Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Audio feedback problem

Status
Not open for further replies.
Mike? Where you at man? I redid my preselector and I got zero intermod or heterodyne. Zero on both. The '3904 is still in the front end. In fact I'm sitting here listening to some guys and one said he was in California, I'm in Atlanta and the radio is right next to my laptop with the PS plugged in. After I got the preselector right, I buzzed the coating off one of the cat food cans and clamped it down tight with some hot glue. Seems to have got rid of allot of computer noise.

I need to put a PL259 connector on it for when I sling up my inverted V antenna and feed that whole thing through coax. Right now I'm sure I'm getting some noise through the antenna. I've only got about 3 inches of some RG-174U cable coming out of the preselector can. After that it's just a raw long wire antenna, about 10 feet.
 
There is plenty of room on the analogue side of things. Do it right! Don't cut corners. In radio....buffer and shield everything you can manage.

My apologies about buffering, you do have Q21 (which is one of the completely pointless stages), and two darlington buffers, which (as AG says) have no need to be darlingtons anyway.

As for screening, I completely agree - good screening is an excellent idea, and also means it makes your layout less critical. Each module though needs to be individually supplied and have suitable decoupling, as many of the individual stages in the modules do.

However, I can't believe you used the phase 'Do it right!' - on something which is done so wrong in so many ways :D
 
I don't think "hot glue" makes good electrical contact for a ground connection as a shield.
Is it possible to solder a ground wire to a cat food can?
 
Nice try both of you. What was the purpose of this design?

To function as a high quality low power receiver.

Dudes, I sat here last night with my computer on and noisy ass power supply stuck in the hole, running, and listened to California station talking in the wee hours of the morning. Well, it may have only been midnight for them since they are 3 & 4 time zones behind me. Yes and I think that all of them were running bare foot (no amplifier) because I heard one or two say they were. That means anywhere from 75 to 100 watts.

The dynamic range of this thing is also quite incredible. So far I have only had one shortwave station that was able to cause the receiver to clip. One. Now WWCR on 3.210 MHz I almost can't stop them from clipping the signal is so strong. But to keep to circuits as simple and few as I possibly can, what I did was make two IF gain controls so that a single AGC loop will most all of the time, do the job.

I have made little enhancement as I go along like yesterday when I rebuilt the preselector hopefully for the last time. But I as of yet have not been able to make a good enough recording to do this thing justice. I always get some sort of white noise where it seems like you can never pick up the atmospheric crackle in the back ground. I don't know why that is. It may be poor quality DSP in the computer. This is one cheap laptop. I paid like 625 dollars for it brand new.

Anyway, I have tested it on the batteries too. Even built a cool trickle charger that uses no processor. When I get this stuff down in my CAD schematics I'll post more. I keep trying to tell you what you see there is a first iteration of the receiver. This is based very closely to that but there are some differences. Obviously the audio since that is about the only thing I have up to date in the CAD. So you have seen both of them and know they are different but similar. The AGC is allot different. In fact in that schematic I have posted, there are indeed some redundant circuits in the AGC loop. But it does quite well on batteries and my solar panel from PEP boys charges the batteries pretty good on a sunny day....camping! contests!

So you guys can bash away but my greatest critic is the radio itself. Being a ham for 36 years I've owned a few but never really anything great. I have had friends who have had nice ones and I've done my fair share of drooling down at HRO & AES. I used to receive Heathkit catalogs and what I would give if I still had them. But I would go into HRO and find a nice Icom or Yeasu and even Kenwood if I could figure out how to use the monstrosity, and I would make comparisons at various known frequencies and time of day schedules. I'm telling you this thing ranks right in there. Can I notch it down to 200Hz bandwidth? No, I don't have DSP because I don't want DSP. I want to run portable and don't need all that digital stuff loading down my battery. But it does give me a noise advantage. I'm sure their techniques of filtering off the digital are quite superb but there is no way you can get rid of all of it. It's gonna leak through in some ground loop or through the power buss somewhere. The rest can be shielded off.

But all & all, this is a pretty hot little rig. When I get some time (soon I hope), I'm gonna do me up another little double ballanced modulator to plug into some of my home brew amplifiers. I'm definately ready for some QSOs. The ultimate radio specs. Think about that one.
 
Last edited:
I have some FRS "walkie-talkies" for adults. Their output power is low and they operate at a UHF frequency so their range is only a few km. But their audio sound quality is great.

I was never a ham. When I talk to people far away I use the telephone.

In "the good old days" I would dial a local beer company, listen to an ad for their beer then use their phone network to make a free long distance call.
 
I have some FRS "walkie-talkies" for adults. Their output power is low and they operate at a UHF frequency so their range is only a few km. But their audio sound quality is great.

I was never a ham. When I talk to people far away I use the telephone.

I think you're missing the point?, talking to like minded people all other the place is what it's all about - and when I was younger we never had a phone.

I was only ever a class B licence holder, so only had fairly short range, but even with 1W at 145MHz I could get 60 or 70 miles. I have used friends shortwave gear, on their licence (as permitted), but never had any great desire for shortwave transmitting - although I have used, and built, various shortwave radio's (and still have a valve Eddystone one here).

But radio ham isn't about quality, it's about communication - and quality travels far less than poor 'quality'. CW (morse) is the best, and goes further than anything else, SSB is next best, giving low quality audio that travels well - after that comes AM, better quality than SSB, but poorer range. Last comes NBFM, best quality of all (but still only voice quality), and the lowest range of all.
 
I think you're missing the point?, talking to like minded people all other the place is what it's all about - and when I was younger we never had a phone.

I was only ever a class B licence holder, so only had fairly short range, but even with 1W at 145MHz I could get 60 or 70 miles. I have used friends shortwave gear, on their licence (as permitted), but never had any great desire for shortwave transmitting - although I have used, and built, various shortwave radio's (and still have a valve Eddystone one here).

But radio ham isn't about quality, it's about communication - and quality travels far less than poor 'quality'. CW (morse) is the best, and goes further than anything else, SSB is next best, giving low quality audio that travels well - after that comes AM, better quality than SSB, but poorer range. Last comes NBFM, best quality of all (but still only voice quality), and the lowest range of all.

Are you trying to be nice???

What a shocker! :eek:

Well, I guess I deserved the criticisms. I got her licked now. When I get the time I got to go back and see what all I've done...lol.

The AGC locks & holds. No more clipping on anything. I got it hitting two stages. One was the first IF and now it also hits a 2nd IF amp.

It became necessary because I hopped up the rf amp a little and actually got my bandwidth switch working. It is cool! I got it so tight in one position that it is very very difficult to tune SSB even. That position will make a great CW filter. But when I hopped up the rf amp the normal SSB became horrendous! I couldn't handle it. Almost everybody that came in moderately to strong caused horrible cracks in the audio. All they had to do was raise their voice slightly. It was terrible! So the AGC is straight now. I had it on brother Stair on I think the freq is 2.890 MHz, and that loud mouth could not make it clip. I had to go back into the audio again, because if any higher frequencies got through they would completely shut off the audio. It's got a much nicer tone now. I got those freqs up a little.

I will attempt some recordings tonight. I found out what was wrong with the in-line recordings. I had the wrong jack. I assumed it would be mono, but it was a stereo input to the PC. I fixed that but do to the jumping around in the audio, the PC was causing feedback at the those high peaks. It should work now.

So, I will defend myself, but thanks guys for being so critical. I do want this to be good. I put too much hard work into it to accept those kind of flaws. I think you will be impressed this time.

You better be....:mad:

Just kidding. I'm sure you will :)
 
Last edited:
Here's what it sounds like now: **broken link removed**

I just installed some diode clams accross the BFO & VFO so that strong signals would not shift the freqs.

That tone is from some oscillator in the laptop screen.
 
I heard extremely narrow audio bandwidth (no bass and no treble) and severe distortion.
 
Diode clam is used to force the preamplifier to remain in the saturatio region to provide adequate gain during operation with relati small signal input data ...

Google can find anything :)
 
Here's what it sounds like now: **broken link removed**

I just installed some diode clams accross the BFO & VFO so that strong signals would not shift the freqs.

That tone is from some oscillator in the laptop screen.

You should have enough isolation that the Rx signal does not affect your BFO/VFO. Not really sure what you have going on. I would guess poor isolation. Diode clamps would just be a bandaid I think, not to mention make your BFO/VFO clipped.
 
If it's not isolation it's poor regulation of the supply.
The high distortion may be too low BFO injection, not enough AGC, or audio clipping...which may just be over recording.

But we all know this.
 
Last edited:
If it's not isolation it's poor regulation of the supply.
The high distortion may be too low BFO injection, not enough AGC, or audio clipping...which may just be over recording.

But we all know this.

Bingo!

The power supply is regulated but when strong SSB stations came in like on 80 meters I would get this frequency shift, sounded almost like a duck. I noticed the power would dip. So I didn't have any regulators at that voltage so I used diodes to clamp those two voltages. It worked out really well. A vast improvement in frequency stability without using PLL & crystals in the BFO.
 
Hi Space Varmit,
I'm also going to try a 40m SSB. Circuit is here :)
A much advanced circuit :p
 

Attachments

  • circuit.gif
    circuit.gif
    21.8 KB · Views: 221
My video player does not show clipping of the audio. The severe distortion might be caused by the many high gain transistor stages that do not have any negative feedback and each has 40% or more distortion.

My video player shows an extremely narrow audio bandwidth which makes the guys sound like female ducks. I don't hear any hum so why cutoff all the low frequencies?
 
Hi Space Varmit,
I'm also going to try a 40m SSB. Circuit is here :)
A much advanced circuit :p

Well much better designed anyway :D

Notice in particular R1 and R2 and associated components for decoupling and regulation, and the absence of lots of un-needed distortion stages.

However, while it will certainly outperform the circuit in question, a proper superhet is really the way to go - but a direct conversion receiver like this can certainly perform.
 
Last edited:
"The severe distortion might be caused by the many high gain transistor stages that do not have any negative feedback and each has 40% or more distortion."

yeah, there's that too :)
 
Last edited:
Hey SV, I found this NF/IP3 cascade excel spreadsheet lurking in the back corners of my PC. I thought it might be of use to you or someone else.

I think it requires enabling macros, but don't worry it is bug free :)

Spreadsheet in next post. xls upload did not work. Sorry.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top