Evolution of Radio Receivers Part 3: Build a Classic 6-Transistor Superhet Pocket Radio

Introduction

In Part 2 we explored the superheterodyne principle and why it became the dominant radio architecture for nearly a century. We saw how converting the incoming signal to a fixed intermediate frequency (IF) dramatically improves selectivity, gain, and stability compared to the earlier Tuned Radio Frequency (TRF) designs.

In this part, we will examine a real, practical implementation that brought the superheterodyne receiver into the hands of millions: the classic 6-transistor AM pocket superhet radio. This simple, elegant circuit became one of the most successful and widely built radio designs in history. It is still an excellent project for today’s electronics hobbyists.

Why focus on the 6-transistor pocket radio?

• It is one of the purest and most representative implementations of a working superheterodyne (superhet for short) receiver.
• Almost every block we discussed in Part 2 (RF preselector, mixer, local oscillator, IF amplifier, AGC, detector, and audio amplifier) is clearly visible and understandable.
• The circuit uses only discrete transistors and standard components, making it an ideal project for hands-on learning and home construction.
• Many readers already own or can easily find these radios (or their kits) at low cost, perfect for experimentation, alignment practice, and modification.

In this article, I will provide detailed circuit description, explain the function of each stage with component values, share practical construction tips, and walk you through the complete alignment procedures step by step. By the end, you should be able to build, align, and understand your own working superhet pocket radio.

Circuit Overview

The 6-transistor pocket radio is a classic single-conversion superhet design operating on the medium wave (MW) band (typically 530–1600 kHz) with a standard 455 kHz intermediate frequency.

Here is how the signal flows through the radio:

Block Diagram (Fig. 1)

The circuit can be divided into the following functional stages:

• RF Input Stage — The ferrite bar antenna is also the inductor that forms the tunable input circuit (RF BPF) with the variable capacitor.
• Frequency Converter — In the classic 6-transistor pocket radio, a single transistor performs the functions of Mixer and Local Oscillator.
• IF Amplifier — Two stages of high-gain 455 kHz amplification with AGC
• Detector & AGC — Diode detector that also generates Automatic Gain Control voltage
• Audio Amplifier — Consists of a driver and a class B push-pull power amplifier to get sufficient volume.

This is an extremely efficient design. Three transistors handle the RF and IF tasks, while three provide enough audio power for comfortable listening. The circuit makes use of Automatic Gain Control (AGC) so the radio has sufficient gain to receive weak distant stations and can handle strong the strong signals from local station without high distortion due to overload.

In the following sections, I will explain each stage in detail, including the role of every important component, typical values used in original designs, and the reasoning behind them. We will also look at how the stages interact with each other.

Detailed Circuit Description

Refer to Fig. 2 the schematic circuit and Fig. 1 the block diagram for the following discussion. I also include a link for you to download a hi-res schematics in pdf.

Converter

Q1 is the frequency converter which combines the functions of mixer and local oscillator. Signals picked up by the ferrite-bar antenna L1 (coil terminals A1-A2) is coupled to the base of Q1 via coil terminals B1-B2. Fig. 3 shows the connection of the ferrite-bar antenna coil and the 2-gang variable capacitor (VC) C3. L1 together with the antenna gang of C3 forms a tuned circuit which selects the desired signal to be received. This is a simplest preselector (the RF BPF in the block diagram Fig. 1) which attenuates all signals from the antenna except the one at its resonant frequency. Coil B1-B2 is wound on the same ferrite-bar adjacent to coil L1 (A1-A2), forming a transformer. In this design L1 (A1 – A2) has 100 turns; B1-B2 has 10 turns. Since impedance transform ratio is the square of turn ratio, the low input impedance of Q1 (1~2 kOhm) is multiplied 100 times. This greatly reduces loading to the tuned circuit and thus maintains reasonable selectivity. Q1 also functions as the local oscillator. It works as follows. Coil B1 – B2 only has 10 turns and so its reactance to the local oscillator frequency is negligible. C4 provides effective RF grounding. Therefore the base of Q1 is effectively grounded for the oscillator signal. Oscillation frequency is determined by L3, C11 and the oscillator-gang of variable capacitor C3. R17 (56 Ohm) sets the oscillation level. C10 provides DC blocking and feeds the oscillator signal to the emitter of Q1, completing the feedback loop. C11 is the padding capacitor which is needed when the oscillator-gang and antenna- gang of the C3 have equal capacitance. To determine the value of C11 involves quite many calculations. This is known as tracking calculation. I will discuss this topic in details in Part 4 of this series. For the current project, C11 is replaced by a short-circuit, as the oscillator gang of the VC is deliberately designed to have lower capacitance so that the local oscillator frequency is always higher than the received frequency by roughly an IF (455 kHz) for its entire tuning range. Such ganged variable capacitor with difference in capacitance were used exclusively in single-band (MW/BC) pocket radios from early 60s to 90s.

At the collector of Q1 there are several signals that include local oscillator signal (fL), incoming RF signal (fR), the sum signal (fL+fR) and the difference signal (fL – fR). As the primary of IF transformer (IFT) T1 is connected in series with the secondary of L3 to the collector of Q1, it selects the difference signal (455 kHz) and attenuates all others. Resistor R6 (100 Ohm) is needed to prevent parasitic oscillation.

IF Amplifier, detector and AGC

The secondary of T1 is connected to the base of Q2 which is the first stage of the IF amplifier. The collector load of Q2 is another IF transformer which secondary drives Q3 which forms the second stage IF amplifier. The secondary of IFT T3 is connected to D1 which functions as an envelop detector to recover the audio signal. R16, C8 and C9 remove the IF (455 kHz) component. While all the IFTs (T1 thru’ T3) resonate at 455 kHz, they have different characteristics due to the difference in turn ratio. There was no standard specifications, except the resonance frequency, as the engineers of the early days radio makers tweaked their circuits trying to squeeze out every dB of performance. Through the years the industry somehow converged to a most accepted design in terms of gain distribution. This resulted in the difference in turns ratio of the 3 IFTs. Fig. 4 and Table 1 illustrates the typical specifications of such 455 kHz IFTs for an AM receiver. These IFTs can still be found in eBay and some electronic component stores, but you have no control on the turn ratios. Fortunately even if the turn ratios are not optimum, the set will most likely still function with slightly reduced performance. It is interesting to note that there was an almost universal agreement right from the start: the first IFT (the one that follows the mixer) has its tunable ferrite core painted yellow; the second IFT painted white; the final IFT (the one that precedes the detector) painted black. Moreover most, if not all, 455 kHz IFTs have an inductance of 680uH, unloaded Q of 80~120, and a internally connected capacitor of 180pF.

	IFT1	IFT2	IFT3	OSC
Color marking	yellow	white	black	red
inductance	680uH	680uH	680uH	360uH
Unloaded Q	90	90	90	100
No. of turns N1-3	175	175	175	122
Turn ratio  N1-3 : N1-2	2.8	3.1	3.7	20
Turn ratio  N1-3 : N4-6	22	18	5	10
Internal tuning capacitor	180pF	180pF	180pF	none

Refer to Fig. 5 for detector action. The recovered audio signal has a DC component which magnitude is proportional to the signal strength of the received RF signal. A strong RF signal causes a large negative DC voltage at the anode of D1. This DC level alters the bias of Q2 via R7. C12 has large capacitance (10uF) in order to remove the detected audio signal. The gain of a transistor is proportional to its emitter current over a range from several hundred uA to several mA. A strong RF signal results in a lower bias voltage at the base of Q2. This results in less emitter current and hence lower gain. As a result, the signal level at the detector input remains relatively constant when the RF input signal fluctuates within a certain range. The net result is a relatively constant volume from the loudspeaker no matter the received station is weak or strong.

Audio driver and PA

The audio signal from the volume-control variable resistor (P1) is boosted by Q4 to drive the push-pull power amplifier formed by Q5 and Q6. This is a typical transformer-in-transformer-out class B push-pull circuit which is most popular for pocket radio receivers of the 60s and 70s.

Sourcing of Key Component

Refer to Table 1, Table 2, Fig. 3 and Fig. 4 for the following discussion.

The major issue is to obtain several critical components that include ferrite-bar antenna (L1), 2-gang variable capacitor (C3), local-oscillator coil (L3) and the IFTs (T1~T3) with appropriate pin-outs. The ferrite-bar antenna coil (L1) and variable capacitor are easy to identify (see Fig. 2). They can still be obtained from component stores or eBay. The oscillator coil and IFTs are more involved. With reference to Fig. 3 and Table 1, we can estimate the turns ratio of the primary/tap/secondary windings of L3 and T1~T3 by measuring DC resistance using a Digital Multi-meter (DMM). The exact value is not important. We care only the ratios. For example, the DC resistance of pin 1 to pin 2 of the oscillator coil L3 is about 0.3 Ohm; pin 2 to pin 3 is about 3.5 Ohm; pin 4 to pin 6 is about 0.5 Ohm. Here the key is that pin 2-3 has much higher resistance than pin 1-2, because pin 2-3 has much more turns; that is what we need. Similar rationale applies to the IFTs T1~T3.

For oscillator coil L3, the winding sense (direction) also matters. For instance, if the secondary winding starts from pin 6 instead of pin 4, oscillation will not be possible. Even though Fig. 3 shows the most widely accepted pin-outs, exceptions do exist. In that case, swapping pin 4 and pin 6 will solve the problem. Refer to the troubleshooting section for details.

Fig. 3 shows the typical ferrite-bar antenna (coil) and 2-gang variable capacitor for pocket radios. These are still available from some component stores. Capacitance of the antenna-gang is 8~150pF; oscillator-gang is 8~65pF. Inductance of antenna (coil) is about 560uH and is tunable within 10%.

The detector diode is preferably a germanium point-contact diode such as 1N60, OA90, etc., which has been obsolete for a long time and so hard to find. Recently so-called 1N60s appear in eBay and some component stores. These new 1N60s look more like a 1N4148 than an old 1N60. See Fig. 6. I bought a few pieces and tested its forward voltage drop (VF) using the diode range of a DMM. To my surprise, VF is 0.26V. (VF of the “old time” 1N60 is 0.29V.) I guess the new 1N60 is actually a small signal Schottky diode, which has higher capacitance. Anyway this is not an issue for detecting 455 kHz signals. We care only about the low forward voltage drop. A small signal Schottky diode is totally acceptable. On the rare instant that you cannot find a Schottky diode, a common silicon diode such as 1N4148 can be used, with slightly loss in sensitivity.

As a concluding remark on component sourcing, I recommend to obtain a radio kit, which is still available from some component stores. This saves you the trouble of improper parts. You may then ask, “why don’t I just build the kit”? My answer is: you absolutely can, but the performance may be inferior to the circuit in this post, and the instructions with the kit may not be as detailed as what I give here.

Construction Tips

I design a PCB to build this radio circuit. Fig. 7 is the PCB pattern viewed from the component side. It can be used as the mask for use with positive per-sensitized blank PCB. Please use the link below to download the pdf file which is the 1:1 PCB pattern. Check and make sure the dimensions are correct. It can be easily done by checking the 60mm width. If you prefer to use PCB fabrication services, I can provide Gerber and NC-drill files. Just drop me a line in the comment field. (WordPress does not allow uploading zip or rar files to a blog.) Fig. 8 is the component layout drawing, viewed from the top side. Don’t forget to connect the only one jumper wire (shown in yellow color). If you copy the PCB, construction is straight forward. Breadboarding is also feasible provided the length of connections are kept to minimum.

Test and Alignment

Tools and Equipment Needed

Refer to Table 3 and Fig.9 for the tools and equipment you may need.

Item	Purpose	Remark
Non-metallic Tuning Tool	To adjust the coils and trimmer capacitors	essential (see Fig. 9)
Lab DC Power Supply	To provide precise supply voltage	essential
Digital Multi-meter	General measurements	essential
RF Signal Generator	To precisely align the set	highly desirable
Inductive Test Loop	To couple RF signal to the antenna of the set	desirable (see Fig. 9)
Oscilloscope	Check oscillator signal, debugging	desirable

Checking DC Voltages

Apply DC power (9V +/- 0.1V) and turn on. Set the volume control (P1) to mid position. Noise and/or audio from a radio station should be audible at the loudspeaker. Tune the variable capacitor (C3) so that no station is received and only noise is heard. This procedure is necessary because we are going to measure the DC voltages at several test points when the set is at maximum gain as AGC has not yet kicked in. Measure the DC voltage at points A through E (refer to Fig. 2) using a DVM. The values should be within 20% of that indicated in the schematics. If not, check for incorrect or misplaced components.

Easy Tune-up (when a RF Signal Generator is not available)

The method described in this section is aimed at hobbyists who do not have a RF signal generator (RFSG). This method will not accurately align the radio, but is good enough to let you receive most stations in the band. Actually I used this method 50+ years ago when I built my first superhet radio. Most hobbyists in the old days did so, as a RF signal generator was just too expensive then. This method works under the assumption that there are broadcast stations close to the low-end and high-end of the BC band. This is usually the case. Moreover we assume that the IFTs are already tuned to 455 kHz by the component makers (usually true).

1. Set volume control to mid position.
2. Check and confirm that in your area there is a broadcast station close to the low-end (530 kHz), and one close to the high-end (1650 kHz). The key idea is to use the stations whose frequency is known to set the band coverage. For example, if there is a station at 567 kHz, use it to set the low-end receiving limit. If not, choose the nearest higher up station, e.g., at 603 kHz. Similar rationale applies for setting the high-end receiving limit.
3. Assume there is a broadcast station at 567 kHz. Set the variable capacitor (C3) for maximum capacitance and then back off a little bit. Mark this position and call this frequency f1. Tune the oscillator coil (L3) to receive this station with maximum audio output.
4. Adjust L1 by sliding the coil (L1) along the ferrite bar for maximum volume.
5. Assume there is a broadcast station at 1512 kHz. Set C3 for minimum capacitance and then back off a little bit. Mark this position and call this frequency f2. Tune the trimmer capacitor at the oscillator gang of C3 to receive this station with maximum volume.
6. Adjust the trimmer capacitor at the antenna gang of C3 for maximum volume.
7. Set C3 to the f1 mark. Verify that the 567 kHz station can be received with maximum volume. Re-adjust L1 if necessary.
8. Set C3 to the f2 mark. Verify that the 1512 kHz station can be received with maximum volume. Re-adjust the trimmer capacitor at the antenna gang of C3 if necessary.
9. Procedures 7 and 8 may need to be repeated for a couple of times.
10. Secure the position of L1 by melting some wax over it to glue it to the ferrite-bar.

Note that we deliberately avoid adjusting the IFTs (T1~T3). Most makers tune their IFT to 455 kHz before shipment. The loading of Q2 and Q3 certainly will detune the IFTs, but only slightly. My experience is that adjusting the IFTs without a signal generator will only make things worse.

Full Professional Alignment Method

1. Set volume control to mid position.
2. Set variable capacitor (C3) for maximum capacitance.
3. Connect a commercial “standard test loop” or home-brew substitute (see Fig. 9) to the output of a RF signal generator (SG). Refer to Fig. 10 for how to position the test loop.
4. Connect an AC voltmeter and/or an oscilloscope to the loudspeaker terminals (pin 4 – 6 of T5). You may replace the loudspeaker with a 8 Ohm dummy load.
5. Set the SG to 455 kHz, RF level -47 dBm, 30% AM modulation, 1 kHz audio.
6. Adjust the IFTs (T1 thru’ T3) for maximum level of the sine wave at the loudspeaker terminals. You may need to alter the SG output level to obtain an optimal audio output. Optimal here means not too strong to cause obvious distortion or too weak that the audio is noisy.
7. Set the SG to 530 kHz, RF level -47 dBm, 30% AM modulation, 1 kHz audio.
8. Tune the oscillator coil (L3) for maximum amplitude of the sine wave at the loudspeaker.
9. Set the SG to 1650 kHz, RF level -47 dBm, 30% AM modulation, 1 kHz audio.
10. Tune the trimmer capacitor at the oscillator gang of C3 for maximum amplitude of the sine wave at the loudspeaker.
11. Procedures 7~10 may need to be repeated several times to ensure the designed receiving range (530 kHz ~ 1650 kHz) is precisely set. The volume setting and SG RF output level may be altered to ensure a distortion free output.
12. Set the SG to 600 kHz, RF level -47 dBm, 30% AM modulation, 1 kHz audio.
13. Tune C3 to the 600 kHz position on the frequency scale. Adjust L1 by sliding the coil (L1) along the ferrite bar for maximum output.
14. Set the SG to 1500 kHz, RF level -47 dBm, 30% AM modulation, 1 kHz audio.
15. Tune C3 to the 1500 kHz position on the frequency scale. Adjust the trimmer capacitor at the antenna gang of C3 for maximum output.
16. Procedures 12~15 are the tracking adjustments and may need to be repeated several times.
17. Secure the position of L1 by melting some wax over it to glue it to the ferrite-bar.
18. Set the SG to 1500 kHz, RF level -47 dBm, 30% AM modulation, 1 kHz audio. Tune to this frequency. Increase volume until the output sine wave just start to clip. This roughly corresponds to 10% THD (total harmonic distortion). The peak-to-peak audio output should be about 4V. This is the maximum audio output and is roughly 250mW into 8 Ohm. DC current consumption under this condition should be about 60mA.

Remark: I use a home-brew test loop to align the radio with good results. The only drawback compared to using a standard test loop is that it is not calibrated and so cannot be used to measure sensitivity.

Troubleshooting Tips

First of all, check and verify that the DC voltages at various points (as indicated in the schematics) are within limits. If not, check for short or open circuits, and then misplaced components.

If DC check is okay but you hear no audio from the loudspeaker for the whole tuning range, the most likely problem is a dead local oscillator. As mentioned in the component sourcing section, oscillator coil L3 is rather critical. If the winding sense is incorrect, you will get no oscillation. The symptom is a complete silence across the whole band. This can be easily confirmed if you have an oscilloscope. Connecting a 10X probe to the emitter of Q1 (point A in the schematics), you will get a sinusoidal signal which amplitude is about 1V peak-to-peak. Any value from 0.7 to 1.5V is acceptable. Tuning C3 will change the frequency and oscillation should be 0.7 to 1.5V for whatever value set for C3. If there is no signal, the oscillator is not working. The most probable cause is incorrect winding sense of L3 (primary vs secondary). To fix it, swap pin 4 and pin 6 of L3. You will need to cut PCB tracks and rewire using jumper wires which must be as short as possible.

If a few stations can be received but the correct tuning range cannot be set, the value of either the variable capacitor C3 and oscillator coil L3 is inappropriate. Check the capacitance of C3. The oscillator-gang should be about 8~65pF. Checking the inductance of L3 is more difficult, as most DMMs do not have inductance range. I will write another post on how to measure inductance (and Q factor) using the affordable NanoVNA. Stay tuned.

Listening Test /Real-World Performance Observations

With the radio fully aligned, it was time for the real test — firing it up in the wild. The difference from the 2-transistor reflex TRF receiver in Part 1 was immediately obvious and deeply satisfying. Weak and distant stations that were marginal or buried in noise on the TRF design now came through clearly. Selectivity was a huge improvement too: adjacent-channel interference that plagued the TRF set was now largely tamed thanks to the multiple tuned IF stages at 455 kHz. Thanks to AGC action, the different in volume among the local stations is not noticeable.

A major drawback of this radio, as well as all pocket radio is lack of tuning accuracy. The frequency dial is small and it is directly attached to the shaft of the variable capacitor. The small dial and direct-drive variable capacitor make precise station selection feel more like guesswork. A few degrees of knob rotation can skip right over a station. Interestingly, pocket radio receivers remain highly popular in the 60s despite this weakness. This might be due to their relatively low price tag and they were mostly for reception of local stations which were not too close to each other.

In larger portables and table-top models, manufacturers added mechanical reduction drives and larger dials. Fig. 11 shows one classic implementation using a cord-and-drum system. I’ll cover these tuning mechanisms in detail in Part 4, along with proper frequency tracking calculations that make multi-band superhets practical.

Overall, this little 6-transistor superhet is still an outstanding performer for its era. It delivers good selectivity, high sensitivity and reliable AGC in a package an average hobbyist can build. For many hobbyists in the early 1970s (including a young me), successfully aligning and hearing distant stations on a home-built superhet was a true “I did it!” moment that hooked us on RF design for life.

Conclusion: A Foundational Milestone in Receiver Evolution

Building and aligning this classic 6-transistor superheterodyne pocket radio is a rewarding milestone for any RF hobbyist. It beautifully demonstrates the principles we explored in Part 2 — frequency conversion, fixed-IF selectivity, and AGC — while delivering performance far superior to the simple reflex TRF design from Part 1. In my early 1970s builds, the moment a weak distant station emerged clearly after careful alignment was pure magic.

This simple circuit, despite using only six transistors, delivers surprisingly good performance for AM broadcast reception and has been a favorite of hobbyists for decades. More importantly, it serves as the foundation for almost every more advanced receiver we will explore in later parts of this series.

In Part 4, we’ll tackle the next big challenge in superheterodyne design: precise frequency tracking across the band, the mathematics behind padding capacitors, and the clever mechanical tuning systems that brought real dial accuracy to portable and table-top radios. Stay tuned — this is where the designs start getting truly professional!
For a complete overview of the series and what’s coming next, visit the Evolution of Radio Receivers Series Roadmap on the series home page.

Whether you’re a student, seasoned hobbyist, or simply nostalgic for these pocket classics, I hope this hands-on project inspires you to build, experiment, and improve these timeless circuits. In upcoming parts we’ll explore frequency tracking, advanced filtering, image rejection techniques, and much more.

How did your build turn out? Share your results, photos, alignment tips, or any modifications in the comments below — I read every one! What aspect of radio receiver design would you like to see covered next?

Happy building, and 73!