This article describes the construction of a medium wave amplitude modulation 7 transistor receiver. It is built from a kit of parts, sourced from ebay, where delivery from WITHIN the UK should be less than £10. The kit itself is of Chinese origin, and with a Chinese set of instructions, much frustration will be had, unless you read this in conjunction ☺
When searching for the kit, it is known as an HX108-2. It uses a very conventional circuit, which has been around well over 60 years! It must be good…
The kit comprises all the parts needed to complete the radio, including a case, loudspeaker, etc. everything except two AA cells…
A very experienced kit builder would complete this inside two hours. But take your time, learn from it, and most importantly, enjoy it. It is a very good introduction to radio circuitry for the more recently licensed members. For a pocket “tranny”, it performs well. Brings back memories of me listening to Radio Luxembourg in the late fifties!
Construction of the HX108-2
Documentation and parts errors:
Note that the component layout drawing is correct, and the battery negative is connected to the on-off switch/volume control. The circuit diagram incorrectly shows the battery positive as switched. There is a break in the supply to the output transformer (B7) centre tap and you need to add a solder bridge. This is not shown on the circuit diagram, see later. The ferrite rod mounting bracket fits the opposite side of the circuit board from the tuning capacitor. I found it necessary to alter the holes in it so that the screws would fit. V1 may not be the stated transistor, see later.
A small soldering iron is a must. I set mine to 272 degrees centigrade, where leaded solder gives a good, clean joint, without overheating the components (transistors and diodes can be damaged by excess heat, although this is much less of a problem that it used to be). Those using unleaded solder will require a higher temperature.
Multi Cored solder. Do NOT use stick solder and acid flux. Unless you want your handiwork slowly eaten away…
A small Philips screwdriver. Used to tighten the screws on the dials.
A very small screwdriver to adjust the trimmers that are mounted on the main tuning capacitor. A metal screwdriver is OK.
Side (or end) cutters.
A trimming tool. In the absence of the correct tool, a very good substitute may be made by trimming a wooden “lolly stick” to make a blade that fits, but is not sharp. This is used to tune the small metal “cans”. (B2 etc.). A metal screwdriver will a) likely break the ferrite screw, and b) affect the tuning so things alter as you take the screwdriver away. Turn, but do not press down when tuning the “cans”. In days gone by, I nicked my Mum’s plastic knitting needles. And so did my Dad!
A digital test meter is useful, but not completely necessary.
The resistors are small, with a wire at each end, and thin coloured bands around them.
Diodes are even smaller, made of glass, with a very thin black band around the positive end.
Disk ceramic capacitors are small brown disks, with two wires coming out of the edge, quite close together.
Transistors are black, “three legged” devices. As a result of good marking on the circuit board, quite difficult to fit incorrectly, although you do need to place the correct types in the correct places. There SHOULD be three different types:
1) 9018G, “V1”. But I got a 9018H. RF transistor.
2) 9018H, “V2”, “V3”, and “V4”. RF transistor.
3) 9013H, “V5”, “V6”, and “V7”. Audio transistor.
The resistor colour code
For resistors, the first thing you will need to know is the colour code: This will identify their values, which are measured in “ohms”.
No colour band (rare these days): +/- 20%
Silver: +/- 10%
Gold: (Very common, and used in this kit): +/- 5%
Red: +/- 2%
Brown: +/- 1% Getting expensive!
The way the colour code is applied to resistors is quite easy. It works like this:
The first colour band represents a number. So does the second. So if the first colour is BLUE, that means 6. With RED as the second colour, you have a 2. The third colour represents a “multiplier”. In simple terms, the number of zeros. So if the third colour is ORANGE, that represents three zeroes, or 000. Blue, 6, red, 2, orange 000. So that resistor has a value of 62000. Known also as 62k. (“k” = kilo = 1000. For lower values less than a thousand, “R” is used. i.e. 680R is a 680 ohm resistor .For higher values,” M” = Meg = 1000000).
The modern way of writing down resistor values substitutes the multiplier for the decimal point. So our 62k resistor above would be written that way. But suppose the third band was red (00) and not orange? The value would be 6200 ohms. Or 6.2k but a small decimal point is easy to miss. Using the multiplier as a decimal point, there is no argument when our blue red red resistor is shown as 6k2 with no decimal point to miss… 620 ohms would be shown as 620R, 62 ohms as 62R, and 6.2 ohms as 6R2, likewise, 3v5 is 3.5 (or 3 ½) volts…
Why is that resistor 62000 and NOT 32,000000? (Orange is 3, red is 2, and blue 000000). The tolerance is the fourth colour band, so that resistor will read as blue, red, orange, gold and have a tolerance of 5%. The tolerance band is the last one to be read… If you are working on a very old piece of equipment, with 20% tolerance resistors, start at the end where there is no space to put a tolerance band. There will be plenty of room to put a tolerance band on the other end. Except they didn’t…
What if you have a resistor coloured red, red, black? That is 22 ohms. Red is 2, red is 2, and black is NO zeroes.
Red, red, gold, gold.. Red is 2, red is 2, and as a third band, gold means divide by 10.(22/10). So the value is 2.2 ohms. And then there is the fourth gold band meaning +/- 5% tolerance
Red, red, silver, gold? 0.22 ohms. Red is 2, red is 2, silver means divide by 100, (22/100) and gold at the end is the +/-5% tolerance.
This kit does use a 51 ohm resistor. Green brown black gold. But nothing lower than that.
On the supplied instruction sheet you will see printed 1 – 0 and the Chinese characters for those numbers. This will enable you to determine the appropriate coils and transformers. e.g. B5 is the third IF transformer, it has a BLACK core and if you look at the number chart for “0”, you will find the Chinese character for black, and from the parts list, you will find the character matches part number 35. In a similar way, you can find the audio transformers, of which there are two, B6 (with green tape over the windings), and B7 (with yellow tape over the windings).
Electrolytic capacitors are black, tubular, and have two wires coming out one end. The white strip down the side indicates the negative connection, and this is important.
Ceramic capacitors have their values stated numerically, but also use a “multiplier” system. If a capacitor has “223” written on it, its value is not 223 picofarads. It is 22,000 picofarads, the three indicating the number of zeroes (“000”). 0.022uF is the same value. So where the parts list shows 0.022uF, you are looking for a small brown disk with 223 written on it. Note that the main tuning capacitor (C1) has a value of 223p shown, and this IS 223 picofarads. That “p” is important…
Further points to consider. Look at the collectors of the transistors. (On the drawing, these are the uppermost connections, except for the output stage, where V7 is shown “upside down). Follow the line upward, there is an “X” on the drawing. This is a break on the circuit board, (two small pads, no holes) and allows you to connect a milliammeter. The values you can expect are printed above the top of the diagram. When you have tested this, you bridge the two pads with solder.
Note that the centre tap of B7 output transformer does NOT have an “X” shown. (That is the middle connection of the transformer, on the left hand side as you look at the drawing). But there is a break that needs to be soldered across the two pads. Else why would they print 4-10 mA… So there are a total of five solder bridges to be applied…
Now you have got this far, check, and DOUBLE CHECK that the batteries are inserted correctly, and that the battery terminals are correctly wired. The positive end of the battery connects to the circuit board as shown on the layout diagram, noting that the NEGATIVE end connects to one side of the on-off switch, also as shown.
V1 is a 9018G, a lower gain device than the 9018H used for V2, V3, and V4. (But V4 is used as a diode). This MAY give trouble if, as in my case, all my “RF” transistors were 9018H, higher gain devices. My radio is quite “lively”.
My B5 (Black, last IF transformer) was “finicky”, adjusting it, the threads sometimes slipped. So for alignment leave B5 alone if possible:
See if you can tune anything, even noise from a local switched mode PSU if need be. Peak B3 (Yellow) and B4 (White) for “maximum smoke” (noise, or music, if you are that lucky!). Thereafter, these should not need adjustment. Peak B5 if you really have to.
If you are getting music, it’s working! However, the alignment is not complete. There are two small trimmers on the main tuning capacitor. Adjust one of them. If you can sharply tune a station in and out using it, this is the oscillator trimmer. Note that. If the trimmer just makes a volume difference, this is the antenna trimmer. Again, note.
For correct alignment, Adjust the RED (B2) core to tune a station to its correct dial position on the lower end of the band. (Towards 535 khz). When receiving the station, move the tuning towards where the station should appear, and tune the red coil until you are receiving it ok. Repeat as necessary. When you have done that, slide the antenna coil up and down the ferrite rod for maximum signal.
Then tune to a station at the top of the band (1605 khz). Again, adjusting the dial and oscillator TRIMMER until the station appears on the dial where it should, then use the antenna trimmer to get maximum signal.
You will need to do the last two steps several times, in that order.
Now, you have learned how to align the “tracking” on a medium wave superhet…!
And hopefully, got a “cheap and nasty” tranny…! (You built it, you’ll love it, believe me!).
This could prove to be an interesting constructional project. So here are the rules:
Marks for each category will start at 10, with marks deducted as appropriate if, in the opinion of the judges, there are deficiencies!
1) Soldering quality. Not too much, or too little solder. No “burnt flux” on the circuit board. No marks will be deducted for the use of unleaded solder, even though it will probably result in a less that perfect appearance.
2) Component placement. All components should stand bolt upright. It is clear which hole the longest wire from the top of a resistor goes. Reversing this loses marks, as will components leaning over.
3) Resistor orientation. If it is decided to place the resistors such that the gold bands are nearest the circuit board, then marks will be deducted per resistor if fitted with the gold band upward. Vice versa will apply; should you choose to fit the gold band highest, then they should all be that way.
4) Electrolytic capacitors MUST be fitted around the right way. They will fail if incorrectly fitted. A white band runs down the NEGATIVE side of the capacitor, and their location on the circuit board has a circle, half of which is white, and is where the negative end goes.
5) Diodes are also polarity sensitive, and must be fitted correctly. The + end of a diode is the end marked by a very thin black band. Get this wrong and lose 3 marks per diode! (And it won’t work).
6) Disc ceramic capacitors. These are not polarity conscious and may be fitted either way. HOWEVER… They have numbers on them indicating their value, and after fitting, these numbers should be visible, not masked by another component if possible. In addition, unless it is necessary to place them so that the numbers are visible, all capacitors should face the same way. (All those with the edge pointing towards the antenna, and all those “across” the circuit board.
7) Glue should be used to secure the large “battery spring”. If it can fall out, that is lost marks. Melting the plastic over the spring using a soldering iron is also acceptable.
8) The antenna coil should be secured to the ferrite rod. A SMALL dob of candle wax is often used, and done AFTER alignment. Do not use a permanent adhesive. A sliver of a matchstick gently introduced between the coil, and the ferrite rod is also acceptable. (Or a shaving from your tuning tool!).
9) General construction. Burning something with the shaft of the soldering iron will lose marks. Do not fit the screw holding the circuit board in, (which will need to be removed for judging). Loose screws on the controls, and a misplaced label for the tuning control, and the cabinet scale will also lose marks.
10) There are four points to be had, if you include with your radio the correct answer to: How you would write a value of 3,300000 ohms, using a letter as a decimal point. Six more points if you get it right for a resistor of 0.56 ohms
This kit, and the competition, is about fun, and in particular learning, whilst giving confidence to the more recent licensees. It offers component identification, and use of the electronic colour code, along with constructional and soldering experience. For those reasons, if the radio does not work, you will only be penalised one mark. Have no fear, one way or another, unless you threw it under a bus, it will eventually work, Hi! ☺ If it works, but not particularly well, maybe you didn’t get the alignment quite right, but no marks will be deducted.
The reason that justifies many of the rules is because if you subsequently have to fault find, the ability to identify components and their values greatly simplifies the process.
Did you check that the expected currents are flowing with a milliammeter? If you don’t have a test meter, just bridge the solder points. There are 5, and any one not bridged will stop it from working. Look very carefully at the wire from the top of the resistors, to where they go into the circuit board. Some are “quite close” to the “cans”, and if they touch, it won’t work.
You did ensure that the battery is connected and installed correctly before switching it on? Check carefully for short circuits on the track side of the circuit board. There are five places they should be, and if there are any more… What about broken wires? The antenna ferrite rod wires are very easily broken, and even the speaker and battery wires are very thin. Check also you have the antenna coil wired correctly, there are four wires but the correct connections are clearly shown on the layout diagram.
Any connections you forgot to solder? Sometimes, there is a “hidden link”. That is to say, the can body connections carry an earth from one place to another, so if you didn’t solder both, another part of the circuit may have an earth missing. Have you got the right transistors in the right place, the diodes around the right way, and all other components correctly connected. With the transistors, some need the middle leg bending slightly towards the “flat” face of the transistors (where the numbers are written). Some need to be bent the other way. Carefully inspect the picture on the circuit board itself, it is very easy to see how they should be installed.
Now, how does this radio work?
First off, a word about how transistors work. A very small amount of current flows into the base, and causes much more current to flow through the collector-emitter part of it. Generally, a small current (called the “bias”) is always passed into the base, causing there to be a “standing current” in the collector emitter circuit. Fluctuations in the base current, by adding a signal to it, cause much larger fluctuations in the collector-emitter circuit, hence we have “gain”. An amplifier…
This radio uses amplifiers. It also uses an oscillator. An oscillator is an amplifier, whose output is fed back to its input, causing it to generate a signal, in much the same way as the “howl” that results on a public address system, when the microphone is too close to the loudspeaker.
First off, the signal is picked up by the ferrite rod antenna B1. This is tuned by C1(a), to the frequency of the station required. As the circuit is parallel tuned, it is high impedance. The input of V1 is low impedance, so just a few turns of wire are used to lower the impedance, and couple the signal to V1.
V1 amplifies the incoming signal. As a result of coil B2, the output of V1 gets fed back into it, and it oscillates, making a signal. This signal is 455kHz higher than the incoming signal, and mixes with that signal. The result is that the output of V1 contains FOUR signals:
1) The original signal. (No longer of interest).
2) The oscillator signal. (Again, no longer of interest).
3) The difference BETWEEN 1) and 2). (We need this).
4) The SUM of 1) and 2). Of no use.
Signal 3) above, is coupled via B3, to V2. B3 is ALWAYS tuned to 455kHz. We are now beginning to get a bigger signal, and we are rejecting those we don’t want. The amplified output of V2, (consisting mainly of what we have tuned to) is fed to V3 via B4. As B4 is also tuned to 455 kHz, further rejection of unwanted signals take place. The amplified output of V3 is fed to B5, where further signal selection takes place.
As B3, B4, and B5 have tuned parallel circuit primaries at high impedance, they have low impedance secondary’s to couple to the next transistor.
V4 is an interesting circuit. This transistor is used as a diode, and has no electrical supply. At this point, the desired signal may be several hundred millivolts of amplitude modulated radio frequency. Being diode connected, it rectifies (detects) the signals applied to it, recovering the audio signal modulation. It has another function however, if it sees a signal that is big enough to cause receiver overload, it causes some of the bias for V2 to be “stolen” via R8. This causes V2 to have less gain, removing the overload. This is known as “AGC”. (Automatic gain control). All receivers need this type of circuitry, although it can be somewhat more complex.
The emitter of V4 contains audio, and rectified RF. The RF is shunted to ground (got rid of), by C8 and C9, leaving audio to be fed to the “top” of the volume control. The “slider” of the volume control feeds audio via C10 to the base of V5, the audio driver transistor, which amplifies the audio and feeds it via B6 to the bases of the output transistors.
The output stage is configured in a way known as “push-pull”. The effect of this is that on one half of an audio signal cycle, V6 conducts, feeding a signal into B7, which couples it to the loudspeaker. At this point, the signal causes V7 to be cut off, and it does nothing.
When the other half of the cycle appears, V6 cuts off, V7 conducts, and feeds that half of the signal into B7 and the loudspeaker.
There are advantages to this circuit, which I first saw around 1959! The use of one transistor would require it to be taking a fair amount of current all the time so that it could both increase the current it took on a positive half cycles, and reduce it to cope with negative half cycles. Battery life would be an issue, as is transistor heating. Indeed, with one transistor, the general efficiency would be 33% at full volume, falling to 0% during no audio, with varying efficiencies as the volume changed.
With that circuit being known as “class “A”, this radio use a class “AB” circuit. It takes a very small amount of current with no audio, and the current rises as the audio gets louder. At high volumes, 50% efficiency is achievable. And the output stage only warms up at high volumes, whereas the class A circuit cools a little at high volume. As a kid listening to the Beatles, my Dad encouraged me to run my radio loud. I now know that it was maybe to overheat my output stage and destroy it, or eat up my batteries faster, Hi!
In the “old days”, at high volumes, a disconnected loudspeaker could result in a damaged output stage. Yes, VSWR also exists at audio… Not a problem with more modern “transformerless output stages. But I would not run this at high volume with no loudspeaker…
Later on, I will be experimenting regards rewinding the antenna coil to make it tune to around 1.91 Mhz, when the radio tuning dial is set to 1Mhz. Because at a DIAL frequency of 1Mhz, the local oscillator (without needing modification) is running at 1.455 Mhz. 455 khz (The intermediate frequency) above that, signals around 1.91 Mhz may be heard. There is a hole in the case, I am assuming the supplied wrist strap could be fitted here. Or a small jack to connect an external antenna to a few turns around the ferrite rod….
Try a few turns to couple into the existing ferrite rod, even if you don’t modify it for top band. You may start to hear distant stations. Or overload poor little V1 to hell and back, Hi!
73 de Stan, G4EGH