Fullscreen HTML5Med Res (??MB)HTML4

Improvised Electronics and DIY Components Part 1 by Dr David Maddison, VK3DSM Image source: www.pexels.com/photo/tools-on-a-desktop-7286026 During World War II, prisoners of war scavenged parts such as razor blades and lengths of wire to build clandestine radios. These allowed them to hear news from home and keep track of the progress of the war. These days, hobbyists in their garages fabricate electronic components such as valves, transistors and even simple integrated circuits. T hese activities fall into the category of improvised electronics. In the hobbyist context, improvised devices are made from scratch, from repurposed parts or from limited resources for fun, education or survival. Sometimes it’s simply a thought experiment: how would you make today’s electronic components if commercial parts weren’t available? Who knows — one day we might need to rebuild society from scratch! Improvised electronics showcases human ingenuity, democratises technology and teaches fundamental electronics principles. It can even have survival value in certain situations. Following the pioneers When the pioneers of electricity and electronics (Faraday, Maxwell, Edison, Marconi etc) made their groundbreaking discoveries and inventions, there were no commercial electronic 12 Silicon Chip components available. They were forced to fabricate everything from raw materials – the same challenge faced by modern-day experimenters and improvisers. The fundamental components needed to make almost any electronic circuit are resistors, capacitors, inductors, diodes, transistors or valves, switches and transformers. Some sort of power supply is also needed, such as a solar panel or battery, along with wire and a means of connecting the components together. Most of these components and techniques can be improvised at a basic level. It wasn’t just those early pioneers who used such techniques; from the 1910s through the 1920s, and for many years afterwards, people built crystal radios. Unlike the expensive valve radios that became available from the 1920s, a crystal set could be built with just a coil of wire, a capacitor, a crystal Australia's electronics magazine detector (often galena) and earphones. In the 1920s, newspapers, magazines and the US Bureau of Standards published guides explaining how to build them. They were also widely built for educational purposes. In Australia, The Sunday Mail (Brisbane) published plans for the “Mystery” crystal set on 3rd July 1932 (see Fig.1). Another early Australian crystal radio is shown in Fig.2. The purpose of this article is not so much instructional but to demonstrate what is possible in the absence of commercially available components (for fun, education or survival). We will look at how these components can be made with relatively simple materials and tools. If you wish to make any of these, you will need to obtain detailed instructions from books or online sources, formulate your plans and apply appropriate safety precautions. Some of the siliconchip.com.au chemicals or techniques involved can be hazardous. This article will concentrate on basic components and techniques. The second and final part, published next month, will look at more advanced DIY techniques, such as IC fabrication. Basic components Basic electronic components that can be made at home include antennas, cells & batteries, capacitors, motors, diodes, sound transducers, inductors, LEDs, magnets, magnetic cores, memristors, resistors, solder, transistors, switches, wire and tunnelling diodes. They can be improvised as follows. Antennas Improvised antennas can be made from a length of wire, say 10-20m, for basic reception. For lower frequencies where more inductance is needed, the wire can be wound around a core (we’ll look at DIY inductors and cores later). Using the improvised magnetic core material described later, something like a ferrite rod antenna is possible. The core should be 10-12cm long and 1-2cm in diameter, then wound with about 30-100 turns of insulated wire (0.32-0.64mm in diameter). This can be coupled with a variable capacitor of around 150-500pF, or a fixed capacitor to receive a specific AM broadcast band frequency (by trial and error if necessary). This creates an LC tank circuit. More rods can be bundled for greater sensitivity. More turns are used for lower frequencies, or fewer for higher frequencies. Thicker wire can also be used, reducing the resistance and improving the Q-factor (improving selectivity by rejecting nearby stations but making precise tuning more difficult). Such tuned antennas can be used for crystal radios. If building an improvised low-power transceiver, a magnetic loop antenna can be very useful to obtain good gain and directivity without needing to construct a complex Yagi or erect a tall mast (although a straight wire antenna can also be used). A small magnetic loop offers reasonable efficiency and sharp nulls for rejecting interference. Peter Parker, VK3YE, has videos on making these at: • https://youtu.be/cg1AXQb3VGM • https://youtu.be/Cv_RnLpZ9gw • https://youtu.be/gRRtlarJnKg Capacitors Capacitors consist of two conductive plates separated by an insulating dielectric material. The simplest possible improvised capacitor uses two pieces of aluminium foil with a sheet of paper, plastic kitchen wrap (cling film), waxed paper, or a thin plastic sheet as the dielectric. Capacitance increases with a larger plate area, closer plate spacing and higher dielectric constant (waxed paper or plastic is better than dry paper). Capacitance is given by the formula C = (εA)/d where A is the area of the plates, d is the distance between them, and ε is the absolute permittivity of the insulator. For improvised radios or tuned circuits, the foil-and-paper or plastic wrap version is simplest, but a variable capacitor will be needed for tuning a radio as well. A simple variable capacitor can be made using overlapping Fig.3: some improvised paper capacitors. Source: www. instructables.com/PaperCapacitor aluminium plates or copper-­laminated fibreglass (ie, PCB substrate) separated by a thin insulator (possibly air). The capacitance is adjusted by changing the amount of overlapping surface area. It is also possible to use two cylindrical pieces of metal (eg, a bolt and a metal tube) with a plastic tube separator between them. Another improvised variable capacitor is essentially two pieces of timber hinged on one side, with their inner surfaces covered in aluminium foil. The gap is adjusted to alter the capacitance. For a crystal radio to tune into the AM broadcast band, a capacitor value between about 150pF to 500pF is needed, depending on the value of the air coil or inductor and the frequency to be tuned. This range works well with common improvised air-core inductors. For example, 77 turns of wire on a 10cm-long, 5cm-diameter former gives Fig.1: the “Mystery” crystal set published in The Sunday Mail, Brisbane, 3rd of July 1932. No newspaper would publish such a thing today! Source: https://trove.nla.gov.au/ newspaper/page/10204138 Fig.2: an early Australian crystal set, circa 1925, made by Fred Smithson in Melbourne. Source: https:// collections.museumsvictoria.com.au/ items/404398 siliconchip.com.au Australia's electronics magazine September 2026  13 Fig.4: after making the ‘sandwich’ for a commercial electrolytic capacitor, the foil is wound into a cylinder and inserted into a can. Fig.5: a Leyden jar capacitor. Source: www.instructables.com/ Leyden-Jar-1 Fig.6: a cross-section of one layer of an electrolytic capacitor. A thin layer of aluminium oxide is formed on the anode, which acts as the dielectric layer. The conductive electrolyte allows electrons to flow between the cathode right up against that oxide layer, maximising capacitance per area. Fig.7: charging a homemade supercapacitor. Source: https://youtu.be/9DZtpBa_GnA 14 Silicon Chip Australia's electronics magazine a 66µH inductance. The resonant frequency formula is f =1 ÷ (2π√LC). With 66µH and 150-500pF, the tuneable range is 876-1600kHz. Other improvised capacitor types are as follows. A common variation of the aluminium foil capacitor is to roll the assembly tightly into a cylinder (rolled capacitor) instead of keeping it flat. This increases capacitance in a compact space and it’s how commercial electrolytic capacitors are made. See Fig.3, Fig.4 and the video at https:// youtu.be/n1hP3HKje3k A Leyden jar is a classic design using a glass jar or bottle (some have used plastic bottles) as the dielectric, with aluminium foil (or copper tape/foil) coating the inside and outside surfaces – see Fig.5. The jar can be partially filled with water or a salt solution to improve contact. It can store a much higher voltage and charge than foil-­paper versions. Capacitors can be made from a length of coaxial cable in which the inner conductor and outer shield are used as the plates, with the plastic insulation as the dielectric. They are suitable for HF and VHF applications. For higher capacitance in a compact size, an improvised electrolytic capacitor can be made using aluminium foil plates, a formed oxide layer on one of the plates, a liquid electrolyte like a saturated solution of baking soda in distilled water, and a porous separator like paper towel or coffee filter (plus a container and wire terminals) – see Fig.6. Once the capacitor is assembled as either a layered or rolled structure, the oxide dielectric has to be formed. The positive terminal of a DC supply is connected to one plate, the anode, and the negative to the other plate (the actual cathode is the electrolyte). The current drops as the oxide layer ‘forms’ over several minutes. There is already a natural oxide layer on aluminium, but this enhances it. The capacitor is then ready to use. Work in a well-ventilated area, as hydrogen and oxygen gases can be produced during forming. Use a low current (limit with a resistor if needed) and avoid short circuits or overvoltage. Supercapacitors can also be improvised. The electrodes can be made from copper, nickel, stainless steel, carbon, or titanium coated with a high-surface-area material like activated carbon or charcoal, separated siliconchip.com.au by a porous material (such as a coffee filter or porous plastic). The porous material is soaked in an electrolyte like lemon juice, salt water, sodium hydroxide (drain cleaner, NaOH), potassium hydroxide (KOH) or dilute sulfuric acid (H2SO4). This creates an electrochemical double-layer capacitor with significant energy storage for a homemade device. A homemade supercapacitor is shown being charged in Fig.7. It uses copper electrodes, a paper separator and NaOH, which is OK for shortterm use, but over the longer term, the copper will corrode and the paper will degrade. The author of the video claims a capacitance of 200F – see the video https://youtu.be/9DZtpBa_GnA Such capacitors can only be charged up to about 1.5V, as water decomposition occurs above that. Strong electrolytes (NaOH, KOH or H2SO4) must be handled with extreme care as they are corrosive and can cause burns. Safer options are sodium sulfate (Na2SO4) or Epsom salts (MgSO4). Some selected capacitance values are shown in Table 1. Carbon rods Carbon rods can be extracted from zinc-carbon (non-alkaline) “heavy duty” D cells or non-alkaline square 6V lantern batteries – see Fig.8. They are useful for a variety of applications in improvised electronics, such as: • electrolysis (eg, splitting water into H2 and O2) • producing intense arcs when connected to a high-current source like a car battery or arc-welding transformer • resistive heating elements • the positive electrode in seawater electrodes, paired with a zinc negative electrode • adjustable resistors • electrical contacts for commutators etc Cells and batteries A battery is a group of cells, so if you can make cells, you can combine them to make batteries. Cell types you can make include: A lemon or potato cell. A piece of copper (from an old coin or copper pipe) and a zinc-coated piece of metal is inserted into a lemon, potato or other acidic fruit or vegetable. The acid acts as an electrolyte. These produce low voltages, are short-lived and mainly useful for science demonstrations. siliconchip.com.au Table 1 – capacitance ranges for improvised capacitors Type Capacitance Notes Coaxial cable 60-100pF/m 50W cables give more capacitance than 75W 500mL Leyden jar 1nF 2L PET bottle filled with salt water 8nF A4 aluminium plates separated by paper 1.1nF; 20nF if rolled into a cylinder Electrolytic using baking soda & water electrolyte 1-100µF Supercapacitor using KOH 1-5F for simple activated Don’t exceed 1.2V-1.5V; charcoal type; up to high ESR 250F Spiral construction gives higher capacitance Fig.8: carbon rods retrieved by the author from non-alkaline D cells and a 6V square lantern battery, which internally had four F cells. Fig.9: an improvised lead-acid battery. Source: https://youtu. be/6VzLatndrPU Alessandro Volta made a copper-­ zinc and saltwater cell into a battery, like a voltaic pile, in 1800. You need an alternating stack of copper and zinc pieces, separated by paper or cardboard soaked in salt water, vinegar or lemon juice as the electrolyte. Copper can be derived from old coins or pipes, while zinc or galvanised steel can be used for the anode. These cells require frequent rewetting. An aluminium-charcoal and saltwater cell is made from aluminium foil or scrap as one electrode and crushed activated charcoal (such as the charcoal from a campfire or BBQ) for the other. Salt water is the electrolyte. These batteries have a long shelf life and are activated only when salt water is added. Commercial versions are used on life rafts. A copper-iron and vinegar cell can be made from copper scrap and steel nails, with vinegar or lemon juice as the electrolyte. They are weak and short-lived. The Daniell Cell is a classic 19th-century improvised cell made with a copper electrode in a copper sulfate solution, a zinc electrode in a zinc sulfate or saltwater solution and both electrodes separated by a porous barrier such as unglazed pottery or Plaster of Paris. Copper sulfate can be improvised by placing copper electrodes in sulfuric acid and passing a current between them. Lead-acid cells can be improvised from lead plates, with one lead plate treated to have a lead dioxide coating, plus separators and a sulfuric acid electrolyte. Access to sulfuric acid is restricted in Australia (even though pretty much everyone owns one or more lead-acid batteries filled with it). The following videos show how to make various kinds of leadacid batteries: Australia's electronics magazine September 2026  15 Table 2 – improvised cells/batteries (voltages are approximate) Cell type Anode Cathode Electrolyte Voltage Current Lemon/ Potato Zinc Copper Acidic fruit/ vegetable 0.7-0.9V up to a few mA Salt water Zinc * Copper Salt water 0.8-1.0V up to tens of mA Aluminium- Aluminium Charcoal charcoal Salt water 0.9-1.2V up to ~100mA Copperiron acid Iron Vinegar 0.6-0.8V up to a few mA Copper Sulfate solutions ~1.1V up to hundreds of mA Copper Daniell Cell Zinc Lead-acid Lead Lead Sulfuric acid ~2.1V fully dioxide charged * galvanised steel can be substituted but the voltage will be lower • https://youtu.be/0zQYJJz89Wo • https://youtu.be/tzyUda3upHI • https://youtu.be/pWO1aNUykq4 • https://youtu.be/6VzLatndrPU (see Fig.9) • https://youtu.be/5B8hL_UNUpk Alum (aluminium sulfate) solution is a much safer electrolyte, but gives a battery with different characteristics and much lower capacity. Lithium batteries can also be improvised. An aluminium electrode is coated with lithium manganese oxide and a copper electrode is coated with graphite. The electrodes are immersed in an organic electrolyte, and a plastic separator is placed between the electrodes. This is shown in the YouTube video https://youtu.be/VZ9KcGG-cps You might have some problems getting the chemicals for this one, and they are hazardous. They would therefore not be suitable in a survival situation. This type of battery is best left for a professional laboratory environment, but we included it to show what can be done. The non-rechargeable batteries mentioned above should work straight away, but rechargeable types need some method of charging them. Key characteristics of the above batteries are summarised in Table 2. DC motors One of the hardest things to improvise is a strong permanent magnet, although we will later describe how to improvise weaker magnets. Strong magnets are required for a high-­performance DC electric motor, although no-magnet DC motor can be made by substituting a fixed magnet with an electromagnet (field windings) powered by the same power supply as the armature (rotor), as shown in Fig.10. This design is known as a woundfield or series-wound motor, and was common in early electric motors. It is still used in many DIY and improvised builds. Diodes There are two main types of diode construction: the earlier point-­contact type (using a metal wire pressed against a crystal) and the more modern junction type (a P-N junction formed within a single semiconductor crystal). The point-contact type can be Fig.10: examples of improvised DC motors that don’t require permanent magnets. Source: https://youtu.be/tBQLGrXohL8 16 Silicon Chip amps improvised with basic materials like a semiconductor crystal and a sharp metal wire (‘cat’s whisker’). These improvised devices are often fragile and have low efficiency, but they do work. For the crystal, galena was common in the past, but silicon, germanium and carborundum are also suitable – see Table 3. A rusty or blued razor blade can function as a crude diode because its surface develops a thin layer of iron oxide that acts as a semiconductor. When a graphite or metal point is brought into contact with this layer, it forms a rudimentary schottky barrier junction that allows current to flow in only one direction. This rectification process can demodulate radio waves, extracting the audio-frequency modulation and converting it into a varying signal that can drive a headphone or earpiece to produce sound. The oxide layer is uneven, so finding the ‘sweet spot’ with the contact point is essential for good performance (see siliconchip.au/ link/acbk for more details). A cuprous oxide (Cu2O) diode can be fabricated by cleaning a piece of copper and heating it to cherry red for 30-60 minutes. A black copper oxide (CuO) insulating layer will grow. Let it cool slowly, then scrape or sand off the black layer to reveal the red/ maroon Cu2O layer underneath. This is a p-type semiconductor material. Make a point contact with a needle or use a blob of solder to make contact (cathode, −) with the red surface and make a connection to the metal base (anode, +). Expect a forward voltage drop of 0.2-0.5V and low current handling. An improvised electrolytic ‘wet’ diode can be made by submerging an aluminium or copper plate (anode, +; aluminium is better) and a lead plate (cathode, –) into a solution of baking soda (sodium bicarbonate) or borax Fig.11: an improvised high-impedance piezoelectric earpiece. Source: https://youtu.be/dWkKFD_yZs8 Australia's electronics magazine siliconchip.com.au Fig.12: a matchbox microphone. Source: www.instructables.com/ Matchbox-Microphone (sodium tetraborate) in water. This was invented in 1902 by Frenchman Albert Nodon and is known as the Nodon valve or rectifier. Another home-brew diode can be made by reacting copper with sulfur. Coat copper with sulfur powder and heat indirectly to melt the sulfur, which will eventually ignite, leaving a dark coating. Sulfur fumes are toxic, so this should be done outdoors with suitable respiratory protection. One end of the circuit is connected to the unreacted copper with a fine point (cat’s whisker) used to find a spot that rectifies. While copper works, galena is superior. Fig.13: an improvised speaker without a permanent magnet. Silicon Chip readers might try improving on this design. One set of terminals is for the electromagnet, the other for the signal. Source: https://youtu.be/ wqGmMFo1jGg Earphones/speakers/microphones Crystal radios require high-impedance earphones because the detector circuit has a very high output impedance and produces extremely little power. A low-impedance earphone would heavily load the detector and collapse the signal voltage, whereas a high-impedance earphone allows the audio signal to develop without significantly disturbing the tuned circuit. The ideal earphone impedance for a crystal set is around 20-100kW. One can alternatively use low-­ impedance earphones with an audio matching transformer having a turns ratio of about 50:1 (a salvaged 230V to 5V transformer would be close). A suitable improvised transformer could be made pretty easily. Improvised high-impedance earpieces can be either piezo-electric or magnetic types. Magnetic earpieces were used with radios but required thousands of turns of wire. A classic high-impedance earphone of 2kW might require 2,000-10,000 turns of 0.1mm diameter or smaller enamelled wire depending on the design – see Fig.11. Note that in general, an earphone or speaker can also be used as a microphone in an emergency. Making a DIY high-impedance earphone is shown in the video titled “Crystal Radio Earphone From Common / Cheap Stuff” at https:// youtu.be/dWkKFD_yZs8 It uses a piezoelectric element salvaged from a beeper commonly found in some microwave ovens or other appliances with a beeper (eg, smoke alarms). Also see the video titled “Make Crystal Earphone/Earpiece for Crystal Radio – Homemade” at https://youtu. be/OVapqz4yBCY Another method is to make your own piezoelectric earpiece or speaker using Rochelle salts, which will be discussed later. A microphone can be improvised from a matchbox or similar small box and some pencil leads – see Fig.12 & www.instructables.com/Matchbox-­ Microphone A modern speaker requires a permanent magnet, but older speakers used electromagnets, and that’s still a valid technique. A basic improvised speaker typically uses two coils: one coil, powered by a DC supply (the electromagnet) creates a steady magnetic field, while the other coil (the voice coil) is attached to the diaphragm and driven by the audio signal, causing it to move in the magnetic field to produce sound. For more details, see Fig.13 and the video at https://youtu. be/wqGmMFo1jGg Inductors and transformers An inductor is a two-terminal device that temporarily stores energy in a magnetic field and resists rapid changes in current. It is usually in the form of a wire coil and is among the easiest devices to improvise. It can have an air, powdered/laminated iron, or ferrite core. A transformer is built similarly, but it has two or more coils on a shared core so that energy can be transferred between them. For an improvised coil former, just about any cylindrical material can be used, such as PVC or other plastic pipe, bamboo, timber dowels, cardboard tubes, or even a bottle – see Fig.14. A Table 3 – suitable materials for making point-contact diodes and transistors Material When used Signal sensitivity Ease of use Suitable for point-contact transistor? Galena (PbS) 1900-1920s High Very finicky Very low gain and stability Silicon (Si) 1920s-1940s+ Good Moderate Less forgiving than germanium Germanium (Ge) 1940s-1960s Very high Moderate Easiest to use Good More stable Not practical due to very low gain and stability Silicon carbide 1900s-1910s (carborundum, SiC) siliconchip.com.au Australia's electronics magazine September 2026  17 Fig.17: a circuit to ‘program’ an improvised memristor with either high or low values. Fig.14: air-cored inductor coils wound on 50mm cardboard tubes. Source: https://analogdial.com/TRFRadio/TRF1.htm Fig.15: an improvised transformer using a nut as the magnetic core. This will have high eddy current losses at any significant frequency. Source: https://youtu.be/ M6D-tZEA4jc Fig.16: an improvised step-up boost converter with the transformer wound on a nut and bolt. Source: https://youtu. be/4bgFEFBHzx0 coil is an essential element of LC tank circuits as used in radio transmitters and receivers, as well as in many other applications, like DC/DC converters. The basic rule for improvised (or any) coils is to use the thickest wire that fits the required number of turns within the available space. The inductance of a coil can be significantly improved by using a core, which concentrates the magnetic field and increases the coil’s effective inductance – more on that later. You can also make a transformer using a nut or a bolt as the magnetic core, as shown in Figs.15 & 16. LEDs A weak, primitive LED can be made by replicating Henry Joseph Round’s 1907 discovery by applying around 18 Silicon Chip 10-100V DC at up to 10mA across silicon carbide (SiC, carborundum) crystals (3-10mm in size) sitting on a conductive foil or in an alligator clip (one electrode) to produce a yellow-­ green glow. The other electrode is a sharp point-contact wire (cat’s whisker) pressed against a crystal. If no glow appears with one polarity, reverse the connections. The junction can work in either direction, although one may produce a stronger glow. This is regarded as the world’s first LED. Modern LEDs use different structures. For more on this, see the video at https:// youtu.be/hUYZfjB3GKs Magnets Magnets can be used in improvised speakers or microphones. An Australia's electronics magazine improvised weak permanent magnet can be made without using another magnet by taking a piece of iron or mild steel (like a large nail), aligning it with magnetic north and at an angle to the horizontal to match the magnetic dip angle in that area (typically 30-70°, depending on the latitude) and repeatedly hammering it. This will align some of the magnetic domains and make a magnet strong enough to pick up paper clips or possibly for use in a speaker or microphone. The magnetism will fade within hours or days, but it can be easily refreshed. If you already have power, say from an improvised battery, you can make an electromagnet. A strong magnetic field produced by a coil carrying sufficient current can permanently magnetise hardened steel. Making a magnetic core For making coils or inductors, a magnetic core will give increased inductance compared to an air core. You can make your own improvised inductor, transformer or antenna core by mixing iron oxide (rust) with a suitable binding material like plaster, epoxy, wax or similar in about a 50:50 volume mix. Solid iron or steel can also be used as an inductor core, but is only suitable for very low frequencies (under 500Hz) due to high losses. Powdered iron cores can be used up to about 10-20MHz and ferrite cores up to around 300MHz. Memristors In our recent articles about Analog Computing (May & June 2026 issues; siliconchip.au/Series/459), we described memristors, devices that can remember the last resistance ‘programmed’ into them. It is possible to improvise one. siliconchip.com.au Nyle Steiner demonstrated memristor behaviour with copper, brass or lead that had been reacted with sulfur, then creating a junction of the metal sulfide coating with aluminium. See Fig.17, http://sparkbangbuzz.com/ memristor/memristor.htm and https:// youtu.be/MlswP_qXbdA at 1kHz, a capacitance of around 150300nF and a leakage resistance of 20MW. Because of their high impedance, they need very little current to drive, but are sensitive to voltage, making an earpiece suitable for weak, high-impedance signals from a crystal radio detector. The energy transfer is efficient in Resistors such a configuration as the impedImprovised resistors can be made by ances match. As they are passive various methods. One is to use a soft devices, no electronics are needed to pencil to scribble a solid pattern on drive them. a piece of paper, as shown in Fig.18. The materials needed are: Electrical connections can be made via • 150g of cream of tartar (potassium paper clips. Another method is to wrap bitartrate; make sure it is real and not copper wire around the ends of a pen- a substitute) cil lead – see Fig.19 and the video at • 50-75g of washing soda (sodium https://youtu.be/kMeR1k1C8_M carbonate), or if not available, baking Another method is to use high-­ soda (sodium bicarbonate) baked at resistance wire, like Nichrome or steel about 200°C for one hour wire, to make a wirewound resistor, • 250mL of distilled water wrapping it around an insulating for• two borosilicate glass (eg, Pyrex) mer (ideally made from a heat-resistant beakers or jars material). The resistor can then be pot• a stirring rod or spoon ted. For more on this, see the video at • a coffee filter and a funnel https://youtu.be/b1VfXZgxI2w • a thermometer Liquid resistors and liquid rheoHeat the distilled water to 80-100°C, stats are simple to make, comprising add 150g of cream of tartar slowly water, a salt solution such as sodium and carefully while stirring, then add carbonate (washing soda), or ideally sodium carbonate gradually in small sodium bicarbonate (baking soda) and amounts – the solution will fizz. Keep two electrodes, perhaps stainless steel or carbon rods. To vary the resistance, the concentration of salt water can be varied, or the amount of immersion of one or both electrodes in the solution can be changed. Such devices have been used commercially in the past, and are even in use today. In DIY applications, use Fig.18: an improvised resistor using them in a well-ventilated area and only a pencil, paper and paper clips. for low-to-moderate power dissipation Source: www.instructables.com/DIYand heat buildup. The potential to Emergency-resistor-an-electronicproduce hazardous electrolysis gases circuit-com like hydrogen/chlorine is an important safety consideration. Rochelle salts Piezoelectric crystals develop an electric charge in response to mechanical stress and can be used for microphones, speakers, earpieces or sensors. Piezoelectric Rochelle salt (also known as potassium sodium tartrate tetrahydrate) crystals are relatively easy to make from common ingredients and can be used for high-­ impedance speakers and microphones. Speakers and microphones based on Rochelle salts can have very high impedances, in the range of 20-25kW siliconchip.com.au adding small amounts until there is no more fizzing. Filter the solution through the coffee filter and funnel it into another glass container to remove any particles. Cover the container and leave it in a cool, quiet spot until the crystals form. Then pour off any remaining liquid. The crystals can then be harvested. Dry the crystals with paper towels. If desired, one of the best crystals can be used as a seed to grow larger crystals by suspending it with fishing line in a fresh solution. The best crystals are the biggest and most regularly shaped with two opposing flat surfaces. Electrical contacts can be made using aluminium foil held in place between the faces of a clamp, with pieces of rubber to distribute the forces evenly to avoid breaking the crystal – see Fig.20. Another mounting method is shown in Fig.21. An article discussing commercial uses of Rochelle salt “reproducers” or speakers from the July 1932 edition of Radio-Craft magazine can be seen at siliconchip.au/link/acbt (Fig.22). “leafcutterjohn” made a sound recording using a homemade Rochelle salt microphone that you can listen to at siliconchip.au/link/acbs Fig.20: a method to hold a Rochelle salt crystal between two electrodes. Normally, the crystal would be mostly clear. Source: https://youtu.be/ RtW277wDrtM Fig.19: an improvised resistor using copper wire and a graphite rod from a pencil. Source: https://youtu.be/ kMeR1k1C8_M Fig.21: a Rochelle salt crystal with wires attached using conductive adhesive. It was successfully used as a guitar pickup. Source: https://youtu. be/8QP7F1VT1rw Australia's electronics magazine September 2026  19 A Rochelle salt microphone was also used as a guitar pickup – see Fig.21 and https://youtu.be/8QP7F1VT1rw Solder Wires can generally be twisted together or held together with some type of mechanical fastener, but soldering is more reliable. Leaded solder is the easiest to use and is made from 60-63% lead and 37-40% tin by weight. Flux is needed to get the solder to wet the surfaces being connected. You can find many simple recipes online, but an effective flux can be made by dissolving citric acid in water to make a concentrated solution. The residue is corrosive to electronics and must be washed away with hot water. A better flux recipe is to dissolve rosin (the purified resin or sap from pine trees) in isopropyl alcohol, producing a non-corrosive, classic flux that’s safer and more reliable, with a residue that doesn’t have to be removed. If isopropyl alcohol is not available, methylated spirits or acetone can be used. However, these may harm some plastics, especially acetone. Fig.24: the structure of the first transistor from Bell Labs. Source: https://w.wiki/KsX5 of transistor construction methods. The first transistor made by Bell Labs in 1947 was a point-contact transistor (see Figs.23 & 24) and this is more amenable to home or improvised construction. The more modern type of transistor is the junction transistor with NPN or PNP construction, which is more reliable and a better performer, but requires high-temperature doping, cleanrooms, photolithography, precise junctions and doping, diffusion and epitaxial growth techniques. This is not impossible at home, but it is significantly harder and less reliable. We will discuss improvised junction transistors later. The first Bell Labs point-contact transistor was built on a piece of n-type germanium. No deliberate doping was done to create a p-type top layer, as shown in Fig. 24. Instead, the emitter point contact injected holes (positive charge carriers) into the n-type base material, creating a localised p-type inversion layer near the surface. The collector point contact then collected these holes. This surface effect effectively formed a PNP transistor structure, while the bulk material remained n-type. The n-type germanium served as the base, while the metal-­semiconductor interfaces under the emitter and collector points created the necessary p-type regions through carrier injection and surface inversion. These surface-effect mechanisms were critical to the point-contact transistor’s operation. However, they also made the device notoriously unstable, noisy and difficult to reproduce consistently. This led to its rapid replacement by the more reliable junction transistors (such as the grown-junction and alloy-junction types), which used intentional doping to create well-­ defined p-n junctions. There’s more on this in a video on how the first transistor worked at https://youtu.be/RdYHljZi7ys Early attempts by amateur experimenters often involved cracking open a point-contact germanium diode, such as the 1N34, to extract the small piece of n-type germanium, which served as the base. Contacts were then made from thin metal foil or phosphor-­ bronze wire. The emitter and collector points had to be extremely close together; some Australia's electronics magazine siliconchip.com.au Transistors (simple) As with diodes, there are two types Fig.22: a schematic of an early commercial Rochelle salt speaker from 1932. A DIY device can be much simpler. Source: https://rfcafe.com/references/ radio-craft/rochelle-salt-crystal-reproducer-july-1932-radio-craft.htm Plastic frame Spring Plastic Emitter wedge lead Germanium Base lead Collector lead Gold foil p-type n-type Metal base Plastic frame Fig.23: a replica of the original Bell Labs transistor. Prototypes can be very messy! Source: https://w.wiki/KsX4 20 Silicon Chip Fig.25: an improvised point-contact transistor attempt with a germanium wafer taken from a diode. Source: https://youtu.be/5o8B0ekgsxY texts suggest as little as 10 microns (0.01mm) to achieve any gain. A critical step was ‘forming’ the collector by applying a brief high-­ current pulse (eg, 200V from a capacitor discharged through a resistor). This locally heats the surface and diffuses dopants (such as phosphorus from the phosphor-bronze) into the germanium, creating a localised p-type region under the contact. Apart from germanium, materials such as galena and pyrite have also been used in similar point-­ contact setups. Disadvantages of these homemade transistors include low gain (often β < 1), instability, and a tendency to operate in the negative-­ resistance regime rather than as reliable linear amplifiers. An improvised point-contact transistor can be made from the salvaged germanium and whisker from a germanium diode with an additional contact made from phosphor-bronze alloy material, as explained in the video at https://youtu.be/vmotkjMSKnI It is important that phosphor-bronze contacts be used; a phosphor-bronze guitar string, heated to create an oxide layer, has been suggested. See https:// youtu.be/5o8B0ekgsxY (Fig.25) for a partially successful attempt at making a germanium point-contact transistor. Also see the video at https://youtu.be/ wjiSzUa0aHs and Fig.26. Some have suggested ‘paper transistors’ can be made with a graphite pencil on paper but these are not real transistors. While they can show basic conductivity, weak rectification or nonlinear behaviour due to contact effects and impurities, they lack controlled doping, current gain and reproducible transistor action. They function more like crude resistors with contact non-linearity or siliconchip.com.au Fig.26: a point-contact assembly for various diodes and transistors with pyrite. It was noted that the contacts had to be extremely close together for the device to work. Source: https://youtu.be/wjiSzUa0aHs rectifying junctions than as true amplifying devices. Transistors (advanced) The more advanced approach to transistors requires specialised equipment but is still within the realm of possibility for a determined home fabricator, at least in countries where chemicals like hydrofluoric acid can be obtained by non-commercial users. In Australia, many of the required substances are far more restricted (alternatives to HF acid might be possible). Most transistors created by improvisers are Mosfets (metal-oxide-semiconductor field-effect transistors) rather than BJTs (bipolar junction transistors). We’ll explain why shortly. Essential equipment to make a Mosfet transistor includes: • a high-temperature furnace or diffusion oven • chemicals for etching and diffusion (eg, hydrofluoric acid, phosphoric acid & phosphorus oxychloride) • photolithographic equipment, such as a UV light source, photoresist, developer and photomasks (often printed on transparency film) • cleanroom-like conditions (laminar flow hood or glove box) • deposition tools (eg, thermal evaporator or sputtering system for metal contacts) One simplified technique to make a basic planar n-channel Mosfet is as follows: 1. Start with a p-type silicon wafer 2. Grow a thin oxide layer on the wafer in the furnace (thermal oxidation) 3. Coat the wafer with photoresist 4. Expose the photoresist to UV light through a mask to define the pattern 5. Develop the photoresist and etch the exposed areas 6. Use acids or plasma to etch the silicon dioxide layer in the desired regions 7. Introduce n-type dopants (eg, phosphorus) via high-temperature diffusion to create source and drain regions 8. Deposit metal layers (eg, aluminium) for electrical contacts (source, drain and gate) Fig.27: a homemade n-channel Mosfet transistor. Two devices were fabricated on one piece of silicon in case one device didn’t work. Source: https://youtu.be/ s1MCi7FliVY Australia's electronics magazine September 2026  21 Fig.28: a simple switch made from split-pin paper fasteners (brads), paperclip and cardboard or thin timber. Source: https://redfernelectronics.co.uk/diy-switches Fig.29: Peter Parker’s improvised Morse key, made from simple components. Source: https://youtu.be/bJ0V9VzoNuM This is essentially the classic planar process pioneered in the 1960s by Jean Hoerni at Fairchild Semiconductor, then taken to Intel by Robert Noyce and Gordon Moore. This process is very challenging at home, requiring precision, proper safety precautions and patience. However, it has been successfully demonstrated by dedicated hobbyists like Jeri Ellsworth (www.jeriellsworth. com). She has a video on making an n-channel Mosfet at https://youtu. be/w_znRopGtbE Sam Zeloof (https://sam.zeloof.xyz) also made an n-channel Mosfet, as described in the video at https://youtu. be/s1MCi7FliVY (see Fig.27). Some experimenters have attempted to make transistors from zinc oxide rather than silicon, but they have generally been unsuccessful. Transistors – BJTs Homemade transistors tend to be Mosfets because BJTs are far less forgiving of fabrication imperfections. BJTs require ultra-precise, thin, lightly doped base regions and perfect junction alignment because they don’t use the easier-to-control field-effect mechanism at the surface. BJTs demand atomic-level control over doping profiles and junction depths, which is extremely difficult without professional semiconductor fab equipment. As a result, when hobbyists succeed in making working transistors at home, they are almost always Mosfets (or simple field-effect devices). In detail: • BJTs require extremely thin junction boundaries with very precise doping profiles. Any deviation will cause failure or poor performance. Mosfets need only two doped regions (source 22 Silicon Chip and drain) of the same type in a substrate of the opposite type; the channel forms under the gate oxide layer via the field effect, so no ultra-thin, lightly doped base is required. • In a Mosfet, the gate oxide is forgiving and can be grown thermally. Imperfections are tolerated as long as it is continuous and insulating. • Mosfets can function with relatively large gate lengths (tens or even hundreds of microns) and crude alignment (hand-drawn masks or contact lithography), while BJTs require the emitter and collector to be very closely spaced and precisely aligned to the base. That is extremely hard without professional photolithography and alignment tools. • Early point-contact transistors (1947 Bell Labs) worked via surface inversion layers, but they were unstable and hard to reproduce. Mosfets exploit the same surface inversion/ channel formation, but the gate oxide insulates and controls it reliably, making them more forgiving for crude fabrication techniques. As mentioned earlier, Sam Zeloof has had no trouble building working Mosfets in his home lab but has never succeeded in making a functional BJT. Jeri Ellsworth and other DIY semiconductor YouTubers have also made Mosfets but never reliable BJTs. Most improvised transistor successes shared online are Mosfets or JFET-like structures. Making a simple p-n junction diode ‘just’ involves a single interface that Fig.30: the ‘dead bug’ construction style. This technique is fast and flexible. For more information, see the video at https://youtu.be/pkrp98hXCUs Australia's electronics magazine siliconchip.com.au can form somewhat accidentally or with basic heating or probing, while a BJT needs two precisely aligned and controlled junctions, which is much harder. Switches and Morse keys Simple switches or traditional Morse keys can be improvised, but should only be used for low voltages – see Figs.28 & 29. Making a ‘haptic’ Morse key from a nail file is explained in the video at https://youtu. be/guRh36xGCJ8 Wire Making your own wire is not for most experimenters, and would not even be considered for all but the worst-case scenarios, such as civilisational collapse. You would have to find or smelt copper ore. Once smelted, copper can be hammered and twisted into a wire-like form, a technique that was invented about 4000 years ago in the Middle East. Zinc-oxide tunnelling diodes In 2001, Nyle Steiner K7NS rediscovered and improvised the ingenious improvised electronic device now called the zinc-oxide tunnelling diode (siliconchip.au/link/acbr). It is a quantum mechanical device you can make yourself with ease. This type of diode was originally invented by Russian Oleg Losev in 1923. It exhibits negative differential resistance (NDR), a property where increasing the voltage leads to a decreasing current in a certain range – see Fig.31. This allows it to function like a tunnel diode for oscillators, amplifiers and even simple radio transmitters without needing traditional vacuum tubes, transistors or ICs. A typical junction shows n-type NDR behaviour beginning at 100200mV. The region where current begins to increase again at around 300500mV. So, for stable oscillation, the device is biased within the 100-400mV range. The NDR provides gain, similar to a tunnel diode, sustaining oscillations in a tuned circuit (eg, an LC tank). Unlike commercial components, finding the ‘sweet spot’ for the device is finicky and it may need to be readjusted regularly. The negative resistance region effectively supplies energy to cancel losses in a circuit. This allows a simple LC tank to sustain oscillations (as in RF or siliconchip.com.au audio oscillators). It can amplify small AC signals by reflecting and boosting them, a process called negative-­ resistance amplification. That can provide RF gain in radios, with energy coming from the DC bias voltage. It can also generate modulated carriers for low-power transmission if the bias is varied. Unlike ordinary diodes that only rectify, this NDR gives the two-­ terminal device the inherent gain and instability needed for oscillation, or conditionally needed for amplification. This mimics the active role of a transistor or valve in simple improvised configurations. All that’s needed to make this device is a piece of galvanised (zinc-coated) sheet, heat-treated with a blowtorch to make a thick oxide layer, and a cat’s whisker of steel or copper wire. The zinc oxide layer creates an n-type semiconductor layer. When the cat’s whisker is brought into contact with the zinc oxide, a metal-oxide-­ metal junction is created; the zinc oxide layer creates a barrier similar to that in a schottky diode. A device is shown in Fig.32. We will discuss radios recently built with this device in the following issue. Improvised construction methods Circuits can be built on traditional timber breadboards, hence the term “breadboarding”. Modern plastic cutting boards can be used instead of timber. There is also the ‘dead bug’ style of construction shown in Fig.30 (on a plastic cutting board, as it happens). It’s also possible to ‘air wire’ components together over a grounded metal plate, as shown in Fig.33. This can work well, using a similar principle to double-sided PCBs with a ground plane. I i1 rdiff < 0 i2 v1 v2 V Fig.31: a current-voltage (I/V) graph illustrating the voltage-controlled negative resistance property of a tunneling diode. Fig.32: a zinc-oxide tunnelling diode made from a simple piece of oxidised, galvanised sheet with a ‘cat’s whisker’. Source: https://ashishrd. com/2024/02/24/the-quantumtunneling-transmitter Salvaging materials If in a survival situation, you can salvage materials from wreckage. For example, wire and solder can be melted off components or circuit boards for reuse. Numerous circuit components can be salvaged, such as resistors, capacitors and transistors, to make simple circuits like a radio. Next month We’ll finish this topic next month when we look at more advanced comSC ponents and techniques. Australia's electronics magazine Fig.33: components air-wired by soldering their leads together over a grounded tin plate. This also gives them some mechanical support. Source: Peter Parker, VK3YE September 2026  23