Fork me on GitHub

Project Notes

#462 HX108-2 AM Receiver

Build and analyse the HX108-2 superheterodyne AM MW radio receiver kit.

Build

Notes

AM/MW radios are of very little use these days in Singapore. There is no local AM broadcasting, just a few stations from Indonesia and Malaysia that can be picked up if you’re lucky. However, there’s still a great deal that can be learned from studying simple AM superheterodyne circuits. I have built a similar kit before - the HX-6B, see LEAP#123 - but I probably know more than I did then, so I’m looking forward to this little opportunity to revisit similar ground.

Although it’s possible to build the entire kit in under an hour, I decided to carefully follow a sub-system by sub-system build from audio out to RF in. Co-incidentally, K7QO is posting an excellent series covering the build of an HX108-2 at the same time … so I am heavily cribbing from there;-)

Tutorials and Resources

The HX108-2 is a pretty popular kit, and a number of very good tutorials can be found on YouTube describing construction and tuning. Here’s a selection…

K7QO HX108-2 AM Receiver: Theory and Construction

K7QO had a series covering the build that appears to have been subsequently re-cut into a series on RF Receivers. The main videos covering this particular radio are part 6 and 7:

RF RECEIVERS – 6 – Audio Amp, Driver and Demodulation Circuits

clip

RF RECEIVERS – 7 – First and Second IF Amps

clip

Other Tutorials

shango066: HX 108-2 am kit radio assembly alignment and test

clip

The Radio Mechanic: an excellent desktop freeform build of the HX108-2 circuit:

clip

The Kit

I purchased my HX108-2 kit from seller on aliexpress. It was generally OK, with no missing parts. The only negative was that the speaker was somewhat corroded.

The basic specification for the kit:

  • Frequency Range: 525-1605kHz
  • Intermediate Frequency: 465kHz
  • Output Power: 100mW
  • Power Supply: 3V, 2xAA
  • Speaker: 57mm diameter 8Ω

Parts

Electronic components:

Ref Item Checked
R1 100kΩ
R2 2kΩ
R3 100Ω
R4 20kΩ
R5 150Ω
R6 62kΩ
R7 51Ω
R8 1kΩ
R9 680Ω
R10 51kΩ
R11 1kΩ
R12 220Ω
R13 24kΩ
W switch and 5kΩ pot
C1 CBM230p var cap
C2 22nF
C3 10nF
C4 4.7µF electrolytic
C5 22nF
C6 22nF
C7 22nF
C8 22nF
C9 22nF
C10 4.7µF electrolytic
C11 22nF
C12 22nF
C13 22nF
C14 100µF electrolytic
C15 100µF electrolytic
B1 antenna BS x 13 x 55
B2 transformer 红 red
B3 transformer 黄 yellow
B4 transformer 白 white
B5 transformer 黑 black
B6 input audio transformer (蓝 blue, 绿 green) 绿
B7 output audio transformer (黄 yellow, 红 red)
D1 1N4148
D2 1N4148
D3 1N4148
V1 9018G
V2 9018H √ 9018G supplied instead
V3 9018H √ 9018G supplied instead
V4 9018H √ 9018G supplied instead
V5 9013H
V6 9013H
V7 9013H
Y 8Ω speaker

kit_parts

Transistors

The transistors are classified by ß(hFE), and it seems many kits substitute different parts of similar capability:

  • 9018G ß = 80-100
  • 9018H ß = 97-146
  • 9013H ß = 144-202

I stuck all the parts in a component testers, and found that one of the 9013H had a much lower ß(hFE) than the others, so I replaced it from spare parts. Might not have been necessary, but avoided a potential issue.

Audio Transformers

There are two audio transformers used in the final stage:

  • B6 input audio transformer (蓝 blue or 绿 green)
  • B7 output audio transformer (黄 yellow or 红 red)

I received a 绿 green and 黄 yellow in my kit. For reference, I measured the resistance of each coil:

  • 绿 green: 220Ω : 104Ω + 104Ω
  • 黄 yellow: 2.5Ω : 6Ω + 6Ω

kit_audio_transformers

IF Transformers

There are four transformer cans used in the design:

  • B2 红 red: local oscillator & mixer
  • B3 黄 yellow: first IF
  • B4 白 white: second IF
  • B5 黑 black: demodulator

While not essential, I decided to tune the IF and demodulator cans to 465kHz prior to installation so they should not require much fine tuning later. Note: 455kHz is the conventional intermediate frequency, but the circuit states it is designed for 465kHz so I went with that.

I tuned the transformers with a 2V peak-peak 465kHz signal from a function generator directly connected to the 2-pin/single coil side of the transformer, and monitored the output with an oscilloscope with the following results:

Transformer Initial Resonant Frequency Tuned Resonant Frequency Output Amplitude After Tuning
B3 黄 yellow 455kHz 465kHz 21.6V
B4 白 white 475kHz 465kHz 17V
B5 黑 black 465kHz 465kHz 6.12V

For B2 红 red, I simply measured its tuning to make sure it was broadly in the ballpark. I’ll fine tune it in-circuit:

Transformer Initial Resonant Frequency Tuned Resonant Frequency Output Amplitude After Tuning
B2 红 red 1455kHz n/a 11.3V

PCB

The PCB has a reasonably good silk-screen to assist with construction:

pcb_front

pcb_rear

Circuit Design

The schematic provided with the kit contains a number of errors. This is a copy that I have annotated with the necessary corrections:

Schematic

The HX6B is a superheterodyne receiver design with two IF stages. The circuit basically maps to the following stages:

  • RF Filter: B1, C1a antenna tuner
  • RF amp: V1
  • Local oscillator and Mixer: B2, C1b
  • First IF Filter and Amp: B3, V2
  • Second IF Filter and Amp: B4, V3
  • Demodulator: B5, V4
  • Audio amp:
    • V5 audio driver/preamplifier
    • B6, B7, V6, V7 push-pull class B power amplifier

A quick sketch from my notes on the functional design:

HX108-2_functional

Test Points

The PCB includes a number of test points with a specified current range:

  • V1 collector: 0.13-0.22mA
  • V2 collector: 0.4-0.8mA
  • V3 collector: 1-2mA
  • V5 collector: 3-5mA
  • B7 high-side: 4-10mA

I believe the way these test points should be used is as follows:

  • build the complete circuit without bridging the test points
  • power the circuit
  • measure the current at each test point - it should be within the limits above
  • after testing, bridge the test point

Since I’m going to build the circuit by stages, I don’t think I’ll be able to make these measurements as I go. I could go back after the kit is complete and un-bridge and test each stage.

Construction

Step 1: Power

Can’t achieve much without power! There are actually two “power rails” in the circuit:

  • the full power of the battery is used to drive the audio amplifier
  • and a ~1.4V rail is established using 2x diode drops to power the RF stages
Ref Item Installed
D1 1N4148
D2 1N4148
R12 220Ω
C15 100µF electrolytic
W switch and 5kΩ pot

Verification:

  • with 3.15V applied (2xAA) the voltage at the anode of D1 measures 1.43V - this is as expected, about two diode drops of ~0.7V √
  • the current drawn is about 7mA √. This is about right: (3.15V-1.43V)/220Ω = 7.82mA

step1

Step 2: Speaker and Battery

Mounting the speaker and battery clip in the housing. Speaker is connected to the PCB, however I’ve left power disconnected for now as I’ll use a bench power supply for testing so I can read off total current and voltage.

Ref Item Installed
Y 8Ω speaker

step2

Step 3: Audio

This step adds the audio pre-amp and power amplification stages, starting from C8/R9 - which is where I’ll inject a test signal to verify things are working. Note:

  • C11 is mislabeled as C6 on the schematic (so I ended up installing both by accident)
  • one of the 9013H tested with much lower ß(hFE) than the others, so I replaced it from spare parts. Might not have been necessary, but avoided a potential issue.
Ref Item Installed
R9 680Ω
R10 51kΩ
R11 1kΩ
C8 22nF
C9 22nF
C10 4.7µF electrolytic
C11 22nF
C12 22nF
C14 100µF electrolytic
B6 input audio transformer (蓝 blue, 绿 green) 绿
B7 output audio transformer (黄 yellow, 红 red)
D3 1N4148
V5 9013H
V6 9013H
V7 9013H

step3 step3-rear

Verification:

  • powered with bench supply at 3V
  • injecting a 0.4V peak-peak 1kHz sine wave at C8/R9 junction (CH1-Yellow in the scope traces below)

First, with the volume set to roughly 20%, total current drawn is about 40mA, and the output (CH2-Blue) is amplified to about 1.55V peak-peak (gain ~ 3.9)

step3-audio-out-40mA

Next, with the volume set to roughly 60%, total current drawn is about 86mA, and the output (CH2-Blue) is already clipping, running at about 4V peak-peak (gain ~ 10)

step3-audio-out-86mA

As I increase the volume, current peaks at about 120mA. Interestingly, the current draw and volume does start to fall off at over say 80-90% on the volume control pot. This probably indicates something in the push-pull configuration is starting to break down (but in a “safe” way).

step3-audio-test

Step 4: Demodulator

The V4 transistor is being used as a detector in this stage.

Ref Item Installed
R4 20kΩ
R8 1kΩ
C4 4.7µF electrolytic
C7 22nF
B5 transformer 黑 black
V4 9018H 9018G supplied instead

step4

Verification:

  • with circuit powered, probe the B5 test point - noise on the speakers √
  • with circuit powered, inject 465kHz carrier with 1kHz AM via 22nF capacitor to the B5 test point - peaked 1kHz tone on the speakers √

Step 5: Second IF Stage

Ref Item Installed
R6 62kΩ
R7 51Ω
C6 22nF
B4 transformer 白 white
V3 9018H 9018G supplied instead

Verification:

  • with circuit powered, probe the B4 test point - noise on the speakers √
  • with circuit powered, inject 465kHz carrier with 1kHz AM via 22nF capacitor to the B4 test point - peaked 1kHz tone on the speakers √

Step 6: First IF Stage

Ref Item Installed
R3 100Ω
R5 150Ω
C5 22nF
B3 transformer 黄 yellow
V2 9018H 9018G supplied instead

step6

Verification:

  • with circuit powered, probe the B4 test point - noise on the speakers √
  • with circuit powered, inject 465kHz carrier with 1kHz AM via 22nF capacitor to the B4 test point - peaked 1kHz tone on the speakers √

Step 7: RF Front-end

All the remaining parts:

Ref Item Installed
R1 100kΩ
R2 2kΩ
R13 24kΩ
C1 CBM230p var cap
C2 22nF
C3 10nF
C13 22nF
B1 antenna BS x 13 x 55
B2 transformer 红 red
V1 9018G

step7

Verification:

  • turn it on, and it works!

Step 8: Final Tuning and Testing

shango066 has a great video on tuning the HX108-2.

Since we don’t have any AM stations in range, I did a rough manual alignment against some digital signals in the local MW band. I read the frequency with a commercial radio, then adjusted the HX108-2 so that the signal appears at roughly the same frequency according to the front dial.

Since I’m tuning with an ugly digital signal, I can’t really tell if the IF and demodulator stages are perfectly aligned, although they seem fine and couldn’t be improved much with fiddling. The main thing I needed to do was adjust the oscillator so that the frequency range roughly lines up with the decal/front dial. I got “close enough for now” by just adjusting B2 红 red:

  • set the variable capacitor to position the front dial at the expected frequency of the signal I was aiming for
  • adjust B2 红 red until the signal is received at best strength/quality

step8-tuning

Step 9: Putting it all together

HX108-2_build

step9-done

Conclusion

My local AM band is completely dead (has been for years) so now I have a perfectly functioning but absolutely useless doorstop;-) But this has been great revision on simple superheterodyne receiver design.

Next steps?

  • next time I travel to a place with AM, I might take it along for a test and some tuning in the field
  • build a low-power AM transmitter for a little local area transmit/receive test

Credits and References

About LEAP#462 RadioAM
Project Source on GitHub Project Gallery Return to the LEAP Catalog

This page is a web-friendly rendering of my project notes shared in the LEAP GitHub repository.

LEAP is just my personal collection of projects. Two main themes have emerged in recent years, sometimes combined:

  • electronics - usually involving an Arduino or other microprocessor in one way or another. Some are full-blown projects, while many are trivial breadboard experiments, intended to learn and explore something interesting
  • scale modelling - I caught the bug after deciding to build a Harrier during covid to demonstrate an electronic jet engine simulation. Let the fun begin..
To be honest, I haven't quite figured out if these two interests belong in the same GitHub repo or not. But for now - they are all here!

Projects are often inspired by things found wild on the net, or ideas from the many great electronics and scale modelling podcasts and YouTube channels. Feel free to borrow liberally, and if you spot any issues do let me know (or send a PR!). See the individual projects for credits where due.