dual gate mosfet mixer help.

Discussion in 'Wireless & RF Design' started by perplexabot, Aug 28, 2014.

  1. perplexabot

    Thread Starter New Member

    Jul 3, 2012
    15
    0
    Hello all. I am working on a mixer using a n-dual gate mosfet. I am trying to simulate it using LTSpice IV. Here is my circuit along with the simulation. Please tell me, if you may, where is it that I am wrong.

    Green: Signal
    Blue: Local Oscillator
    Red: Mixer output

    [​IMG]

    Thank you for reading and your time.
     
  2. k7elp60

    Senior Member

    Nov 4, 2008
    478
    69
    Most of the schematics I have seen for duel-gate mosfet mixers has some differences.
    1. G2 has a high value resistor from it to the source(100k). The resistor from Vcc to the G2 is not there.
    1. G1 needs a ground return. With your circuit I would add a 100k resistor.
     
    perplexabot likes this.
  3. perplexabot

    Thread Starter New Member

    Jul 3, 2012
    15
    0
    First of all thank you for your valuable input. Second I would like to apologize for not labeling my gates. Here is another screen shot with labeled gates and hopefully the correct usage of your corrections. What do you think?

    [​IMG]
     
  4. k7elp60

    Senior Member

    Nov 4, 2008
    478
    69
    I would recommend removing R1 & R6.
    Change R5 to 100K
    C2 needs to be connected between V2+ an G2. In other words C2 should be in the circuit like C1.
     
    perplexabot likes this.
  5. perplexabot

    Thread Starter New Member

    Jul 3, 2012
    15
    0
    Thanks again. So here is what I get with your recommendations. I had this output waveform earlier and that is when I decided to dc bias G2.

    Here is the latest version with your improvements:

    [​IMG]

    Weird what is happening on the negative half cycle of the sin wave :confused:

    THANK YOU!
     
  6. k7elp60

    Senior Member

    Nov 4, 2008
    478
    69
    Sorry for the delay in responding to your question. This type of mixer has actually four outputs. From your diagram I calculate the two input frequencies are 13.3Mhz and 4 Mhz. The four output frequencies are 13.3Mhz,4Mhz,17.3Mhz and9.3Mhz. In order for the mixer to work correctly there has to be a circuit on the drain to select which frequency you want. Usually it is either the sum of the input frequencies(17.3Mhz) or the difference of the two input frequencies(9.3Mhz) If I have done the frequency calculation correctly a tunned ckt is required. I have used a similar circuit at audio frequencies and used a low pass filter on the output. In this case off the shelf components can be used(a uh inductor and a pf variable capacitor in a parallel resonate circuit. I can show you a formula for calculating the values if you like.
     
    perplexabot likes this.
  7. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,346
    perplexabot likes this.
  8. perplexabot

    Thread Starter New Member

    Jul 3, 2012
    15
    0
    So this is a unbalanced mixer. I didn't know that. Thank you. I will filter it with a LC circuit and see what happens with the output. I believe since the LC are in parallel w=sqrt(1/LC), this will be the resonance frequency.
    Thank you for the that, I didn't know feedback was provided to both the gates, I have incorporated that.
     
  9. perplexabot

    Thread Starter New Member

    Jul 3, 2012
    15
    0
    Hey, so here is what I currently have, and it still doesn't seem to work : ( Anyone have some info?
    So for this simulation I changed the LO and signal frequencies.
    For the signal I went for 1370KHz (a local radio station) and for LO I went for 915KHz.
    Sig - LO = 455KHz = IF

    I then proceeded to calculate the parallel LC values. So f = 1/[2pi*sqrt(LC)]. I chose a random value of an L that I have in my component box, which is 100 uH. So C = 1/[(2 pi f0)^2 * L], where f0 is the resonance frequency (455 kHz). I got C = 1.22 nF. That is how I chose the values below.
    Here is a screeny:

    Red wave: Signal
    Green wave: LO
    Blue wave: Output (off of R4)
    [​IMG]

    What do you think?
    Thank you for reading.
     
  10. vk6zgo

    Active Member

    Jul 21, 2012
    677
    85
    Shift C6 to the Drain of U1,& delete R4.(or if your simulator won't let you,increase it to 1 MegOhm.)
    C5 ,L1 are the Drain load for U1
    Duplicate C4 on the Drain side of R3
     
Loading...