Crosstalk Removal

Discussion in 'General Electronics Chat' started by Management, Oct 31, 2007.

  1. Management

    Thread Starter Active Member

    Sep 18, 2007
    306
    0
    I have this instrument that uses air calibration to remove crosstalk between a drive output and receive input. There seems to be a limit on the amount of crosstalk the instrument can remove under air calibration. I was told that the input frequency was 2kHz (as well as teh drive frequency) but the first opamp in the received path is an amplifiying bandpass filter centered at 2.9 kHz. I was also told that if it was set to exactly 2kHz that sometimes the maximum crosstalk limit that can be removed was exceeded and that 2.9kHz solved this problem. What I would like to know is how is this solving the problem? :confused: I can make a link between this bandpass and the crosstalk elimination during air calibration.

    If it is not something general/fundamental or is application specific then forget I asked. Thank you in advance though.
     
  2. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Air calibration? That is a new term to me. Is this something to do with a diplexed radio? The "drive output" and "receive input" do not compute well.

    If you can go into some more detail about the instrument/circuit, it would be helpful.
     
  3. Management

    Thread Starter Active Member

    Sep 18, 2007
    306
    0
    Thank you for responding. I apologize. There is a sensor that uses a drive coil and a receive coil. It is used to measure conductivity of liquids. Air calibration means that the sensor is not dipped into an liquid, hence the "air" term. so the intrument powers the drive coil and depending on the received signal and resistance measurement calculates the conductivity.

    But I am only referring to the time that we calibrate for air meaning a zero conductivity measurement. So during air calibration the is some crosstalk between the coils and it is eliminated to get a value of zero for conductivity.

    I hope this helps. What I am curious about is teh relationship between the bandpass filter and the crosstalk between the coils. I don't understand how change the center frequency relates to the amount of crosstalk between the coils. Or better yet, does this bandpass lessen the amount of crosstalk before the circuit evens measures it. The drive coil signal is 2kHz and the received signal is 2kHz but the bandpass is 2.9kHz. If set to 2kHz would the bandpass filter pick up too much of the crosstalk? Note, this bandpass is the first thing seen by the received signal.
     
  4. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    What allows you to conclude that the bandpass filter is set for 2.9KHz? Is this a setting that you can read off of a display on the instrument?

    hgmjr
     
  5. Management

    Thread Starter Active Member

    Sep 18, 2007
    306
    0
    I checked the circuit configuration and component values myself with hand calculations. Also verfied in it in spice and matlab.

    What is getting me is that in a newer design of the smae instrument that particular opamp used was changed. Also, with this newer instrument the bandpass filter has a center frequency of 2Hz. So I am confused as to why this new instrument has uses a cf of 2kHz and the older one uses 2.9kHz. The whole schematic was basically unchanged. It's still roughly the same preamp. So now I am trying to understand why the older instrument uses a cf 2.9kHz and it's relation to the crosstalk picked up during air calibration. This might be a little too specific situation but any help or incite as to what to look for would help.
     
  6. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    It is possible that the new unit was designed with the newer generation opamps that have lower inherent noise and so the overall system noise has been lowered.

    I don't have an obvious answer to why they would design a 2.9KHz bandpass filter to pass a 2KHz signal.

    Can you provide a diagram of the stage in question? I can try to duplicate you calculations.

    hgmjr
     
  7. Management

    Thread Starter Active Member

    Sep 18, 2007
    306
    0
    Just coming back and reporting as I go through this problem that I have. It has been confirmed that it was designed for 2.9kHz.

    But I have another question but first a little background. These measurements are done with sensors that can be interchanged between instruments. Inside the sensor there are two coils. One drive and one receive. If on the receive path (instrument connection) there are some series inductors ... would that contribute to the crosstalk that occurs?
     
  8. Management

    Thread Starter Active Member

    Sep 18, 2007
    306
    0
    Possibily but the instrument in completely independent from the sensor. The smae sensor was used on both instruments. What can be done in the an instrument to reduce crosstalk between two coils in the sensor? Or you probably can't do anything but try to remove it in the instrument when you calibrate? Assuming the instrument can remove a good amount if there is a lot of crosstalk, am I thinking currectly in this matter?
     
Loading...