of outputs, one set DC-coupled and the other AC-coupled. The capacitor is charged and discharged via pot VR8, trimpots VR9 & VR10 and diodes D5 and D6. It works as follows. Assume that initially the timing capacitor is discharged, and it is being charged by a current to pin 4 of IC1 through D6. IC1’s output will be a linear negative-going ramp to counteract the increasing charge of the capacitor. The integration needs to be stopped at some point, so the op amp output is fed to a second op amp, IC2, configured as a comparator with hysteresis. When IC1’s output reaches the lower hysteresis voltage, set by trimpot VR7 and associated components, the comparator is triggered and its output goes negative, which is fed back to IC1’s input via potentiometers VR10, VR9, VR8 and D5, which is now forward-­ biased. This causes the timing capacitor to start discharging, resulting in Practical Electronics | February | 2026 a positive-going linear output ramp from IC1. This continues until IC1’s output reaches the upper hysteresis voltage of the comparator, and the output of IC2 switches again, producing a negative-going ramp from IC1. Thus, the process of charging and discharging of the timing capacitor and switching of IC2’s output continues indefinitely to produce an upward and downward ramp, plus a coincident square wave from the output of IC2. Varying the duty cycle/ symmetry The upward and downward slopes of the triangle or ramp are determined by the charging and discharging currents through the two arms of VR8. If VR8 is at its midpoint, the slopes are equal and a triangular wave is produced. If VR8 is off-centre, the currents through D5 and D6 are un- equal, and a sawtooth waveform is produced. Since the sum of the resistances to D5 and D6 and to IC1 is the same at any setting of VR8—equal to the total resistance of VR8—the period of the ramp, or triangle, will be constant regardless of its shape. (This is not quite true because of the non-ideal schottky diode characteristics and non-ideal characteristics of VR8, but it’s pretty close.) The setting of VR8 also determines the duty-cycle of the square wave/ pulse from IC2, since it depends on the periods of the upward and downward triangle wave ramps. To vary the frequency, the square/ pulse output voltages from IC2 are adjusted by VR10 over a range of approximately 3:1. I chose this range to allow for precise setting of the frequency and to reduce non-ideal effects of the components. 35 Constructional Project To cover a wide range of frequencies, a series of nine charging/timing capacitors can be selected by rotary switch S1, as shown in Table 1. Note that there is a 330pF capacitor always connected between pins 1 & 4 of IC1, and this is the only timing capacitor that is used on the highest (10-30kHz) range. It also adds to the switched-in capacitances on the 3-10kHz and 1-3kHz ranges, but for lower frequency ranges, its value is too small to have any real effect. To obtain a precisely symmetric triangle or 50% duty-cycle square wave, the potentiometer’s centre detent has to be pretty close to the point where the resistance from the wiper to each end of the track is identical. I have found that for a typical pot, the resistances of the two arms are not equal when set at the detent; furthermore, the detent generally has some ‘wobble’. Also, PCB-mounting potentiometers with a centre detent are not readily available. So, to ensure a symmetric waveform, the S3 “Symmetry” switch can be switched to its “50%” position, engaging VR11 and its 43kW series resistor for equal charging and discharging currents, and thus a fixed 50% symmetry. In the other position, S3 enables variable symmetry, as described earlier. Table 1 – Timing capacitors S1 Freq. range Capacitance 1 1-3Hz 3.3μF 2 3-10Hz 1μF 3 10-30Hz 330nF 4 30-100Hz 100nF 5 100-300Hz 33nF 6 300Hz-1kHz 10nF 7 1-3kHz 3nF or 3.3nF * 8 3-10kHz 2 × 330pF 9 10-30kHz 330pF * 3.3nF might make the 1-3kHz band too low in frequency Table 2 – Li-ion battery options Type & size Voltage Capacity 6F22, “9V” ~8V (use two) 6001300mAh 10440 ~3.7V 350(use four) 1000mAh 14200/ 14250 ~3.7V ~300mAh (use four) 14500 ~3.7V 800(use four) 2500mAh 36 The final task is to produce a sinewave, and the method must work over the entire frequency range of the generator. In other words, it must be frequency-­independent from 1Hz to 30kHz. This requires some non-­linear circuit elements. There are various methods, but I chose a simple one. Feeding the triangle wave to four diodes—two for positive and two for negative, plus a couple resistors—can reasonably approximate a sinewave. These diodes (D1-D4) should be closely matched, ideally from a single order and adjacent on a tape. This technique will never achieve a perfect sinewave, but it can come close (see Scope 3). The waveforms square/pulse, triangle/ramp, and sine are selected by S2 and then buffered by op amp IC3 before being sent to the output terminals. Both direct and capacitor-­isolated outputs are provided. S2 is arranged with a pattern of square, off, triangle, off, sine for two reasons. Firstly, it provides some isolation among the waveforms, and secondly, having an off position or positions can be handy during use. Because the sinewave from the shaper has the lowest amplitude of the three waves, the output op amp gain is adjusted, via trimpot VR2 (“Sine”), to accommodate the sinewave. Then the square/pulse and triangle/ramp amplitudes can then be adjusted via trimpots VR3 (“Tri”) and VR6 (“Sq”). The wave amplitudes may be adjusted to either have equal peak amplitudes or equal RMS amplitudes, as desired. One reason for choosing equal RMS (root-mean-square) voltages is that each of the waveforms would deliver the same power to the load at the same setting. Power supply I wanted the waveform generator to be battery-powered for easy portability as well as electrical isolation. The two batteries need sufficient voltage for the 5V voltage regulators (REG1 and REG2), meaning about 5.5V minimum, and preferably 7-8V. The specified batteries are “9V” (actually about 8V) lithium-ion rechargeable types. Alternative rechargeable lithium-­ ion batteries are listed in Table 2, but check the capacity. I don’t recommend using 14500 (AA-size) cells, as four are required, in two holders, and it’s difficult to fit those into the specified enclosure. Compared to 78L05 & 79L05 voltage regulators, the ADP3300-5.0s have a much lower dropout voltage and lower quiescent current use for lower battery drain. They also have the ability to drive dropout LED indicators (LED1 and LED2 in this circuit) and provide a more accurate regulated voltage. The specified LEDs are high-­ brightness types for operation at low current and thus lower battery drain. The more accurate voltages, coupled with low-input-offset voltage op amps, reduces the need for compensation-­ adjustment circuitry. The ADP3300-5.0 is used for both the positive (IC4) and negative (IC5) voltage regulation. Thus, the batteries do not have a common connection. If you use USB-rechargeable batteries with a double charging cable, be sure to remove the USB cables from the batteries before switching on the Waveform Generator as the circuit does not have a common battery connection, whereas the USB charging cables do have a common battery connection. The current drawn from each battery is about 18mA each polarity, depending slightly on the frequency and waveform. Thus, the “9V” 600mAh batteries should provide about 20 hours (or more) of operation per charge, as confirmed by my trials, or twice as long for 1200mAh batteries. A 220W load increases the current up to 26mA for a square wave output at 6V peak-to-peak, or several milliamperes lower for the other waveforms. Part choices/variations Two different parts are specified in the parts list for VR8, the Symmetry adjustment potentiometer. The P0915N version is better as it results in smaller frequency shifts at the extremes of symmetry/duty cycle, on the order of about 1-2%. Using the PTV09 version will probably result in larger frequency shifts. However, if using the (better) P0915N version, its terminals will need to be reformed or trimmed and the two projections on the bottom— not the mounting tabs—will need to be removed so the pot will sit directly on the PCB. Since its shaft is smooth, you can drill out a knurled knob for a clean fit. Practical Electronics | February | 2026 Versatile Waveform Generator Photos 1 & 2: this PCB was assembled with the five SMDs on adaptor boards. Note how the miniature banana sockets on the right are soldered to the pads on the top of the PCB. I glued the 9V rechargeable batteries to the bottom of the enclosure and connected them to the PCB using standard battery snaps. Unfortunately, potentiometers typically have a resistance tolerance of ±20%. Consequently, the values of some resistors may need to change depending on the actual resistance of the pots you get. 1. VR8’s nominal value is 100kW. If yours measures above 100kW or below 92kW, you should ideally change the value of the 43kW resistor. Halve the measured value of VR8 and subtract 5kW, then pick the closest available value to use in place of the 43kW resistor. 2. VR10’s nominal value is 1kW. If its value is below 935W or above 1.03kW, you should ideally change the value of the 390W resistor. Multiply VR10’s actual resistance by 0.4 and then pick the closest available value to use in place of the 390W resistor. A good alternative combination of op amps is AD8065 for IC1, either AD8051 or AD8091 for IC2, and AD8033 or AD8065 for IC3 (the AD8033 comes in a smaller package than the others, so will be more tricky to solder). For the five surface-mount ICs, there are two mounting techniques: (a) directly on the PCB as surface mount, or (b) using adaptor boards with pins and receptacles. The main advantage of using adaptor boards is that you can unplug the ICs for testing and it’s easy to replace them later (eg, for experimentation). If you decide to use the adaptor boards, you can prepare them by first inserting five pins, long end down, in the appropriate pattern into a stably mounted DIL socket – see Photo 3. Then place an adaptor board, with the surface-mount pads upward, onto the pins and solder each pin (Photo 4). With the pins attached, solder the IC to the pads using your preferred technique. There are a few ways to do it, either with a regular iron or hot air; the construction procedure below goes over our preferred method. Make sure that the orientation of the IC is correct (see Photo 5). For the op amp ICs, finding the correct orientation is straightforward— they only have five leads. For the regulators, it’s a bit more tricky as they are rotationally symmetrical; refer to the construction procedure below for instructions. Inspect with a magnifying glass to verify that all leads have been soldered correctly. Pin sockets need to be inserted into the PCB to receive the adaptor board pins. It’s best to temporarily attach the adaptor board, solder those socket pins to the main board, Photo 3: using a DIP socket as a jig to hold the PCB pins. Photo 4: soldering the PCB pins to the SMD adaptor board. Photo 5: soldering the SMD IC to the adaptor board. Practical Electronics | February | 2026 IC mounting 37 Constructional Project then unplug it before you power it up. Construction The Waveform Generator is built on a double-sided PCB coded 04104251 that measures 101.5 × 73.5mm. The following instructions assume you will be soldering the three op amp and two regulator ICs directly to the PCB pads. If you want to use adaptors instead, the procedure is not terribly different except that you will be soldering those parts to the adaptors, then fitting the adaptors with pins and soldering matching sockets to the sets of five through-hole pads arranged around each chip location. Start by soldering the five SMDs. In each case, spread a thin layer of flux paste over the PCB pads first. The op amps, IC1-IC3, each have five pins with two on one side and three on the other, so the correct orientation of each should be obvious. Place the part on the board, tack-­s older one pin and check that the device is flat on the board and each lead is centred over its pad. If not, remelt the initial solder joint and gently nudge the part into place. Repeat if necessary until it is nicely aligned, then solder the remaining pins. Add a small amount of flux paste to the first pin and touch it with a clean soldering iron tip to reflow the joint. Given that these leads are quite close together, you may have accidentally bridged two or more pins. Use a magnifier to check. If you have, it’s quite easy to correct: simply add a small amount of flux paste to those pins, put the end of some solder-wicking braid on top and press it down onto the board and pins with your soldering iron. Wait for a few seconds until the solder melts, then drag the wick away from the pins and lift it and the iron off the board. That should leave behind just the right amount of solder. REG1 and REG2 are similar to IC1IC3, but they’re a bit more tricky because they have three pins on each side. That means you’ll have to figure out which of the two possible orientations is correct. The PCB is missing a pad on one side because pin 2 of these devices is not used. Examine the chip under magnification and find the pin 1 indicator in one corner. Rotate it so that corner is next to the missing central pad, then tack-solder one pin. Proceed with soldering as for IC1-IC3 but of course you can skip the pin which has no corresponding pad. You should still check for bridges to pin 2 (however unlikely they are) and fix them if present. If you manage to solder them in the wrong orientation, simply remove the middle pin and resolder it on the other side of the adaptor. Now move on to fit the throughhole resistors and diodes. The orientations of the resistors do not matter but the diodes do, so make sure their cathode stripes face as shown in the overlay diagram (Fig.3). Also, don’t get the similar-looking 1N4148 (standard silicon, D1-D4) and BAT41 (schottky, D5 & D6) diodes mixed up. Note that the resistors used are smaller than the standard 1/4W or 1/2W types generally used in our projects. As 1/4W resistors won’t fit in the specified case, we recommend you use 1/6W or 1/8W miniature body resistors. There are many resistor values used, so refer to the colour code table in the parts list or use a DMM set to measure ohms to ensure they go in the right locations. Follow with the capacitors, none of which are polarised except for the two larger electrolytics. Their longer (positive) leads face each other, as shown by the + marks on Fig.3. While many of the ceramic capacitors are 1μF types, there are quite a few different values, so don’t get them mixed up. The two larger 1μF 250V caps go near the output terminals as shown, laid over as otherwise they will be too tall to fit in the enclosure later. Next, fit the trimpots. There are eight in four different values, so again, make sure the right ones go in the right locations. Note that the footprints for the trimpots on the PCB have four pads, while the trimpots have three pins. This is to allow you to use either the common 3362P types or the less-­ common 3362R reversed version. Fig.3 shows the correct orientations for 3362P trimpots, and the PCB also has “P” and “R” labels on the two possible locations for the central pin. If using 3362R trimpots, rotate them 180° compared to what’s shown in Fig.3, so the central pin goes into the pads marked “R” on the PCB. Testing If you are using adaptors for the op amps, you can test the board before connecting any of the expensive op amps to the circuit. Connect the batteries, plug in the two regulators Fig.3: the three ICs and two regulators are shown soldered directly to the PCB here, but they can also be attached via SMD-to-DIL adaptors, using the rows of holes above and below each of those devices. Watch the orientations of the ICs, diodes, electrolytic capacitors, trimpots and rotary switches. The two LEDs indicate both when it is switched on and also whether the 9V batteries are still OK. Also note the way the batteries are wired – there is no reverse polarity protection! 38 Practical Electronics | February | 2026 Versatile Waveform Generator and switch the power on; both LEDs should light up. When connecting the batteries, it is best to have the power switch off; otherwise, accidentally touching a connector with the wrong polarity could damage a voltage regulator. Using the output ground (“COM.”) as a reference, measure the voltages at pins 2 & 5 of IC1 (you can use the larger through-hole pads or sockets rather than trying to probe the SMD pads). Pin 2 is at top centre and should measure -4.98V to -5.02V, while pin 5 is at lower-right and should measure +4.98V to +5.02V. If not, switch off and check for faults. If you’ve soldered these ICs directly to the board, you can still perform this test, but there is a risk of damaging the ICs if something is wrong with the regulators. So check the orientation of REG1 & REG2 carefully before switching on, as well as the polarity of the batteries and their wiring (you can do this by probing the battery terminals on the PCB with a multimeter). If everything checks out, and you have socketed the ICs, switch the power off and plug in IC1, IC2 and IC3. Make sure they’re all orientated correctly, with the sides with two pins facing towards the bottom of the PCB. Set the Amplitude control (VR5) to maximum and the Waveform switch (S2) to square wave. Set Symmetry (S3) to the 50% position, and all trimpots to around midrange. When power is switched back on, there should be a square waveform—or nearly so—at the output, centred at 0V. Troubleshooting Are both LEDs on? If not, the batteries, voltage regulators and associated circuitry need attention. If they’re on but there’s no output, check that the Waveform switch (S2) is not at one of the off positions and that the Amplitude control (VR5) is not at or near minimum. Try adjusting trimpot VR7 (“Hyst”). As usual, if you run into any problems, check that the ICs and diodes are all in the correct orientations. Remove the ICs, if using adaptor boards, and verify the supply voltages again. Check that the resistors and capacitors are all the correct values. Look for unsoldered pins or wires, and for solder bridges on both sides of the PCB. If you’re still stuck, check the output of IC1 at pin 1 (upper right). Practical Electronics | February | 2026 Parts List – Waveform Generator 1 double-sided PCB coded 04104251, 101.5 × 73.5mm 1 Serpac 131,BK plastic enclosure [Mouser 635-131-B] 1 panel label, 104 × 74mm 2 9-position vertical rotary switches, 18t split shafts (S1, S2) [Taiwan Alpha SR1712F-0109-15K0A-N9-N-027] 2 miniature PCB-mount vertical DPDT toggle switches (S3, S4) [Nidec ATE2D-2M3-10-Z] 4 miniature 2mm banana sockets [Amazon B08M3BCSB1] 5 SOT-23-6 to DIL breakout boards (optional) [SparkFun BOB-00717] 25 0.51mm diameter PCB pins (optional) [DigiKey ED90325-ND, Mouser 575-90810001508] 25 0.51mm diameter PCB pin sockets (optional) [Mouser 575-3016015152127] 2 9V rechargeable batteries [eg, 600mAh EBL6F22] (BAT1, BAT2) 2 9V battery snaps with flying leads (BAT1, BAT2) 5 knobs to suit the 18t spline shafts of S1, S2, VR5, VR8 & VR10 4 3mm inner diameter, 1mm-thick plastic or fibre flat washers 4 No.4 × 8mm self-tapping screws 4 stick-on rubber feet Semiconductors 2 AD8065ART op amps, SOT-23-5 (IC1, IC3; see text for other options) 1 AD8091ART op amp, SOT-23-5 (IC2; see text for other options) 2 ADP3300ARTZ-5 low-dropout 5V linear regulators, SOT-23-6 (REG1, REG2) 1 high-brightness 3mm red LED (LED1) [Kingbright WP710A10SRD/J4] 1 high-brightness 3mm green LED (LED2) [Kingbright WP710A10ZGDK] 4 1N4148 or equivalent 75V 200mA signal diodes (D1-D4) 2 BAT41 or equivalent 70V 15mA schottky diodes (D5, D6) Capacitors (all 50V radial multi-layer ceramic, 2.5mm pitch unless noted) 2 330μF 6.3V low-profile (5mm tall) radial electrolytic [Panasonic ECE-A0JKS331] 1 3.3μF 25/50V X7R ±10% [Murata RCER71E335K2DBH03A] 2 1μF 250V X7R ±10% [Murata RDER72E105K5B1H03B] 12 1μF 25/50V X7R ±10% [Murata RDER71H105K2M1H03A] 1 330nF 25/50V X7R ±5% [Kemet C333C334J5R5TA] 1 100nF 25/50V NP0/C0G ±5% [Murata RCE5C1H104J2A2H03B] 1 33nF 25/50V NP0/C0G ±5% [TDK FA14C0G1H333JNU00] 1 10nF 25/50V NP0/C0G ±5% [Kemet C315C103J3G5TA] 1 3.3nF NP0/C0G ±5% [Murata RCER5C1H332J0DBH03A] 1 1nF NP0/C0G ±5% 3 330pF NP0/C0G ±5% [Kemet C315C331J3G5TA] 1 100pF NP0/C0G ±5% [Vishay K101J15C0GH53L2] 1 47pF ±5% [TDK FG18C0G1H470JNT00] 1 33pF NP0/C0G ±5% [Vishay K330J15C0GF53L2] Potentiometers (all 9mm vertical plastic pcb-mount 18t spline shaft types) 1 5kW linear B-type (VR5) [Bourns PTV09A-4030U-B502] 1 100kW linear B-type (VR8) [DigiKey 987-1708-ND – see text] 1 1kW linear B-type (VR10) [Bourns PTV09A-4020U-B102] Trimpots (all 3362P-style miniature top-adjust) 3 2kW (VR1, VR2, VR6) 3 5kW (VR3, VR7, VR9) 1 1kW (VR4) 1 10kW (VR11) Resistors (all ⅛W miniature axial 1%) 2 470kW 1 3.3kW 1 100kW 2 2.2kW 1 43kW 1 1kW 1 27kW 1 470W 2 22kW 1 390W 1 3.9kW 1 330W 39 Constructional Project Fig.4: a pure sinewave shaped like this will have a low distortion figure, well under 1% THD. Try to get the output of your unit to match this as closely as possible. If there is a triangle waveform, then IC1 & IC2 are working and IC3 may need attention. If you’re getting strange waveforms, verify that the schottky and regular diodes have the correct orientations. Check the values of the following components: the filter capacitors across VR3 and series diode pair D1 & D3, IC3’s feedback capacitor, and compensation capacitor across the 2.2kW resistor from IC1’s output to VR7. Set-up and calibration Calibration requires the following steps in sequence. 1. Set the Frequency Band switch (S1) to the 1-3kHz position. Set the Frequency pot (VR10) and all trimpots at approximately midrange. 2. Connect an oscilloscope to the lowest lead of a capacitor below S1, using the output common terminal as the reference. 3. Set the S3 Symmetry switch to the 50% position and apply power. A triangle wave should be displayed on the oscilloscope. Adjust trimpot VR7 (Hyst) so you get exactly 4V peak-topeak. The triangle may be slightly asymmetrical; that will be fixed in step 5. 4. Connect the oscilloscope to the direct output terminal, set the Waveform switch (S2) to square wave mode and adjust VR5 for maximum amplitude. A square wave should be displayed on the oscilloscope. 5. Adjust trimpot VR4 (Balance) for an exactly symmetrical square wave. A multimeter with a duty-cycle measurement option would be useful here, or use a similar oscilloscope measurement. Adjust VR10 (Frequency) if necessary. 6. Set S3 to its alternative Vary position. Adjust trimpot VR9 (“Sym”) so you get slightly less than 5% duty cycle with VR8 fully anti-­clockwise and slightly more than 95% duty cycle with VR8 fully clockwise. The duty cycle can be pushed from 2% to 98%, but frequency shift may increase. 7. With S3 still in the Vary position, adjust VR9 (Sym) for an exactly symmetrical waveform. Note the frequency. Set S3 back to the 50% position and achieve exactly the same frequency by adjusting VR11 (50% Freq). 8. Set S3 back to the 50% position and S2 to sinewave. An approximate sinewave should be displayed. Sinewave adjustment 9. Adjust trimpot VR1 (THD) to achieve the cleanest possible sinewave. You can trace Fig.4 onto tracing paper, baking paper or clear plastic and place it over the oscilloscope screen as a guide. Alternatively, if your ‘scope has a spectrum analyser mode (or you have a spectrum analyser) adjust VR1 for minimum harmonics. If you are not fussy, forming an approximation to a sinewave on a ‘scope screen may be good enough. If using a spectrum analyser, I suggest setting the Wave Generator frequency to 1kHz and the analyser frequency span to cover the audio range. Momentarily switch to triangle wave mode and adjust trimpot VR4 (“Bal”) to minimise the second (2kHz) and all other even harmonics. This should only require a slight readjustment. Switch back to sinewave mode and adjust VR1 (“THD”) to minimise the odd harmonics. Then adjust trimpot VR1 (THD) to minimise the odd harmonics. VR7 (Hyst) may also be adjusted a slight amount, but this will also alter the frequency bands. When you’ve finished, all even harmonics should be approximately 60dB lower than the fundamental and all odd harmonics (starting at 3kHz) should be at least 45dB lower than the fundamental. Adjust the amplitude setting as necessary to avoid overloading the spectrum analyser. A sinewave THD close to 1% should be achievable. Wave amplitudes 10. Leaving the ‘scope connected to the direct output and S2 in the sinewave position, set VR5 (Amplitude) to maximum. Now you have a choice of equal peak voltages or equal RMS voltages for the three waveforms. For equal peak voltages, decide on what maximum you want and adjust VR2 (Sine) to that maximum. I do not recommend greater than 6V peak-topeak. Next, set S2 to square wave mode and adjust VR6 to achieve the chosen maximum output level. Switch S2 to triangle wave mode and adjust trimpot VR3 (Tri) to achieve the same maximum level. Alternatively, to set the waveforms to equal RMS voltages, use Table 3 or an RMS-reading device (multimeter or oscilloscope). 11. Check that VR10 (Frequency) varies the frequency over a range of at least 3:1 and check the minimum Fig.5: the controls are quite complicated so you’ll need this panel label to understand what they all do. It will also help you locate the holes for the switch and potentiometer shafts, LEDs and banana sockets. You can download it as a PDF from our website and print it at actual size (1:1). 40 Practical Electronics | February | 2026 Versatile Waveform Generator and maximum frequency for each band. The bands should overlap. If the minimums are not low enough, decrease the value of the 390W resistor. If the maximums are not high enough, adjust VR7 (Hyst) slightly and return to step 8. The frequency bands will likely not track by exact factors because of the typical variations in capacitance of the timing capacitors. That’s why these capacitors (all the ones that connect to pin 1 of IC1) should have a ±5% or better tolerance, if possible. In the worst case, you may need to replace one or two caps or parallel them with lower-value capacitors. 12. With S2 (Waveform) set to triangle wave and S3 (Symmetry) at the Vary setting, rotate VR8 (Symmetry) to both extremes to check that the triangle wave becomes a clean downward or upward ramp/sawtooth, and recheck that, on the square wave setting, the output becomes a pulse that varies in duty cycle between 5% and 95%. Enclosure preparation Fig.5 is a front panel label that can also be used as a drilling guide. You can download it from siliconchip.au/ Shop/11/1823 We have instructions on preparing and attaching panel labels online, see: https://pemag.au/Help/FrontPanels With the panel label attached, the holes can then be drilled through carefully. The final hole sizes are 3mm for the LEDs, 8mm for the potentiometers, 10mm for the rotary switches, 4mm for the toggle switches and 2.5mm for the banana sockets. If possible, I suggest punching the small holes. I also suggest countersinking the small holes on the inside of the enclosure for easier insertion of the LEDs, switches and banana receptacles. The mounting post on the top part of the enclosure that is near rotary switch S2 needs to be trimmed back a bit to allow room for the switch. The anti-rotation tabs on the tops of the rotary switches and pots need to be removed. Insert the LEDs and banana sockets into the PCB with the LEDs in the correct orientations, but do not solder them yet. Temporarily fit the PCB into the enclosure using a 1mmthick non-­c onductive (eg, plastic or fibre) spacer or washer on each mounting post. Top tip: use super glue to stick the washers in place temporarily (either to the enclosure or top of the PCB) so they don’t slide out as you’re trying to assemble everything. Adjust the LEDs and banana receptacles as desired, then solder the LEDs, and tack-solder the sockets quickly to avoid melting the plastic. Remove the PCB and solder the sockets to the upper surface of the PCB, being careful to maintain their position. You can now screw the PCB into place in the enclosure on the 1mm spacers. Do not use panel-mount hardware on the rotary switches or VR8. After considering several mounting methods for the batteries, I simply used a little epoxy to attach them to the lower part of the enclosure, with a piece of thick paper in between should I ever want to remove them. You could also consider foam-cored double-sided tape, although it may not be strong enough to hold them long-term. Usage notes The square wave or pulse rise and fall times are approximately 90ns (see Scopes 1 & 2). There is a barely noticeable non-linearity in the triangle waves at the three lowest frequency bands. I attribute this to the capacitors, which are X7R for these bands. The higher-frequency bands use C0G/NP0 capacitors and look perfectly linear to my eye. Using C0G or film capacitors for the higher-value timing capacitors would eliminate the slight non-linearity, but they are too large to realistically fit. For an explanation of capacitor types, see our detailed article on capacitors in the December 2024 issue. Scope 3 compares the Waveform Generator’s quasi-sinewave (mauve) to a pure sinewave (yellow) at 1kHz; the pure sinewave was generated by sending the Waveform Generator quasi-­ sinewave through a three-stage RC filter. Table 3 – peak vs RMS voltages Waveform RMS formula Peak for 1V RMS Peak for 2V RMS Square/pulse Vrms = Vpeak 1V 2V 1.73V 3.46V 1.41V 2.83V Triangle/ramp Sine Vrms = Vpeak ÷ √3 Vrms = Vpeak ÷ √2 Practical Electronics | February | 2026 Scope 1: a 30kHz pulse with a duty cycle of 2%, from setting “Waveform” to square/pulse and the “Symmetry” control fully anti-clockwise. Scope 2: a 30kHz ramp, from setting Waveform to triangle/ramp and Symmetry control fully anti-clockwise. Scope 3: a pure sinewave (yellow) with the generator’s output overlaid (mauve) at 1kHz. The total harmonic distortion (THD) is around 1% if it’s properly adjusted. There is a slight phase shift between the two waveforms. There is a frequency shift, up to 1-2%, as the symmetry/duty cycle is varied between 5% and 95%. This appears to be a peculiarity of the potentiometers; in particular, carbon-­ element potentiometers. Cermet pots have much less shift, but they are considerably more expensive. A likely additional contributor is the non-ideal characteristics of the PE schottky diodes. 41