455kHz IF Signal Generator with AM Modulation

Introduction

In order to learn and understand how radio receivers work, I am building an AM radio from an ELENCO AM radio kit. Part of the testing of the build of the  kit (see here) requires you to hook up a signal generator. The signal required is a 455kHz carrier wave containing a 1kHz amplitude modulated signal. According to the kit instructions, when testing the radio IF circuit, the signal generator's IF AM signal should cause your radio kit to produce an audible tone, if you built it correctly.

After much searching and weeks of breadboarding different circuits that didn't really do what I wanted, I stumbled across a circuit in a 2008 edition of an Australian magazine called Silicon Chip (see here). The circuit is called the Minispot 455kHz Modulated Oscillator By Mauro Grassi. I assume it's free to copy as there's no copyright information on the web page. However, if you reproduce the circuit here, please acknowledge it's source. 

The circuit consists of a classic relaxation oscillator using a pair of BJTs with C-R timing circuits to create an audible tone at around 450Hz. This is then fed through a third BJT where the 455kHz is mixed in from a ceramic resonator. 

The ceramic resonator is a fairly common part, often used in remote controls. Strangely enough, the week I discovered the circuit, I had been given a Toshiba DVD player to repair. it was not repairable but the remote control did yield a 455kHz ceramic resonator. If you're not so fortunate to have an old remote control to cannibalize,  you can easily get a bag of 10 ceramic resonators from eBay for around £2 GBP. Alternatively, try your local recycling centre - my local centre has a 'second use' shop where they re-sell many perfectly good, unwanted items. There I found TV/DVD remote controls for 50 pence each.

Schematic

The schematic below is one I have re-drawn from the original, after breadboarding the circuit to check it worked satisfactorily. The re-draw was because I changed transistor types and capacitor values. If you click on the picture, a full size will pop up for viewing/printing.

My re-drawn schematic. A slightly modified version of the original from the Silicon Chip magazine.

Parts list

X1      455kHz Ceramic Resonator

D1  1N4004

SW1 SPST switch

Q1,2    BC548B transistor

Q3 2N2222 transistor

LED1 5mm Red LED

R1 1.5k

R2,3 33k

R4,5 1k

R5 470R

R7 22k

R8 10M

C1,2 47nF

C3 220uF Electrolytic

C4,5 68pF

C6 27pF

extras : 9v battery, battery clip & wire for the antenna

Circuit notes

I modified the output capacitor from the original 22pF to 27pF as that's all I had. I don't believe it has affected the functionality.

I also modified the choice of NPN BJTs from the original schematic to suit what I had available. I tried using a BC549 for Q3, but found the 450Hz modulation wasn't as clear when viewed on my oscilloscope. Swapping out for a 2N2222 (which from testing showed less gain than the BC549) made the modulation look much cleaner on my oscilloscope.

For the antenna, I grabbed a scrap 4" piece of single core wire. I calculated a proper 1/4 wave antenna for 455kHz would be in excess of 500 feet! A bit silly. I'm not sure if I should use a few feet of coiled wire instead... I guess there's insufficient output to worry about reflected signal from the end of the antenna. (I'm just a beginner so don't know much about radio antennas yet. The result of the 4" wire is that my portable Sony AM radio is tuned in to the centre of hte AM band, it screams when you hold the antenna to the back of the case - so I guess it works OK)

The circuit draws about 12mA from a PP3 9v battery. Half of that is for the LED to show it's switched on! without the LED it's more like 6.5mA. If you plan on running the circuit for long periods (I can't think of any reason why you would want to!) then leave out LED1 and R6.

As built, it produces an audible tone at around 455Hz. This is lower than I intended, but much more peasant to listen to while tuning the IF coils on your radio :) If you want to increase the frequency of the audible tone, reduce R2/C1 & R3/C2. if you want to better understand the relaxation oscillator part of the circuit, search the internet for "2 transistor oscillator", there are lots of resources that explain it clearly. 

Finished circuit on copper strip board

I built the circuit using copper strip board. 1.5" x 4" is plenty. You could make it more compact if you like but I wanted the component layout to be visually similar to the schematic. I think when you're learning electronics,  it's worth an extra few pennies to have a layout that you can easily relate to your schematic. Troubleshooting my dodgy soldering seems to be a regular part of my learning electronics.

The finished circuit powered up

Probing the signal output
Note I laid out the circuit to be visually similar to the schematic.
Q3 and the 455kHz crystal are at underneath the relaxation oscillator part of the circuit.

Checking the carrier frequency is 455kHz
The crystals are nominally 455kHz but are affected by capacitance and temperature.
My frequency counter shows 455.1kHz which is good enough for me.

My oscilloscope shows the 450Hz audible modulation on top of the 455kHz carrier signal. 
At 0.2V/div, you can see the signal is around 1.6Vpk-pk

Copper strip board layout

The images below were created to help me visualize the layout of the schematic on copper strip board. they're here for reference if you want to use them. you could make the circuit a lot more compact, but as I previously stated, the aim was to produce a board I could easily follow with the schematic.

Top side - shows component layout and which strips have 9V, GND and ANTENNA

Top side - shows 'ghost' of components and breaks and solder points from underside

Under side - shows breaks and solder points