Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

  • State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 378 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

FDC2212: Spec clarification

FDC2212: Sensor affected by another sensor on the line

EMRE ÖZDEMİR
Intellectual 505 points
Part Number: FDC2212
Other Parts Discussed in Thread: ISO1050

Hello everybody we have a problem we try many things but we couldn't solve it so we need your help.

Our system has sensors and a main board to communicate with sensors and pc.

At sensors side we use ISO1050 Isolated CAN Transceiver.

And to isolate the power we use 5V converter with LDO to 3.3V

sensors register values are configured as 

0x8c40 ->0.196mA current drive

0x820c-> DEGLITCH filter : 3.3Mhz

and the filter is same as FDC EVM (L = 18 uH, C =33 pF)

we use one sided plate because its difference value at same distance better than two sided.and we want to detect both side.

Sensors values affected when another sensor added to the mesh.Its values is stable if it is the only sensor at mesh.

I added the graphs ,the BLUE line is sensor value. these graphs are the same sensor in two conditions.

first graph : one sensor 

second graph : two sensors

So,we didn't understand why this happens, although we did isolate the whole system.And we need suggestions about how could we handle this problem.

Best Regards

Emre.

  • 0
    TI__Mastermind 23810 points
    Emre,

    Are the two sensors physically far away from each other?

    If so, then perhaps grounding is causing the problem. Are you also isolating ground? This app note explains the phenomenon, which is observable on our EVM:

    www.ti.com/.../snoa952

    Possible solutions include:
    - larger local ground
    - connection to earth ground (or larger ground)
    - make sure that PC is connected to earth ground
  • 0 in reply to Clancy Soehren
    Intellectual 505 points

    Hi Clancy ,

    We check distance effect and there is no problem.Does not affect whether they are far or close.

    We found that  the cap at LC tank is creating this problem.When we changed it with a new one this effect decreased but does not removed completely, its acceptable.

    So my next goal is making detection range and difference value getting more.

    There is a post that says the cap can get removed to increase detection range.I tried this but detection gets removed.The data it sends does not change according to the approach.

    If my sensor LC filter is same as EVM and  DEGLITCH filter : 3.3Mhz

    my max difference value is around 80 000 and signal oscillates at around 5000 value.

    If you have any idea about how can I decrease oscillation and increase difference value ,it would be helpful .

    Regards, 

    Emre.

  • 0 in reply to EMRE ÖZDEMİR
    TI__Mastermind 23810 points
    If you remove the cap, then only parasitic capacitance will be there. This will cause the oscillation frequency to increase (possibly out of range), and you will definitely need to increase the deglitch filter.

    Strategies to increase detection distance are:
    - Increase sensor size
    - Reduce noise (for example, better grounding, filtering, improved board layout)
    - If you are using shielding on the other side of the sensor, increasing the distance between the sensor plate and the shielding plate can help.

    For the oscillation, it looks as if it is about 25Hz (based on the time scale you provided). Is there anything else in the system that could be causing that? Slow moving noise like this can sometimes be related to the power supply (especially is using the mains supply from a wall outlet). Are you able to try running this experiment from a battery to see if that oscillation goes away?
  • 0 in reply to Clancy Soehren
    Intellectual 505 points
    I am reading data every 25 ms
    read 4 times and calculating average and send the value at every 100 ms

    -We can't change the sensor size.

    -At start we used two layer pcb and bottom layer is gnd in this stuation we get difference value around 15000
    and second we remove the gnd from bottom layer and difference value increased around 60000
    but this oscillation problem and change in one sensor affects the other sensor problems come out.
    and now to remove this effect I added 2 cm x 3 cm gnd plane at this point the difference value is around 40000
    in this setting, the oscillation problem is removed and effect from other sensor value is decreased
    I am trying to calibrate this gnd plane size to better performance and remove the effects.

    I am measuring these difference values from same distance.I put and sponge on the sensor and I am putting my hand on this sponge.

    As a result, I find that the system is not working properly if It doesn't have enough gnd space.
    I experimented with mains and battery power and I saw that there was no difference between them.
    The DC DC converter doesn't filter this noise.
  • 0 in reply to EMRE ÖZDEMİR
    TI__Mastermind 23810 points
    Your conclusions and experiments make sense. Please let us know if you have other questions.