Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

How exacly 80m ARDF works?

Status
Not open for further replies.
The 'bulk' capacitors (supply rail decoupling capacitors) have virtually zero effect on reception sensitivity but they do have some effect on the audio quality and stability. For example, the LM386 draws very little current when quiet but the current increases with volume level. That in turn causes a slight drop in the supply voltage which can then decrease the volume. Under worst circumstances, you get a feedback effect where the amplifier makes a noise like a bad 2-stroke engine, the capacitors help by keeping the voltage up during peaks of current demand so lessening the effect. The value (their storage ability) is what matters but in this application the exact type you use is unimportant as long as it has >12V rating.

The VFO is basically an amplifier and a tuned feedback circuit. In this case you DO want it to oscillate. The feedback path is between pins 1 and 2 but the amplifier is all inside the MC3361 so it will not work if it is removed I'm afraid.

Brian.
 

I am preparing to rebuild the receiver from the 100% new components, so I can compare reception with the existing one.

I am also trying to get some kind of antenna.
I am going to buy wire soon, can anyone tell me what's the best option? The price does not really matter.... is thicker wire better?
 

Thicker wire is better but the difference it makes is very, very, very small, at least in a receiving system. When the wire is used to transmit there may be points along it where there are high voltages or high currents so reasonably thick wire may reduce losses but wire of say 1mm diameter is still good for several hundred Watts.

I really wouldn't worry about the antenna too much. Ideally it should be 0.25 Wavelengths long (~21.4m for 3.5MHz) but many other factors influence it's performance. If you can keep even a short length of wire clear of obstructions and away from sources of interference it will work far better than an ideal antenna in a shielded location. At 3.5MHz and 7MHz the signal will pass almost unhindered through brick, concrete, glass and wood but try to avoid running it along metals or beside mains power cables as these are a source of interference.

Try to add a ground connection if it is possible. Mains Earth is better than nothing unless you are in a building shared with other users as the Earth wire can carry interference and actually make matters worse. The best ground connection is to a metal rod driven into the Earth but if you are on the top floor of an apartment block that may not be easy to do! A reasonable alternative is metal water pipe but you must be sure the connection to the ground is metal all the way, most modern water supplies use plastic pipes which makes them useless as an electrical ground. NEVER use gas pipes, they might work but the risk of explosion or harm to anyone working on the pipes, even some distance away isn't worth taking.

Brian.
 

Okay, so as mentioned earlier I have made the very same receiver once again myself.

The only difference is that I used BB329 instead of BB104G.
I have also used 1000uF instead of 100uF supply voltage filtering capacitor and 470uF instead of 100uF LM386 supply voltage filtering capacitor.

I have found the following layout on some kind of polish forum:
lidiaBADboard.png



And I decided that I will use it because I dont trust my own designs as I am very inexperienced, but....
... (after I etched it, drilled, started soldering)...
... it has turned out that the layout above has a fatal mistake in a VFO pot connection.

I had to cut the tracks and add one jumper wire, but I managed to fix it.
I have skipped the capacitors marked on the picture because they are not present in the current version of schematic.

I have also used 10uH SMD inductor instead of axial one one the picture.

Then I used osciloscope to check the frequency and I decided that I must add extra 22pF capacitor so it has the correct VFO frequency range.
After that I tested it with the test crystal signal generator.

The test signal sound is clear and loud, even clearer than in the previous one, and I can clearly hear "disortions" while tuning. I heard them also in the previous one, but not so clear.

Then I made first test with 3-4 m transformer wire antenna at around 1PM and..... nothing.
I still can't receive anything... does it really mean that I have to have a better antenna?


Also, what's the purpose of extra capacitors at MC3361?
 
Last edited:

WARNING: not only is the positive end of the tuning control not connected to a voltage (which I think is what you fixed) but the voltage regulator has the wrong connections. Check the connections on the 78L08, with the legs facing towards you and the flat side facing upwards, they are OUT, GND, IN from left to right. If you didn't notice that before powering up, you may have damaged the 78L08 but everything else should have survived.

The two capacitors circled in blue do absolutely nothing. They are part of the FM discriminator circuit which is what the MC3361 is really supposed to be used for but you are not using that part of the circuit at all.

The capacitor circled in orange will have the effect of making the LM386 amplify a little more, particularly at higher audio frequencies so it will make the sound slightly 'hissy'.

When you say 'distortions', can you describe exactly what you can hear or maybe record a few seconds of audio as you tune across the band and post it here please. Its easier to diagnose a problem when you can hear what you are observing there.

Incidentally, did you notice the board says MC3661 as the first IC. It is just another error on the board, there is no such device as the MC3661, it should say MC3361.

Brian.
 

WARNING: not only is the positive end of the tuning control not connected to a voltage (which I think is what you fixed) but the voltage regulator has the wrong connections. Check the connections on the 78L08, with the legs facing towards you and the flat side facing upwards, they are OUT, GND, IN from left to right. If you didn't notice that before powering up, you may have damaged the 78L08 but everything else should have survived.
That's the very good observation. Infact, the 78L08 got hot instantly along with 7812 from power supply just as I started first test (only power supply section soldered, no VFO, MC, N whatsoever). That's the first thing I fixed already, I just somehow forget to note that in my post.
It seems that I also had/am having a cold joint issue around LM386 but I fixed it for now.


When you say 'distortions', can you describe exactly what you can hear or maybe record a few seconds of audio as you tune across the band and post it here please. Its easier to diagnose a problem when you can hear what you are observing there.
Those are samples recorded just (at the middle of the night) with mobile phone. The setup is new receiver board with a few meters transformer wire antenna (around the room) and GND connected (kinda accidentally, but it made signal louder) to oscilloscope GND/earth (but the oscilloscope input was NOT connected to the board during recording). The board was powered through the 7812 vreg.

The mp3's are zipped, because forum does not accept .mp3 extension.

NoTestSignal.mp3
View attachment NoTestSignal.zip
00:04 - I power on the receiver
00:05-00:50 I slowly turn pot from most right side to most left side
00:51 I switch the CW/SSB switch
00:52-end I slowly turn pot from most left side to most right side

WithTestSignal.mp3
View attachment WithTestSignal.zip
00:04 - I power on the receiver
00:05-00:38 I slowly turn pot from most right side to most left side (00:23 to 00:27 you can hear the test signal sound changing)
00:39 I switch the CW/SSB switch
00:40-end I slowly turn pot from most left side to most right side (00:56 to 1:03 you can hear the test signal sound changing)
 

Strange!

The noises are quite normal and the test signal tone changing is exactly as I would expect as you tune through it's carrier. The CW/SSB switch doesn't actually do anything to the reception by the way, it just drops the level of higher audio tones on the basis that Morse code doesn't need such a wide audio range. It doesn't change the sensitivity or tuning in any way.

The only thing I can think of is that the antenna filter isn't passing the signal to the MC3361. Bear in mind that your test oscillator is probably producing a huge signal level compared to outside stations. I checked a moment ago and the 80m and 40m bands are particularly quiet right now but they are very busy in the evenings. Can you try again in the evening when the bands are busier and make another recording please.

Brian.
 
  • Like
Reactions: Garyl

    Garyl

    Points: 2
    Helpful Answer Positive Rating
I will try again, but now my wire has arrived and I am going to try somewhere outside the city.

This is the wire I have:
IMG_20170830_122842.jpgIMG_20170830_122903.jpg
I have cut 20 meters of single wire from that. I am going to test both receivers outside right now. I'll post here again when I come back.
 

My test with 12m~ antenna outside the city was succesful. I've managed to receive first CW and then even SSB, but voice was very squeaky and hard to underestand!

The setup was around 12 meters of wire, starting at the roof, height 2.31 and connected to the wooden pole at 2.5m.
The ground was connected to old metal rod that I put into ground (not very deep).

I had issue with LM386 on new receiver (suspect cold solder joint, but maybe something else is temporarily turning off the LM386, but after a while it started working and worked good for several hours until I had to leave).

At first I connected antenna and ground to old receiver. I've managed to receive CW stations, but didn't manage to read it. It seemed kinda like there are two signal received at once.

Then I tried SSB, but no luck.

After that I switched to new receiver. I also managed to receive CW.
Then after a while (I straightened wire, maybe it helped, I also moved with the receiver to the inside the shed) I managed to receive SSB. I am 100% sure that it was voice, but most of the time it was squeaky. There were times when I could understand something, but squeak/high pitch tone seemed pretty persistent no matter what I do.

I've tried tuning, changing volume, changing the antenna pot, etc, but no luck...

All tests were made between 6PM and 8PM.

Still, I can confirm that VFO "stabilizes" after a while, just as the original circuit author wrote in the description.

What do you guys think? I can provide some sample recordings from those tests if it's needed. Is this squeaking normal or do I need to fix something? Maybe add shielding to VFO.... or metal enclosure for entire circuit...

PS: Please note that I received all this stuff outside the city, so it does not solve my problem with receiving stuff in my work room... but maybe it at least proofs that both (?) receivers are working?
 

I find I need to dial around on the BFO to find a position of the knob where I can understand SSB speech.

Until I purchased the right kind of shortwave radio, I only heard SSB voices sounding like Donald Duck on my cheap radios which lack BFO.
 

Brad: this is a direct conversion receiver so there is no BFO as such but it should be possible to resolve SSB by adjusting the main tuning to beat with the correct sideband.

Garyl: this is good news but I can't understand why you hear squeaking sounds. What you should hear is the pitch of the sound, whether CW or SSB, starting high pitched then getting lower as you tune to the source's carrier frequency the getting higher again as you tune through it to its other side. For CW you can just tune for the best tone, bearing in mind that you get the same tone at each side of the signal frequency but one may carry less interference than the other. For SSB, again you should hear voices that change in pitch as you tune through them but this time, only one of the side of the signal should be understandable, the other side may seem to have the correct voice tones but it will be 'spectrally inverted' so high and low pitch are mirrored.

Possibly the problem is the tuning range isn't quite right. CW stations are usually centered around the low frequency end of the band (3.5 - 3.6MHz on the 80m band) and voice usually around the remainder (I think 3.6 - 3.8MHz in your country) so maybe your tuning range is off to one side of where you want it. On the assumption that you were hearing CW properly, you need to make the tuning range slightly higher by reducing the inductance of the tuning coil, the one between pins 1 and 4 of the MC3361. It probably only needs a slight adjustment so try spreading the turns slightly wider and see if it helps.

Brian.
 
Until I purchased the right kind of shortwave radio, I only heard SSB voices sounding like Donald Duck on my cheap radios

I realize that this circuit is very simple, but I've seen multiple videos of people getting better, understandable voice quality on it. You can search for such videos on youtube. They are using the very same schematic as I.


On the assumption that you were hearing CW properly, you need to make the tuning range slightly higher by reducing the inductance of the tuning coil, the one between pins 1 and 4 of the MC3361. It probably only needs a slight adjustment so try spreading the turns slightly wider and see if it helps.
Unfortunatelly this coil is in SMD package. The circuit description says that 1210 SMD coil is required for the stability of the VFO. It's said that the author experimented even with various SMD cases and chosen 1210 as the best one.
Also it's said that the VFO stabilizes itself after 30 minutes.

I can setup a better antenna (but still a bit far from my homeplace), between two poles, around 35 meters distance. The pole height is 4m. Should I try to put antena higher, for example on a wooden sticks on top of the poles?

I will try to post some recordings soon, with both SSB and CW.
 
Last edited:

The antenna is more than adequate already, do not go to much trouble trying to improve it.

There is nothing in the VFO to stabilize it at all. I suspect what the author found was that after 30 minutes the temperature of the MC3361 stabilized and that has some influence on the frequency. The 1210 should be no better than any other type, the number just refers to the length and width of the package but it is only the inductance that really matters.

Brian.
 
  • Like
Reactions: Garyl

    Garyl

    Points: 2
    Helpful Answer Positive Rating
There is nothing in the VFO to stabilize it at all. I suspect what the author found was that after 30 minutes the temperature of the MC3361 stabilized and that has some influence on the frequency. The 1210 should be no better than any other type, the number just refers to the length and width of the package but it is only the inductance that really matters.

You are right, the temperature was mentioned here. It was also noted that even after a while the frequency changes itself around 100Hz per 30 minutes.


Would putting everything in the metal enclosure help much?
I've seen various parts of PCB covered in metal enclosures in the old CRT Televisions.
 

No, the enclosure wouldn't help. In some products, like CRT televisions there were circuits producing tens of Watts of interfering signals only a few cm away so screening was essential but with no powerful sources of signal so close, it would make no difference to your receiver.

The usual way of compensating for frequency drift is to select capacitors with equal but opposite drift characteristics so the tuning stays fairly constant as the temperature changes. Unfortunately, capacitors with special temperature properties are difficult to find these days and even if you could source them you would have to find how much drift you needed to compensate. I would be a difficult task and certainly not worth doing. Most VFOs are frequency stabilized using a PLL (Phase Locked Loop) in modern designs because it is relatively easy to do and gives excellent results but it would make your receiver much more complicated.

Brian.
 

Today I have made another desperate attempt at receiving something at my home location and finally got a little success!

I've remembered that there is old, unused Ethernet (CAT5 cable) going from the house roof to my room (4 to 2nd floor). It was just cut near at my window, so I grabbed the end, took out the isolation, soldered all the little wires together and used it as antenna.
I have also connected the ground terminal to the CO2 pipe (heating system, not gas!).

This way I can receive some voice, right now the quality is not good enough, but I didn't try for long. Tomorrow I will try a bit more.

So it means that the receiver was working correctly after all, it just needed a better antenna....

Anyway, is there any way to amplify signal from the antenna?


PS: I have also received some kind of strange music, who is transmitting that? :D
 

The strange music will be from one of the commercial broadcast stations on a nearby frequency. Direct conversion receivers can pick them up but have difficulty recovering the audio frequencies. If you tune through an AM station you should hear a high pitched whistle which drops to a low pitch them rises again as you continue tuning. The optimum point for music is exactly in the middle of the two tones where the pitch is lowest but without being able to lock to the signal, you will find it difficult to get the notes to sound right, they will tend to be shifted up or down slightly in pitch.

I have a receiver here that uses a synchronous AM detector, not far different from the way a direct conversion receiver works but it has a PLL tracking the signal to produce a tuning voltage to 'lock' the frequency exactly.

Brian.
 

The strange music will be from one of the commercial broadcast stations on a nearby frequency. Direct conversion receivers can pick them up but have difficulty recovering the audio frequencies. If you tune through an AM station you should hear a high pitched whistle which drops to a low pitch them rises again as you continue tuning. The optimum point for music is exactly in the middle of the two tones where the pitch is lowest but without being able to lock to the signal, you will find it difficult to get the notes to sound right, they will tend to be shifted up or down slightly in pitch.
So you say that was a normal AM radio?

Please hear, this is the only sample I took yesterday:
View attachment 2017_09_02_23_12_38.zip
The only word I can recognize in this recording is "Ukraine". I started tuning again at the end of the recording.
 

That doesn't sound too far from what I would expect but it does seem the tuning range is very restricted.

You need to understand how different types of transmissions are sent and the limitations of your receiver:

An AM broadcast (like domestic radio stations) has three component parts, a lower sideband, a carrier and an upper sideband. The sidebands are mirror images of each other with the carrier in the middle. The sidebands are created by the modulation and even if there isn't any (silent station) the carrier is still present.

A DSB (double sideband) signal is like AM but the carrier is missing, it can be removed by filtering but in most cases it is electronically canceled. The transmission is only the two sidebands which have no 'reference' to recover the modulation against. The receiver has to substitute its own carrier before the modulation can be heard.

SSB (single sideband) is a more efficient version of DSB. It recognizes the two sidebands are identical but opposite 'twins' so they carry duplicate information. Usually by filtering, one of the sidebands is removed so the other sideband and the carrier are both missing. Again, the receiver has to substitute it's own carrier before any modulation can be recovered.

Now looking at the receiver:

It is a 'direct conversion receiver' which means it converts the incoming signal directly down to audio frequencies. In contrast, most receivers are 'Superhetrodyne' types which do it in two or more steps with the middle frequency being fixed so it is easier to filter out adjacent signals and interference. That type of receiver is more complicated to build though.

What your VFO does is provide a reference signal, it then goes with the incoming signal to a 'mixer' stage and comes out as the sum and difference of the two frequencies. For example, if the incoming signal is on 3.600MHz and the VFO is tuned to 3.601MHz, the difference of 1KHz would come through the loudspeaker. Note that a signal on 3.599MHz will also produce a 1KHz output at the same time though! For receiving Morse code you don't tune exactly to the signal, you tune some distance from it so the keying of the carrier at the transmitter comes though as a keyed tone (difference of the frequencies) to your ears. That's why adjusting the VFO makes the pitch of the sound go lower or higher, you are moving your reference and hearing it move against the fixed frequencies coming in from the antenna.

For receiving AM you will always get a background tone, it's the difference between your VFO and the carrier from the station. With some care you can tune the VFO exactly to the carrier so the tone goes very low in pitch but it is very hard to remove it completely and the result is the voice or music is 'out of tune'.

For receiving DSB, if you tune the VFO to exactly match the point where the missing carrier would be, in theory you can recover the audio perfectly but for technical reasons I won't explain just yet, it rarely works well. Note though that if you are off tune, one sideband will be closer to the VFO than the other so you get a mixture of higher pitch and lower pitch versions of the modulation instead of both being the same.

For SSB, again you tune to the frequency where the missing carrier would be but this time because only one of the sidebands was transmitted, you do not hear the opposite sideband at all. The effect is that as you tune through the signal the pitch of the audio will sweep from too high to too low and it's up to you to find the most intelligible point. A drawback to direct conversion is it can't tell which sideband is which so you hear the transmission twice at two adjacent tuning positions, one has the high and low pitch of the audio reversed which makes it impossible to understand, even if the voice pitch is approximately right. The other tuning position is the same incoming signal but with the wanted sideband so it should be possible to understand. It takes practice to get the tuning just right though!

Brian.
 
  • Like
Reactions: Garyl

    Garyl

    Points: 2
    Helpful Answer Positive Rating
It is a 'direct conversion receiver' which means it converts the incoming signal directly down to audio frequencies. In contrast, most receivers are 'Superhetrodyne' types which do it in two or more steps with the middle frequency being fixed so it is easier to filter out adjacent signals and interference. That type of receiver is more complicated to build though.
Isn't this method also used in TV tuners? I am planning to use one in the FM receiiver, afaik they output the "intermediate freq" which I have to somehow convert to audio...


But let's back to the receiver of this topic. So indeed a second tuning knob would be useful...
What would be the best way to add it to schematic discussed here? Just add a second, for instance, 1k pot next to 10k pot?


Also, I am planning to do a 40m/80m band version of this receiver.
I know that I must change antenna filters and VFO for other band.
As for antenna filter, I realize I will have two separate filters, right? And just... switch them somehow?
But what about VFO? I guess it would be simples with DDS...

Does a wire length from antenna filter to mixer chip matters much?

I also wonder which parts of circuit I can put on separate boards without having too much stray inductances impact...
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top