VCO too noisy in input

Discussion in 'General Electronics Chat' started by abforce, Feb 3, 2014.

  1. abforce

    Thread Starter New Member

    Nov 14, 2013
    19
    0
    Hey guys:

    I am using a VCO which type is UMG-123-D14-G. Now I find it is too noisy in the output node to make the input signal readable. The input signal is given by an OpAmp and is good while separated from VCO.

    How could I decrease the noise in the input of the VCO? Is there anything I can do with either the OpAmp or the VCO?

    P.S. The VCO input capacitance seems to be too big for my OpAmp.
     
  2. Papabravo

    Expert

    Feb 24, 2006
    10,180
    1,800
    Not familiar with this part. Can you give us a link to your datasheet? Can you also tell us what opamp you are using? Would a schematic diagram be too much to ask for?
     
  3. abforce

    Thread Starter New Member

    Nov 14, 2013
    19
    0
  4. ScottWang

    Moderator

    Aug 23, 2012
    4,858
    768
    How do you connect the op Amp and VCO? (solder, breadboard,other)
    How about the distance between the output of OP amp and VCO?
     
  5. Papabravo

    Expert

    Feb 24, 2006
    10,180
    1,800
    Can you identify the source of the noise and the coupling mechanism? Is the opamp the only one in the package? Is there another unconnected amp in the same package as the one you are using?
     
  6. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,293
    1,262
    What is the tuning frequency?
     
  7. abforce

    Thread Starter New Member

    Nov 14, 2013
    19
    0
    I connect the VCO with a shelled wire. about 5 cm long, because the two boards are seperated.
     
  8. abforce

    Thread Starter New Member

    Nov 14, 2013
    19
    0
    I find the noise is mainly at 59MHz, which is the working frequency of the VCO. I have no idea how they are coupling because the VCO has an offset and the same noise waveform even no input is connected. I think it just kind of 'pollutes' the input in some sense?
     
  9. abforce

    Thread Starter New Member

    Nov 14, 2013
    19
    0
    Tuning frequency is 9 kHz. Is there anything I can do with it to improve?
     
  10. ScottWang

    Moderator

    Aug 23, 2012
    4,858
    768
    Have ever try to connecting the input of VCO to GNG and then how is the output?
    And which Op Amp are you using?
     
  11. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,293
    1,262
    I can only offer some ideas:

    Put the IC in a metal screen to shield it. Obviously a ground plane under it.
    Use something like this to filter Vcc coming into the little box.
    http://avx.com/docs/Catalogs/w2fw3f.pdf

    Use shielded wire and RF connectors at input and output.
    Put the connectors right next to the pins.
    http://www.mouser.com/Connectors/RF-Interconnects/RF-Connectors-Coaxial-Connectors/_/N-89mcy/

    Place a small cap across Rf to roll of the frequency of you drive op amp to limit the bandwidth to your 9Khz signal.

    Hope one of them helps. Are you sure it is not just your scope picking up the signal?
     
  12. Papabravo

    Expert

    Feb 24, 2006
    10,180
    1,800
    The point of my question was to gauge your understanding of the situation. Noise is a multi faceted phenomena. In order to mitigate it's effects you have to identify the source and the coupling mechanism.

    Identifying the source can be done by isolating the source e.g. turn off the florescent lights or isolating the pickup e.g. surround with a metal cage.

    There are two indirect coupling mechanisms. Capacitive coupling passes high frequencies from the source to the system under test. Inductive coupling passes low frequencies from the source to the system under test.

    The direct injection of noise from a noisy source can also occur.

    When I was a young engineer there were guys who didn't understand the fundamentals whose stock solution was "hang a cap". Needless to say most of them ended up in Marketing makeing many times the bucks I was ever able to haul in.
     
  13. abforce

    Thread Starter New Member

    Nov 14, 2013
    19
    0
    Nice Idea. I tried to connect VCO to ground and also to power supply. Seems it will be less noisy but still some noise being there. The bias may be disappeared . I used a Ad8036.
     
  14. ScottWang

    Moderator

    Aug 23, 2012
    4,858
    768
    The ideal output is zero volt when the input is connected to GND, so the problem now that you have to check the power itself, it's not easy to get a pure DC without noise.

    How many volts of noise do you get?
    The DC Characteristics of AD8036:
    2 mV Offset
    10 V/C Drift
    Ultralow Distortion, Low Noise
    –72 dBc typ @ 20 MHz

    What kinds of power are you using?
    Maybe you can attach the board you have.
     
  15. ScottWang

    Moderator

    Aug 23, 2012
    4,858
    768
    I'm just thinking maybe you can using the battery then it will avoid the noise come from the power, and then check dose it still have the noise and it will more easier to find out where is the noise came from.
     
Loading...