CN0359: Float of signal from Instrumentation Amplifier

We have set up EVAL-CN0359 and have the following problem. 
We are measuring the voltage from J5 pins 2 or 3.
These feed AD8253.
We see the expected distorted square wave on the scope but it is floating up and down between the rails.
Page 18 of the datasheet for AD8253 possibly implies that the inputs are not stabilized by a bias current path to ground in the CN0359 layout.
See also fig. 6 of your article by Kitchin concerning "Common Problems When Designing Amplifier Circuits".

Any advice here? T
hanks--Fritz
Parents
  • 0
    •  Analog Employees 
    on Mar 26, 2019 10:45 AM over 1 year ago

    Hi Fritz,

    Is there a conductivity probe connected on J5? If there is, can you send me the product page for it? Also, can you send a snapshot from the signal captured on your scope? I'll try to recreate the cause of the floating effect you're seeing

    It is correct that a current return path is needed, especially for AC coupled signals. The CN0359 does not use AC-coupling on the signal. The current driving the conductivity probe enters through one of the electrodes (the one connected to pin 1 of J5 and/or pin 2 if it is a 2-electrode probe) and has its return path via the trans-impedance amplifier U19.

    Regards,

    Nikko

  • Nikko-I am attaching pics of the scope shots.

    The first one is a time lapse showing the gradual drift of the overall signal. The other shots are of pins 1,2,and 3 w.r.t ground.

  • 0
    •  Analog Employees 
    on Mar 28, 2019 12:58 AM over 1 year ago in reply to fsonnichsen389

    Hi Fritz,

    I was looking at the datasheet of your conductivity cell you have sent. The CN0359 board had come out from the factory with the following settings: 

    1. Set EVAL-CN0359-EB1Z to following setting:
    2. Excitation voltage: 1V
    3. Excitation frequency: 100Hz
    4. T-H setup time: 10%
    5. T-H hold time: 1%
    6. Temperature coefficient: +2.00%
    7. Cell constant: 1.0000/cm

    I was looking at the datasheet of your conductivity cell you have sent and it seems that it is not matched.


    Could you try changing the settings and let us know if this made any difference?

    Thanks,

    Glenn

  • Glenn
    Thank you for your reply and sorry for the late response. We have been deliberating on this and I wanted to try to thoroughly understand the problem. Here is some more focused information:

    We are running the system within spec for our probe. The IST sensor mandates less than V=0.35 (e.g. 0.7 Vpp). We set f=1kHz and leave the timing parameters alone. We set the cell constant to that in the IST datasheet (e.g 0.68)

    We are measuring the output of diff. amp. U15 with respect to the circuit ground.

    With the above setup we see the down drift shown in the scope shoots that we posted.

    If we crank up the voltage beyond the recommendation by IST, to 2Vpp, the drift seems to go away. This however ultimately damages the sensor and we have seen this.

    The only other clue is that adding a 1M resistor to pin2 of the conductivity cell connector seems to slow but not stop the drift.

Reply
  • Glenn
    Thank you for your reply and sorry for the late response. We have been deliberating on this and I wanted to try to thoroughly understand the problem. Here is some more focused information:

    We are running the system within spec for our probe. The IST sensor mandates less than V=0.35 (e.g. 0.7 Vpp). We set f=1kHz and leave the timing parameters alone. We set the cell constant to that in the IST datasheet (e.g 0.68)

    We are measuring the output of diff. amp. U15 with respect to the circuit ground.

    With the above setup we see the down drift shown in the scope shoots that we posted.

    If we crank up the voltage beyond the recommendation by IST, to 2Vpp, the drift seems to go away. This however ultimately damages the sensor and we have seen this.

    The only other clue is that adding a 1M resistor to pin2 of the conductivity cell connector seems to slow but not stop the drift.

Children
  • 0
    •  Analog Employees 
    on Apr 3, 2019 6:59 AM over 1 year ago in reply to fsonnichsen389

    Hi Fritz,

    I would like to check a couple of things on the board.

    1. Reference voltage on pin 9 of U15 w.r.t. ground. This should be ~1.65V

    2. Short pins 2 and 3 of J5 to ground and measure output of U15. This should give you ~1.65V at the output of the amp. 

    3. What is the coupling and impedance of the probe used? This should be DC coupled and set to high impedance to avoid loading of the probe.

    4. Can we try increasing the frequency used for the measurement? We are currently at 1kHz and the specified range of your cell is from 100Hz to 10kHz.

    Regards,
    Glenn

  • Glenn

    1- pin 9 shows 1.64

    2- The output of U15 with 2&3 at gnd shows 1.828V

    3-Our probe is 10M 11pf

    4-We did the above at 0.3V and 1kHz. At 5.1kHz it is roughly the same, the U15 output showing 1.838V

    We did all of this on a new board. Looking at the output of U15 at 5.1kHz, 0.3V and with the IST sensor, we still see the gradual drift to the rails.

    Thanks-Fritz

  • Any further action on this? We are experiencing plating on our probes, very likely due to this phenomenon. Are you equipped to set up this circuit in your lab and duplicate our results?

    Thanks

    Fritz

  • 0
    •  Analog Employees 
    on Apr 10, 2019 1:44 AM over 1 year ago in reply to fsonnichsen389

    Hi Fritz,

    Apologies on the delay.

    Based on the data that you sent. The system looks to be working fine. I am thinking if the probe that you have could be not functioning the way it is. This could be the overvoltage condition on the earlier part and the added plating on the probes. What kind of plating do you have on the probe?

    Do you happen to have an extra probe to try this out? We have the board but do not have your exact same probe.

    Regards,

    Glenn

  • Glenn-we had replaced the failed probe some time ago and are running 2 or 3 of you boards with fresh probes, all exhibiting the same behaviour. We also have run the CN0359 with a commercial TOPAC probe and it similarly shows this problem.

       I do not quite understand how the system can work with the common mode drifting between the rails and indeed when it does this an error condition is raised by your software.

    Where are you located? We can send you a few test probes.

    Meanwhile have you tested this with a probe of your own and watched the voltages with a o'scope? If you see the problem you can continue analyzing it. If not we can buy the same probe and do some tests with it here.

    Also is it possible to communicate via telephone? We lose a whole day between each exchange otherwise and we can quickly try the things that you suggest over the phone.

    Thanks you

    Fritz