Mixer Problem

Discussion in 'Wireless & RF Design' started by PRS, Mar 18, 2010.

  1. PRS

    Thread Starter Well-Known Member

    Aug 24, 2008
    989
    35
    I built the mixer in figure A in 1992 and put it away. I blew the dust off it and tested it today. It seems to work just fine. I set it up for experimenting as in figure B and took measurements.

    For the rf voltage I used a signal generator with a sine wave at 0.8 Vpp.

    For the lo voltage I used a function generator with sine wave at 0.8 Vpp.

    I watched the voltage at the output of the IF strip, which was tuned to 455kHz.

    Holding rf at 150 kHz I got good strong 455 kHz sine waves at the output with the LO at 305 kHz and at 605 kHz, as expected. So I know the mixer is good.

    But, I tried to modulate a 455 kHz sine wave with a 1 kHz sine wave, expecting output as per figure C. What I got was a nondescript mess. I tried varying the amplitudes of the rf and LO signals, but just got more garbage.

    Question: Is this the wrong kind of mixer for modulating audio onto rf?
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,647
    2,346
    Hello,

    The RF signal should go on the transformer on the left.
    The LF signal should go on the diode part (center connection).

    See what happens then.

    Greetings,
    Bertus
     
  3. PRS

    Thread Starter Well-Known Member

    Aug 24, 2008
    989
    35
    You nailed it, Bertus! How did you know I was hooking it up wrong? I'm now getting excellent results. A 1lHz triangle wave input is perfectly reproduced at the output of the IF strip. :)
     
  4. bertus

    Administrator

    Apr 5, 2008
    15,647
    2,346
    Hello,

    When you read the datasheets of some Double balanced mixers, you will see where the signals have to go.

    Greetings,
    Bertus

    PS sorry for the late response, I had a lillte holliday.
     
  5. PRS

    Thread Starter Well-Known Member

    Aug 24, 2008
    989
    35
    Thanks again, bertus. I hope your little holiday was fun.
     
  6. bertus

    Administrator

    Apr 5, 2008
    15,647
    2,346
  7. PRS

    Thread Starter Well-Known Member

    Aug 24, 2008
    989
    35
    bertus, that is very interesting stuff to me. Thanks. My mixer was homemade and just a copy of the two-transformer diode ring as portrayed in a University Communications text I kept. After studying the behavior of this mixer, I came to realize that symmetry is very important. Otherwise there is distortion. Next time I build one it will be physically perfect and I'll use an IC diode ring to make sure they're matched. I looked back in my notes and found there to be asymmetry in my waveform output vs input. I suspect it was either non symmetrical transformers or mismatched discrete diodes.
     
  8. bertus

    Administrator

    Apr 5, 2008
    15,647
    2,346
    Hello,

    Did you also have a look at active mixers?
    There are the NE602 / SA612 (with build-in oscilators).
    Or the MC1496.

    I have attached the datasheets for more info.

    Greetings,
    Bertus
     
  9. PRS

    Thread Starter Well-Known Member

    Aug 24, 2008
    989
    35
    I checked out your links and found these devices to be cheap so that I can afford them. Thanks very much again. I looked for such online but I had no luck. I will certainly buy these devices and use them. Right now I have a CA3039 diode IC that I intend to use as a diode ring mixer with two carefully built center-tapped transformers. Any advice on the inductances or the AL of the toroid cores and the number of windings?
     
  10. bertus

    Administrator

    Apr 5, 2008
    15,647
    2,346
  11. PRS

    Thread Starter Well-Known Member

    Aug 24, 2008
    989
    35
    Thank you again, Bertus. You have nailed it with these links. I am most grateful.
     
Loading...