Connect with us

noise problem with differential analogue input

Discussion in 'Electronic Design' started by jsmith, Jun 27, 2007.

Scroll to continue with content
  1. jsmith

    jsmith Guest


    I have noise problem with my differential analogue input circuit.

    Input Part #1 :
    Input Part #2 :

    All inputs wires are coaxial ones(but very long; about 15 meters)
    Case/Chassis has connected to Earth of system.
    Also, there is an averaging function for 50Hz noise at

    I will be appreciated with your comments in this


    Thanks in advance
  2. I wonder why you are using coax for a differential signal, this wont
    help, twisted pair should be better.What is the required bandwidth?

    What CM rejection are you getting?

  3. Tom Bruhns

    Tom Bruhns Guest

    What exactly do you mean, "I have a noise problem..." What sort of
    noise? Can you put an image of a `scope trace on a web page,
    illustrating the noise? If you short the two inputs together and bias
    them to an appropriate common-mode voltage (perhaps ground), do you
    still have noise? What sort of diodes are those, and why are they
    there? (Hint: if they are standard signal diodes as the symbol
    suggests, they won't do much to clamp the signal within the common-
    mode range of the amplifiers.) Have you done anything to insure the
    input signals stay within the common mode range of the amplifiers?
    (Do the input signals share a common "ground" with the amplifier
    circuit, and are they nominally near 0V?) One common mistake in
    applying this sort of amplifier is having an input signal with no
    ground reference, and the amplifier inputs "float" to a DC level
    that's inappropriate, outside the common mode range of the amps. A
    resistor from each input to ground can often cure that problem. The
    resistors in the schematic do not do that.

    Like Martin wrote, it would be more usual to feed signals through a
    shielded twisted-pair. Since the "gain" of the amplifier is about
    -12dB, I'd be surprised if you really even needed the shielding, if
    you're not in an electrically/magnetically noisy environment.
  4. jsmith

    jsmith Guest

    50 reads/sec (averaging about 194 ADC samples[10bits] each 20
    milliseconds for each read)
  5. Whats the ambient electrical noise like?

    Ok, just had another look at the circuits, the diodes seem to be
    reversed biased al the time, so they dont actually seem to do

    i think you need to dump the coax, and try some screed cable, proper
    screened microphone cable would do for a test, or even CAT 5.

    A quick tweak would be to drop the value of R76 a bit, and put a
    small pot in series, with 1% R's your CMRR will only be in the 40dB
    range, the pot will get you into the 70dB or more range, but this wont
    solve having he wrong cable type

  6. jsmith

    jsmith Guest

    Thanks for your post and Martin ones,

    I try to put an image of a scope trace, tomorrow.
    All signals (6) are DC voltages between 0 to 10 volts. In fact, they
    are from resistors of voltage dividers that show voltage of some High
    Voltage capacitors.
    Ground of all capacitors (6) have connected to same point (earth and
    my boards chassis) that connect to G inputs. 5 capacitors (signals)
    have posetive charges and 1 capacitor (signal) has negative charge.
    If I short the two inputs together, my system shows (true) Zero
    Diodes are 13 Volts/1Watt Zener for opamp protection from possible
    high voltage inpouts.
  7. Eeyore

    Eeyore Guest

    What is the signal source ? Impedance, signal voltage etc. What's the source of the
    interfering noise ?

    Your configuration isn't a true differential input btw and that won't help if the noise
    level is high.

  8. It looks as if it's 100R per leg, but this guy needs a bit of tuition
    in drawing circuits (sorry John)

  9. jsmith

    jsmith Guest

    The main problem was fluctuation of common lines that solved with
    sharing all commons with some resistors to GND of circuit. but it till
    suffer from some (about +/- 2%) reading fluctuation especially at one
    of the inputs. Because of some field problems.unfortunately, I have to
    record scope trace of signal fluctuations in few next days.
  10. neon


    Oct 21, 2006
    the coax cable must be grounded a the source and nowhere else. the amps must have their ground separete comming from the source.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day