Glissade project. These are also known as potentiometer violins, signal generators, audio generators or variable tone generators. Dik Mik of Hawkwind (RIP) played one during the band’s early years, albeit the one he used was a more complex professional piece of Deutsch equipment: some of us need to start somewhere! There's a new hoofbags movie on the way on the subject of electronic circuits’ construction methods. In this we explore a method that has been around for a while. Single or double sided plain copper clad circuit board is cut into strips using a mini circular table saw or PCB guillotine. Then it’s cut into short sections with side cutters and glued onto a baseboard. These form secure heat resistant insulated points onto which component leads can be soldered. The original method of doing this was the drawing pins stuck into wood method, as described below on this hoofbags Web page. Unfortunately, wood is a poor insulator and can cause unstable or non working projects, especially if FETs are used, as in high impedance input stages. I chose a simple circuit as an applied means of demonstration for this construction method, but it would also work with any other application or project. If our viewer has difficulty translating an electronic schematic or circuit diagram into a working piece of electronic equipment, here is a great place to make a start.

The Allen Gordon Low Noise Treble and Bass Pre Amp. Allan Gordon a retail outlet, presumably owned by Allan G, selling guitars, music keyboards and similar, closing during the 80s, Walthamstow, London, E17. Since then, it’s been demolished and private residence is in its place. This was bought from there and it’s just about as simple as a tone control system can get. There’s no boost facility, only bass and treble cut. Twangy, who owns it told the Hoofbag that one of his bands it was used in, the members took turns and argued a bit over who’d use it next. Hoofbag reverse engineered the thing, schematic is presented here. The device takes advantage of the high impedance and low noise characteristic of the FET, a 2N3819. There's a 6.35mm input socket and the unit is powered up when a lead is inserted. the output is a long wire with a jack plug on the end for connection to an amplifier. There will be a YouTube demo uploaded soon and any interested individual can build it., although procuring a 4 or 5 Meg potentiometer, bass control, might be difficult: eBay. Please take into consideration that this design could be copyright, owned by AG, so selling them could be a controversial issue.

These templates are printed on an A4 sheet that can be trimmed and stuck onto a suitable baseboard using double sided sticky tape or another chosen adhesive. From left to right: the glissade, the A4 printout that can be trimmed for the amp also and a photo of the finished one watt output stage that may be copied.

Designing a Simple Transistor Circuit

This may be of interest to electric guitarists who want to become more of an anorak about their craft!

Here we will deal with a ubiquitous transistor BC547, for use in audio frequency applications. It’s the same as a BC 107 NPN, although in a plastic TO92 package for cheapness. Supply is a 9 volt PP3, bias: class A and common emitter configuration. For simplicity let's ignore the 0.1u and 10u capacitors and consider only the DC conditions, it’s all bog-standard Ohm’s Law; V = R X I or R = V /I or I = V/R, and another transistor quantity: hfe. Seen on digital multi-meters, its purpose is to test transistors for gain amounts. Also known as Beta, hfe is the ratio of current (I) difference between the collector and base of a transistor and the higher this figure, the higher the gain. The required collector current depends on the signal level that our transistor needs to work at. This could be up to thirty milliamps (30mA) driving a 65 ohm loudspeaker in an intercom for example. At the opposite extreme transistors used in low level pre-amp stages, may require only 100 micro-amps collector current. In fact, 100uA – 30mA is the approximate usable working current range of a BC547 although it will pass 100mA max Ic, @ full saturation, for our purposes, more than 30mA makes it overheat, causing burns, and at less than 100uA the gain drops below a useful level: In such circumstances we’d need to select an alternative type. We will design a one stage, single transistor amplifier, useful as a guitar tube-driver: This will enable the instrument to push an amplifier into overdrive distortion. 1 mA will suffice for the collector and next comes Ohm’s Law. In the diagram there’s an X and is the junction of collector, RB, RC and one end of the 10uF. When working, point X should be half supply voltage: 9/2 = 4.5Vwith respect to the grounded emitter.  In practice, this could be anywhere between 3 – 7 volts but half supply is ideal. The current, as mentioned, is 1mA: same as 0.001 amps. Press ‘ON’ on your calculator and enter 4.5 (half 9) then press divide - by your current which is 0.001 (1milliamp). You’ll get 4500. This is your value for RC. You’re unlikely to find a 4.5k so instead we use 4.7k; nearest preferred value: hereon abbreviated NPV.

It doesn’t take much imagination to occur that you can feed an output of one stage into another one and get more volume, as done below. This can be done by connecting two or more of these circuits in cascade. From a design point of view it is better to start from the premise that we require a clean signal so we’d then select appropriate resistance values to pass a higher (25mA) current for the last stage: the first stage is left as above and the second would have 180R as RC, 47k as RB. We could substitute the 180R collector resistor with a high impedance loudspeaker. Commonly available loudspeakers of this type are 65 – 80 ohms. It's not capable of marshal 200 watt performance: the best this simple amplifier can manage is a quiet practice amp for use in a personal space, primarily, presented as a means of grasping fundamentals of amplifiers/effect pedals electronic operation.

Stain’s Three Transistor Guitar Amplifier

Like it says  on the bocs ! Drawing pins come in varous sizes but these little cheaper ones are best for these projects. If you use the more common larger ones, they need to be spaced out a bit more, which will involve using a bigger bread-board.

Drawing pin as it looks when separated from it's mates and taken out of the bocs. They need to be kept in a tin with a tight fitting lid to prevent escape: Standing on them barefoor by accident is excruciating !

Calling all inventors: please invent this! Solder fumes contain nasty substances that irritate noses. We need to dissipate them into the surrounding air away from our face, the fan positioned close to our workpiece. Solder extraction gear is costly: this simple device, consisting of an old angle poise lamp frame and a computer fan does the trick.

Wire: so much of this thin solid core stuff is thrown out by telecomm engineers and sold off cheap in electronics shops that it's worth collecting up. It's perfect for these 'pintronics' projects.

This simple amp can be used for other purposes: forming the basis of an intercom or amplifying a crystal radio. Being limited in output power, it overdrives easily giving a crunchy sound desired by some guitarsts. The BC 337-16 is available; 99p for 25 (eBay). They’re obsolete, because the hfe, Beta or Gain is low: the examples I bought from eBay were only 40 - 70. However, the low gain enables simple construction such as the old-fashioned drawing-pins-on-wood method: unsuitable for BC 109Cs or similar high gain types as we may get resistive coupling causing transistor bias problems. This produces poor sound quality/unwanted noises from the speaker: wood is not a great insulator, according to Sir Clive Sinclair, getting worse in damp environments.

For improved screening, also necessary for high-gain transistor  types, we'd need a metal case because wood provides no screening. Metal working makes hard work of our project and is costly. Circuit instability makes noises at the speaker and consumes a power amplifier’s limited output by emitting inaudiable frequencies. Some circuit designers use shunting capacitors in selected places but it may cause other instabilities and reduces the HF response. Terminating resistors can be used, also adding to circuit complexity: to be discussed later. Unstable circuits, though, can be used for sound effects. During project planning, the thought occurred of using Germanium transistors. However, a typical type number such as OC81, is £2.50, being rare.

!

Transistor hfe or gain does more to affect input impedance of an amplifier than make what it's working in louder. Electric guitars have inbuilt volume/tone potentiometers, typically 250k and there's also the parallel wired pick-up coil, so combined matching would probably be less than 100k. For this reason there’s no point in having a 1 Meg input impedance amp for a guitar. If you wanted to connect a Xtal mike or pizo record player P U, then this order of impedance is appropriate.

A hoofbags music/sound engineering trick is to feed an output of one of these amps, via screened lead, to a remote speaker in a box, also containing a microphone. The microphone’s output is then sent back via another long screened lead to our Studio 100. Bassman Sid then gets his over-driven bass sound, rich in harmonics and sustain, desired for Electric Cheese Trolley music.

Interstage Decoupling: On the diagram's left side is a 2.2k resistor and 100mfd capacitor. These filter out any voltage fluctuations caused by Tr2 and Tr3s operation that would affect transistor 1. Missing these components out may lead to 'motorboating', manifesting as an unpleasant noise from the speaker. The right hand 100mfd capacitor helps extend a battery's useful life.

As a continuation of the ‘designing simple transistor amplifiers’ section on this hoofbags Website, we have three of the above circuit (at the start of this section) strapped together on one bit of 5" long wood. This stuff is sold as baton for boxes and is 3/8ths thick, 1‘n half inch wide.  In mm, to be Euro friendly: 127 mm long X 44 mm wide and 7 mm thick. If you used thinner stuff the drawing pins might cause splits, but just about any lump of wood would do. Balsa is unsuitable. One amplifier stage feeds into the next (cascade) and provides enough volume for a transistor guitar amp that is loud enough for practice purposes, and more. It can form the basis of a fuzz pedal, and a means of connecting to a bigger amp, in which case this would be the pre-amp, will be investigated and described in an update. This amp as used here with it's speaker is about as loud as a small transistor radio and will work from a battery such as an Ever Ready PP3 or PP9.

 Wire

We require tinned copper wire and the old style BS3036 used this it's usually suppoied on a card and can probably be bought from electrician shops or even hardware stores. Being tinned isn't essential, a lot of telephone wire comes as plastic coated copper. This works fine. Also eBay: do a search for telephone wire.

Here at hoofbags, and coz we're skint, we like doing stuff on a budget, and it’s also why we still use an old but functioning Amstrad Studio 100 'what Sid rescued from a skip'. If you attempt this ultra-budget design a beginner will understand fundamentals of audio electronic amplification. This may lead to a reader eventually being able to design their own circuits and enable self-repair of amps such as Pignose and effects pedals.There’s another issue: if you go to Maplins you'll see 0.1 matrix board for soldering chips and other parts to, for a self-build. You can use one of the excellent Antex miniature solder-irons but these are hardly hot enough to melt this modern lead-free stuff we’ve all got to buy. When it eventually melts, it's supplied in too-thick a gauge to be used with 0.1” and the chances of causing adjacent track shorts makes the entire endeavour anything but an enjoyable craft hobby. I've lost my camera USB lead. When I find it or get a replacement I will present powerful LM383 and LM386 designs that can be built in this way.

For these projects I've used a Black Spur 30W at £2.00 from ESK. They’ve got a fat bit that's hot enough for unleaded solder. If you’re new to soldering, a temptation is to try conveying molten solder on the iron tip to the joint, using 'melted solder as glue’, plonking it onto a joint. This is incorrect! You must, in the case of a drawing-pin, heat one up already in the base-board, to the point that it will melt the solder upon contact. This is to be done on component leads, too. Don't worry too much about solder damage to transistors or other components: it’s difficult to even deliberately cause component damage when soldering lead-outs. This caution was propagated because early germanium types were more fragile and a budget transistor would cost about a day’s wage.

 Take a bit of 1.5" X 3/8ths" softwood baton and saw it to 5" long. Using a pen, mark it out with 4 vertical divisions, as in the photo with 1" sections. Draw a centre-line horizontally. After that, draw another two that are a quarter of an inch from the top and bottom edges. The line inter-sections are drawing-pin insertion guides.

With drawing pins inserted into the baton: our baseboard, it should look like the right-hand photo. We need to add more pins and should look a bit like the rather out of focus photo below. (appologies!)

Wood can withstand solder-iron tempratures for some time; resulting heat damage is minimal.

It's entirely possible to build the amp from careful examination of the large-view photo avaialble from clicking on the hyperlink picture: right.

Component List:

A high impedance minature loudspeaker: of any value between 65 - 100 ohms. lyn3009 eBay Item number: 360026730614

Three of  BC 337-16 transistors. auntyant2 eBay. Any BC 337 will work but not sound as good. Alternatively: adjust base-bias resistor values.

18 of the smaller size drawing-pins, aka: thumbtacks.

Two of 100 micro-farad capacitors.

A 47 micro-farad capacitor

A 1 micro-farad capacitor (both  can be substituted with 10 mfd but almost any value would work)

1N 914 or 1N 4148 diode (almost any similar type would work)

Resistors: 15k, 3.3k, 270k, 2.2k, 5.6k, 560k.

Other Bits: bit of wood, ball point pen, ruler, 1metre of screened cable with a guitar type jack plug on the end, thin tinned copper wire of about 26 swg or similar, about 1 metre of single flex to wire up the speaker and battery, 9 a volt battery or mains adaptor set to the appropriate voltage.

An essential item: a means of powering up and testing projects in development. Maplin part No: JK09K with crockadile clips attatched. It enables a £3.99 Uniross AD100061 multi-voltage power plug to be used. PSU polarity is determined by the co-axial/two pin pluggie bit's direction. If project don't work upon competion, reverse crock-clips.

Power supply socket connection details (+ pin)

Get your anoraks on, we've some  points to consider: Although this is a simple circuit, there's a surprising amount of stuff going on. For example, if using ohm's law, we were to work out a speaker impedance at 25 - 30 mA it would resolve at around the 160 ohm range instead of 65. However, using this order of collector load would limit output power and cause the circuit to function incorrectly. When the circuit is making a sound, there is an AC (alternating current) content superimposed upon the DC, (direct current) flowing through our BC 337. This AC frequency, determined by which note your guitar is producing, will cause the collector current to fluctuate. Viewed on a suitably configured oscilloscope, it'll be seen that the current will increase to perhaps 41 mA on the peaks and decrease down to, say, 19 mA on the troughs. When reproducing a guitar signal, an overdrive is often accidentally or deliberately opted for so, these fluctuations may be much more. It is therefore appropriate that a speaker load impedance is specified that will more accurately represent a correct load during these transitory peaks. There are other issues occurring: those of thermal runaway. It is a fact that the hotter a transistor gets, the more current flows between its collector and emitter. Unless kept in check, the resulting increase of current causes a further increase in temperature. This cycle will continue, assuming the presence of an uninterruppted supply current of course, until the transistor reaches saturation or the bias conditions render it un-useable. Here, the problem is eliminated because we take our base bias at the junction of the

transistor’s collector, and its load. Now, as the transistor warms up, conductivity increases, and has the effect of a reduced voltage at the collector junction. This causes a matched reduction in the base bias current. It behaves in the same fashion as a speed governor on a steam engine. This arrangement is commonly known as ‘collector-base’ biasing, because the base bias resistor is strapped between the collector and base. In older electronic hobby books, one may encounter simple bias arrangements where the base resistor goes directly to the supply rail. Probably, these circuits were designed at a time before the complicated scientific principles behind thermal runaway were fully understood. I’ve noticed an improvement in eBay available 65 ohm 3” speakers recently. No bad thing at all! It is often considered that, in place of a suitably high impedance speaker, an audio coupling transformer and 8 ohm speaker may be substituted: This may work, or perhaps problems could be encountered. A suitable transformer, especially if it’s large, may not have a sufficiently high primary winding resistance to the DC, which will disable the installed thermal stability measures. If a transformer is opted for, a good choice would be a commonly available Eagle Electronics LT 700, which has a primary resistance of about 72 ohms. However, comparable output results might not be achieved. There is a commonly used transistor biassing arrangement where the collector load matters a lot less. It's called the voltage divider type and uses about twice as many resistors and an extra capacitor. It will be explored in an update. Watch this space! :o)      

The 3 transistor design above is only capable of between 10 - 30mW. That's enough volume for practice purposes but if the guitarist needs an amp for busking, it would need to be louder. Another problem with the 3 transistor design is that it uses up the battery at the same rate irrespective of the volume setting: being of 'class A' biasing configuration. The 5 transistor design suggested here is volume verses battery life inversely proportional because of it's 'class B' biasing. It's capable of 300mW RMS with a 9 volt supply and about 600mW with 12v. It may be possible to get more at higher voltage: up to 14 although higher than this may cause component damage. I'm using 2 mini transformers: LT44 and LT700. These are stocked by Maplin's occasionally, and other sources such as through an eBay search. T1 and T2 should be BC337-16. T3, T4 and T5 can be almost any NPN TO92 plastic general purpose audio, however, for best performance they should be 3 of the same type number and specification. Needless, BC337-16 is ideal. The LED, blue in this instance, serves three purposes in this circuit: it drops 2 volts for intersatge decoupling, lets the user know that it's on and it looks nice. Naturally, in the finsihed project, LED may be front panel mounted and interconnected via wire.

This amplifier's design has been around for long enough: it can be found in every transistor radio up until the late 70s and the legendary Pignose 5 watt is reasonably similar.

Current flow: under 2mA, through LT44's primary is limited via the 2.2k resistor. Without this (via direct wire connection) current woud be about 6mA and the poor little LT44 doesn't like it very much. It will over-heat and go open circuit. The LT700 data sheet recommends a 3.2 ohm speaker, 3 ideal. 3 ohm speakers are a bit rare so 4 ohm will do. An 8 ohm is useable but there will be a reduction in much needed volume. a good size for this would be about 4 - 5". Any size would do. Note also that transistor T4 has its flat side facing the opposite way from the others:  For T4 the flat side must face to the right. Also important is the correct LED connection. In this example the flat denotes the cathode and negative, which faces left. If the light don't come on the amp won't work: try re-soldeing the LED the other way 'round.

There're transistor interstage transformers in existence smaller than LT44s  / LT700s but all are still referred to as minature. One issue about these components' size is that, in order to get enough turn's winding within transformer coils, the wire must be very thin: the wire of an LT44's primary, at about 650 ohms, is as thin as hair! Thinner wire means higher resistance and that reduces the maximum current that can flow  through it. Consequently, in combination with the limited mass of the transformer's iron core, the output power is limited. Larger transformers can be used in this circuit without modification of any other components, with exception of the 2.2K which can be lowered for bigger transformers if necessary and T3, T4 may need heatsinking. These transformers must be of the type specified for transistor driver and output and can occasionally be salvaged form very old transistor radios. 2 or more watts RMS is easily feasable.

When I first built it, it didn't work. However, I tweeked it a bit, changed a few components and it sprang into life. One problem with RF oscillators is that they're very fussy about component type. In this, my first attempt, I used a standard potentiometer for the 50k. These have a metal back cover and the capacitance between it's internals and surrounding components reduced the circuit 'Q' so the oscillations wouldn't start. Using a skeleton pre-set did the trick. Another issue with a radio transmitters is it's not actually very obvious if it has started working or not: you have to twiddle the pre-set potentiometer at the same time as adjusting your radio's tuning. Not the easiest of tasks! The initial finished article is shown in the pane to the right. It is built on a paxoline board with 'Stick On Wiring from :

It should be remembered that using a transmitter of this type in your part of the world may be subject to strict prohibitions: it's up to you to research this before attempting to buid this.

In the UK, AM transmittion is definitely not allowed. We only get away with it here because the power output from this device is so low that it can't adversely affect any other radio or electronic equipment that is being used at a safe distance: it will only transmit up to about ten feet away using a good receiver and aerial.

The supply must not exceed 3 volts and you must not attatch an aerial to the collector of Tr 2, where it connects to the 4700pf capacitor to extend transmittion range. In fact exceeding the 3 volt power supply limit will prevent it from working, so it is pointless to do this unless you make careful adjustments to the resistance values, which you also certainly mustn't do!

There are a few errors in the schematic/circuit diagram: the 150 ohm emitter resistor should be 100 ohms and an audio input, say from an mp3 player must be connected to the left hand side of the 0.1 mfd input capacitor (far left of this diagram).

For garuanteed results it's best to carefully examine the finished board because this is the one that was actually tested as working so ya can't go wrong!

If you're at a college, school or know someone who works with electronics, you probably wouldn't buy solder in small quantities like this. On the plus side, small solder bottles like these can be bought from local hardware shops if you're out 'n about. Another is that the packaging materials can be recycled. The M. T. tube can be used as a coil former for this transmitter. A coil made this way will also work in a radio receiver. Coil winding is a learned skill, just like most

Beginners sometimes find it difficult to wind coils. They get to the end of a winding endeavour only to accidentally let it go at the finish and the whole thing becomes unwound. A way of preventing this mishap is to wrap adheasive tape around the former, sticky side outwards: this will hold the windings in place until it can be terminated by threading through holes at the ends of the former. This coil is made from two four feet lengths of solid core telephone wire. These are tightly twisted together before winding onto the former so that a centre-tap is provided. Enamel coated wire will work just as well but care must be taken to remove the enamel at the coil ends.

The M.T. solder tube is fastened into a vice. Care should be taken to avoid over-tightening to prevent breakage. Next, a 1.5mm hole is drilled at one end. Another one is drilled 10mm down. These holes are used to tie the coil wire ends.

For this transmitter we need a centre-tapped inductor that constitutes the 'tank' coil. In combination with a 470 pf capacitor, it 'rings' or ocsillates at our desired radio frequency in the middle of the MW AM band and is powered by our second 2N2222 transistor. An 8 foot length must be divided into two 4' ones. Each end of the two lengths must be stripped and then re-attached by tightly twisting.

Our coil's Q or sharpness is determined by the neatness of its windings. How close the windings are to each other and the material from which the former is made all have an influence. For ease and chepaness we use telephone wire. It appears that the coil wire length is more critical that the wire thickness. Most types of insulated solid core wire will do.

Another means of obtaining a tank-coil inductor is to use a section of expired felt tip pen. These pens are very cheap, sold in packs of about 50 for a few quid. The coil wire is 26 swg enamel insulated, again two 4' lengths joined at the centre-tap.

Below is the Stylophone circuit diagram as printed in the book that I got with my Stylophone.

Above is the Stylophone circuit diagram. Please note that T1 is not a conventional transistor but a unijunction device. These devices have two bases, B1 and B2 and Emitter. Stylophones are useful as a tone generator in electronic music, not just a toy. Every model of Stylophone has an audio output socket for output to an amplifier or recording device.

Liz Costa. 2E1 FQN

Newham General Hospital, NE London

John (Cap'm) once worked as a private mobile radio (PMR) engineer: For those who don't know, it's the network of radio receivers and transmitters used by cab companies, motorcycle dispatch firms and similar. We were on 24 hour emergency call-out. The transmitters are often located on top floors of high buildings and Newham General Hospital's top floor is one such location. When visiting it for the purposes of transitter adjustment, I (Stain) accompanied John, usually as the toolbox carrying sidekick/misfit. I took a camera up there on one occasion and took these: click on thumbnail for bigger picture. It is quite a big area: an entire floor, the size of the hospital's footprint, devoted to electric fans for ventillation purposes, and other instruments labled 'Autoclave 1, 2 and 3': unusual, I thought because this equipment's usually installed in the operating theatres. The radio transeiver that John was working on had a slight fault although not adversely affecting the equipment's functioning: a volume control had a tendency to not allow the radio 'through' conversations that were emitted from a loudspeaker, to be effectively muted. It's quite spooky when you consider that what all of the cabbies were talking about was audiable throughout this floor, at night and in the dark! I was concerned that this may be a problem and was informed by a security guard that, it once had them all on a wild goosechase because the transeiver was near a ventillation duct going directly down to the morgue: could  the dead had re-animated?  The radio could occasionally and clearly be heard from within there also!

Stain's Pictures - Canary Wharf 50th Floor late 80s