Documenting a TRF Radio build

Thread Starter

Anonymous User 30

Joined Sep 4, 2017
38
Greetings,

I'm new to electronics and am interested in the field just as a hobbyist. I've read the Complete electronics self teaching guide by Boysen and Kybett and am now moving to Basic Antennas ARRL and Building your own transistor radios by R. Quan. To be honest im attracted to RF technologies and once I can get a handle on building my own radios I would like to get my ham licence.

I decided to document this project here as I would have many questions along the way and one thread should suffice in asking all questions.

So to begin, this circuit below, should I be picking up modulated AM signals on the oscilloscope? I believe I have constructed a bandpass filter operating as an antenna tuner:

TRF1.JPG

I constructed this circuit so I could detect AM signals on the oscilloscope, in an effort to understand how to construct (bias) the amplifer for the next stage. Unfortunately all i'm receiving is noise on the output. R1 is grounded to the oscilloscope, C1 is a variable capacitor and L1 and C1 are tuned around 1kHz. This frequency is the centre of our AM broadcast band - Sydney, Aus.

Any advice would be appreciated.
 

Papabravo

Joined Feb 24, 2006
14,253
I don't know about Australia, but in the continental US the AM Broadcast band is 550 kHz. to 1620 kHz. The center or geometric mean would be 943.927 kHz. Your 1 kHz filter is off by several orders of magnitude. I have no idea if your oscilloscope can receive small the small currents and voltage in a RF signal. Unless you have a nearby station you're going to be looking a microvolts and nanoamperes.
If your scope has a 50 Ω input you should probably use that. The 1 Meg plus 15 pf of a typical 10x probe is too much impedance to overcome.
 

Ylli

Joined Nov 13, 2015
976
I assume you meant 1 MHz and not 1 KHz. 1KHz is an audio frequency, not a radio frequency. This is a very basic band pass filter, and it will have very limited rejection of signals away from it's resonant frequency. In other words, when looking at the output on a scope, you are really seeing a composite of all the AM radio stations in your area. It will just look like a bunch of noise.
 

BobTPH

Joined Jun 5, 2013
2,505
Use a parallel tuned circuit ( L and C in parallel between antenna and ground). Use 50 ft of antenna, and you might see something on the scope.

The MK 414 is an amazing little 3-pin chip tha makes a reasonable TRF receiver.

Bob
 

MrChips

Joined Oct 2, 2009
21,682
Describe your antenna. How long is it?
As Bob says, you need to be able to tune a resonant L and C parallel circuit.

Here is your basic crystal radio circuit.

1598057041605.png

Even this will work but you cannot select the radio station. All strong signals will be received on top of each other.

1598057112124.png
 

Thread Starter

Anonymous User 30

Joined Sep 4, 2017
38
Papabravo - I mean't 1000KHz and the inductor and capacitor values work as that being the resonant frequency. My scope unfortunately has only 1M Ohm inputs.

Yii - Thanks for your comment, I'll look into more sophisticated band pass filters.

What do you think I should turn my attention to. Is it worth building the amplifier now with careful biasing?
 

Thread Starter

Anonymous User 30

Joined Sep 4, 2017
38
Mr Chips, thank you for that.

Could you describe the antenna I should be constructing? The ARRL book I have read essentially describes the construction for large dipoles. Im would like to hold off on building something too large at the moment.
 

Thread Starter

Anonymous User 30

Joined Sep 4, 2017
38
[/QUOTE]
Use a parallel tuned circuit ( L and C in parallel between antenna and ground). Use 50 ft of antenna, and you might see something on the scope.

The MK 414 is an amazing little 3-pin chip tha makes a reasonable TRF receiver.

Bob
Thanks for that info, definately the antenna construction is not as straight forward as I thought it might be, I underestimated the weakness of the signals.

You constructed a series RCL filter.
http://www.maxmcarter.com/classecalcs/lccalc.html
Without the values it is a guessing game.
Thank you. L = 230uH VC = 60 - 160 pF R = 33Ohms.
 

Ylli

Joined Nov 13, 2015
976
For the AM broadcast band, about the only kind of resonant antenna that you could build would be a magnetic loop. As far as an 'outside' antenna, the best you can do is just the longest and highest wire you can manage. As far as the following circuitry, that type of an antenna will just look like a high impedance voltage source.
 

sparky 1

Joined Nov 3, 2018
302
The filter given has a Q of about 4.4.
As you change the Q value in the calculator. You you can see the affect on the other values.
In TRF frontend with preselectors and Q multipliers also have regenerative effect. Sort of an art in signal amplification.
There are some articles in radio magazines years ago. With a lot of tuning some weak signals could be improved.
There is more information on peak tuning. Since you can move the notch however there is a null. I believe the null could remove
or displace the signal improperly so that it is picked up in another section where it can be heard. It is common receive unwanted QRM
around 455kHz. It used to be recommended to make a simple 455 kHz oscillator for radio work.

The BFO ( beat frequency oscillator ) used to be variable. Adjusting BFO also helped with reducing unwanted interference.
The modern day radio has fixed frequency BFO. In radio audio having a strong presence over a microphone often times means using
an audio filtering to adjust bass and treble to get your voice heard over the noise. The Baxandall circuit is an example of peaking and nulling
in audio where adding more sensitivity at certain frequencies improves the quality of the sound.
 
Last edited:

MrChips

Joined Oct 2, 2009
21,682
As Bob recommended, about 50ft of wire strung horizontally outdoors should give you enough signal.
You should see about 50mV when the parallel LC circuit is tuned to a radio station.

Here is my basic crystal radio.

1598183584736.png

1598183520798.png
 

Thread Starter

Anonymous User 30

Joined Sep 4, 2017
38
Hi,

Interestingly I bought a crystal radio kit from my local electronics retailer just to see how it works and couldn't get a signal with it. I actually took it back to the store and was given another for free, but still no luck. I decided to build the crystal radio from scratch with Mr Chip's design and I got a few stations:

crystalset.jpg

The ferrite rod inductor was a previous inductor I used, but it was a bit temperamental so I replaced it with a 220uH choke.

I now have the information I need from the oscilloscope, which was a probe before and after the diode respectively:

DSO00017.jpg

Vpp at around 600mV. Frequency incorrect.

Here is the probe after the diode roughly 50mV Vpp:

DSO00018.jpg

Now i need to turn my attention to building the amplifier, does anyone have any suggestions or warnings regarding pitfalls? Should I build the amplifier after the parallel LC circuit (before diode) or after the diode? Any suggestions on what transistor to use? Thanks.

EDIT: The stations getting through are perhaps one or two. Is this due to inferior LC parallel circuit, or just the strength of the signal getting through?
 
Last edited:

MrChips

Joined Oct 2, 2009
21,682
Good to hear that you got it working.
The signal before the diode is RF. The signal after the diode is AF.
Since all the impedances are high you have to make sure that you do not load down the circuit with low impedances.

You can start by amplifying the RF signal with a JFET (junction field-effect transistor).
FETs in general are voltage sensitive devices with very high impedances at the gate, i.e. they take very little input currents.

You can try this circuit. I will add a BJT (bipolar junction transistor) AF amplifier later.

Crystal Radio JFET.jpg
 

Thread Starter

Anonymous User 30

Joined Sep 4, 2017
38
Good to hear that you got it working.

You can try this circuit. I will add a BJT (bipolar junction transistor) AF amplifier later.
Thanks for that circuit. I decided to go straight to the AF amplification because I cannot find that FET, I believe it to be obsolete.
I got success with a little amplifier I made to drive a 6W RMS 4 Ohm speaker. Schematic of the circuit is here:

TRF 1 BJT.JPG

Photos here:

File_000 (8).jpeg

File_001.jpeg

Here are the oscilloscope tracings of the input signal into the amplifier and the amplified output going to the speaker:

DSO00020.jpg

I understand what you mean by not loading down the signal with low impedances. I report back with another FET and try the circuit you suggested. Hopefully I can get a two stage amplifier to work. I noticed selectivity of stations improved with the amplifier.
 

Thread Starter

Anonymous User 30

Joined Sep 4, 2017
38
Ran into trouble. I've decided to hold off on the FET amplifier and try to walk before running. I've decided to try a two stage BJT amplifier.

Here is the circuit below:

2 stage BJT.JPG

The first problem is the AM signal is disappearing when the AM signal is being fed into the base of the BC548 and reappears when disconnected from the base (probed after the diode). Look forward to your help.
 

MrChips

Joined Oct 2, 2009
21,682
Where did you find this circuit?
All your resistor values are way too low. For example, the combination of R1, R2, and R3 sets the input impedance to be less than 100Ω while the crystal radio prefers to see an impedance of greater that 10kΩ.

Hold off on your two-stage AF amplifier design for a while until you learn how to select appropriate resistor values for your amplifier. I don't have the time to show you right now, maybe later or if another AAC member can show you how.
 

sparky 1

Joined Nov 3, 2018
302
The design of an antenna preamplifier takes into account the signal strength of the desired reception.
A weak signal can easily be buried in noise with too much amplification. Let's say you would like to amplify 0.40 microvolts rms
or -115 dBm having 50 Ohm impedance, that would be exceptional !

As a first goal tackling a low signal level of 0.5 - 3 uVrms is not that easy but it can be rewarding to someone inclined that way.
That could be a distant AM station barely legible late at night. If that signal is 2 uV and you amplified it 4 times or 6 dB
that 8uV might be just enough signal not too much noise. So having a long wire antenna and ground can be used as a weak signal for testing
a preamplifier without lots of equipment.

Some jfets have a gate threshold near zero which makes them very useful. Radio is full of tales and fantasy circuits, many bad antenna circuits.
Because you know the operating point for this is low and are able to make a small amplification at that level then you will no doubt want more amplification. There are times when only a slight improvement is all that is needed. Other times the noise is unavoidable and 50 dB is better than having the station fade in and out. Also adjustable gain along with the selective higher Q can greatly improve the crystal radio performance.
 
Last edited:

Thread Starter

Anonymous User 30

Joined Sep 4, 2017
38
Where did you find this circuit?
All your resistor values are way too low. For example, the combination of R1, R2, and R3 sets the input impedance to be less than 100Ω while the crystal radio prefers to see an impedance of greater that 10kΩ.

Hold off on your two-stage AF amplifier design for a while until you learn how to select appropriate resistor values for your amplifier. I don't have the time to show you right now, maybe later or if another AAC member can show you how.
I am using a 4 Ohm speaker for this amplifier, not the crystal headphone. Would it cause an issue? EDIT: I see the issue now, this is because of the voltage divider effect. V Zin / (Zin * Z source). Tried it with the original amps biasing resistors and managed to preserve much of the signal. Will get back to you on this.
 
Last edited:

MrChips

Joined Oct 2, 2009
21,682
How to design an AF speaker amplifier

Let us begin the conversation by discussing the importance of impedance.
What is impedance?
Impedance is the measure of the property of the device to impede the flow of DC and AC.
The unit of impedance is ohm, the same as for resistance.
Resistance is the same as impedance except that resistance is the same for DC and all AC frequencies.
Impedance changes with AC frequency.

When we are dealing with DC circuits or pure resistors we use the term resistance.
When we are dealing with AC circuits and devices that may have inductance and/or capacitance we use the term impedance.

Ok, that is just a preliminary to the following discussion.

Loudspeakers are essentially inductors. They have DC resistance and AC impedance. Your choice of a loudspeaker with a rated 4Ω impedance is not the best choice. The most common impedance for loudspeakers is 8Ω. Modern systems use 4Ω speakers because they produce twice the amount power with the same voltage.

Power = V * V / R

For your application you want to go with higher impedance, not lower. You can get loudspeakers with higher impedances such as 16Ω or 32Ω. Transistor and tube amplifiers generally have high impedance outputs, usually greater than 1000Ω. In order to use such amplifiers with a 4Ω or 8Ω loudspeaker a matching transformer can be used. However, we will avoid that route with proper amplifier design.

On the other end of the scale of speaker impedance, the earphone you got with the crystal radio set is a high impedance earpiece. The transducer element is a piezoelectric device that has an electrical model as a capacitor. The DC resistance is ∞ (i.e. infinitely large). The AC impedance is also very high. This is desirable with a crystal radio because the earpiece with high impedance will respond to voltage and take very little current.

So what to do?

One solution is to find high impedance headphones. I have headphones that measure 250Ω per side with a DMM set to ohmmeter range. These were got from the airlines when you could purchase their headphones for in-flight listening. If you can find similar headphones that would be a good place to start. Stereo headphones generally come with 3 connections, tip, ring and sleeve. The left and right speakers are wired in-phase with a common connection to the sleeve. If you join tip + ring and use the sleeve as common the speakers will be in phase but the combined impedance will be half that of each speaker. If you use just the tip and ring and ignore the sleeve the combined impedance will be doubled but the speakers will be out of phase. Since the headset is placed against the ear you will not be able to notice the difference.

1599740744933.png


An alternative is to find high impedance headsets, 1000Ω and higher that are built specially for crystal radio listening.

1599741732878.png

1599741620073.png
 
Top