Listen to Latest SolderSmoke Podcast

Sunday, December 23, 2012

Woe is me! Why Podcast is Late: Parasitic Oscillations!



I'm building my FOURTH JBOT and this one is giving me more trouble than all the others combined.  
Here are the details of my tale of woe:
20 meter rig.  Double Sideband.   The JBOT is fed by a simple two-diode, trifilar toroid singly balanced modulator.  NO DIPLEXER TO SPEAK OF.  At the output of the balanced modulator there is a 1000 uH choke to ground and a .001 uF cap to the input of the JBOT.  AF amp is an LM-386.  VXO is a very simple MPF-102 one stage ceramic resonator VXO with no buffer stage.  7 element low pass filter (three coils, four caps). 
All the transformers are wound on FT-37-43 cores.  T3 is four such cores stacked 2X2.
 
The arrangement works beautifully into a 50 ohm dummy load.  But as soon as I connect it to an antenna (a dipole out in the yard, fed with 50 ohm coax) the output signal gets ugly.  
Looking at it on the scope, instead of the nice figure eight pattern (similar to the two tone test pattern of SSB) I get ugly fuzzy outcroppings from the peaks.  Looking at the signal more closely I can see that in addition to the 14.2 Mhz signal, there is another LF oscillation at around 250 kHz. 
I've been struggling with this.  I can't get rid of the LF oscillations. The leads are all short and the inputs are far from the outputs.   I've beefed up shielding, and decoupling.  I've sacrificed a chicken to Papa Legba. Nothing seems to help.  
I THINK the feedback/oscillation is taking place in the JBOT itself -- not through the
audio amp or the balanced modulator or the VXO.
I watch the RF and the AF inputs to the balanced modulator to see if there is any difference between the stable situation (with the 50 ohm dummy load) and the unstable situation (with the antenna),  I don't see any differences. 
I even put an antenna tuner between the final and the antenna anb made sure that the antenna looks like 50 ohms non reactive.  This seems to help a bit, but the ugly instability is still there. 
Help me! 
\


Our book: "SolderSmoke -- Global Adventures in Wireless Electronics" http://soldersmoke.com/book.htm Our coffee mugs, T-Shirts, bumper stickers: http://www.cafepress.com/SolderSmoke Our Book Store: http://astore.amazon.com/contracross-20

4 comments:


  1. Can you walk the circuit with your scope backwards, from the antenna, to see where the LF oscillation is starting? Can you disconnect stages and drive the remaining ones with lab gear or other bench detritus to see if you isolate what stage the problem is starting in? (e.g. drive the PA with another transmitter via a pad?)

    Shots-in-the-dark:
    - check the antenna/lead with an analyzer or other transmitter to make sure they are OK.
    - Try it on a battery, just in case it's some power supply weirdness.
    - try an appropriate choke on the antenna lead. It may be radiating back into the final, due to some kind of mismatch. (Not a solution, but a clue if it helps.)

    My shaky hypothesis: It's starting when you switch to an antenna and the PA has the "worst view" of that switch. Unless it's caused by something radiating from the coax back into another stage, it's the most likely culprit.

    Good luck and 73,
    Bob KD4EBM

    ReplyDelete
  2. looks like the resonant 250KHz created by the 1000uH choke and the total series 1000pF coupling cap and JBOT input capacitance . 1000uH choke and the 1000pf resonance is about 160kHz .add the input capacitance will raise it to 250kHz.
    I would try a 1k ohm resister instead of the choke since the impedance of the choke is about 820 ohms at 14.2MHz.
    good luck and merry Xmas
    wayne VA7AT

    ReplyDelete
  3. I concur with Bob KD4EBM, check the antenna.
    Another idea is use the antenna tuner between the amp and the dummy load to add some mismatch and reactance. If you cannot get it oscillating with the mismatch then that will let the antenna off the hook.
    I know you said that you cannot see the oscillations at the input but can you ground the amp input at the mixer. That will verify that unwanted RF is not getting into the mixer.
    And try grounding the base of the next stage with a capacitor.
    Can you see the oscillations at intermediate stages of the amp?
    Does is start oscillating at power up or after the you put an audio signal in?
    Cheers
    Mark
    VK2WU

    ReplyDelete
  4. Hi, I am having a identical problem with my Beach 40 DSB rig. I also have bypassed all the power connections, inputs and outputs are as separated without completely rebuilding. Everything looks perfect on the scope until I couple the final into the circuit, then the parasitic oscillations start. I realize it is because there is more RF present with the final hooked up. It looks perfect on the scope while hooked to the dummy load, but when I hook up the dipole with a very good match at that frequency I'm testing, the oscillations start. I'm baffled, any ideas? Maybe it will be better when I enclose it in a box.

    Joel
    KB6QVI

    ReplyDelete

Designer: Douglas Bowman | Dimodifikasi oleh Abdul Munir Original Posting Rounders 3 Column