• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 40 subscribers
  • Views 324 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:

TIDA-00527: TIDA00527

WEBENCH® Tools/TIDA-00527: TIDA-00527

divyanshu bawa
Expert 1325 points
Part Number: TIDA-00527

Tool/software: WEBENCH® Design Tools

Dear Sir, I am trying to make this application more towards the hardware compatible. As I have to put manchester coding . I am trying to balance dc by sending a 1khz wave on the bus line. and if any data will be sent and change in dc line may detect the data. I have seen many modules which they majorly giving the hardware compatibility irrespective of the firmware. Can u help me in getting some better way for Dc balance?

Moreover i calculated the value for inductor and capacitor with my exact data signal rate. But it does not work fine.

  • 0
    TI__Guru 62910 points

    Hello Divyanshu,

    From your message I understand that you are seeing signal integrity issues in getting your data across the L/C filter that is used by this design - is that correct?  If so, can we please take a look at both the application schematic as well as the signal waveforms at the transmit and receive sides?  This will give us a better understanding of the specific problem so that we can help to suggest a solution.

    In general for this design to work we recommend use of high bit rates and DC-balanced signaling.  I'd like to better understand your method of using a 1-kHz wave on the bus line to help with this - can you please provide further details?  My initial thought is that 1 kHz may be too slow unless the L/C values are increased beyond what is used in the reference.  Additionally, another method for DC balance you may want to investigate would be a data encoding scheme such as 8b/10b coding.

    Regards,
    Max

  • 0 in reply to Max Robertson
    Expert 1325 points

    Hi Max

    Yes, You may be right on this part. If I had a data rate of 115200. With what combination of L/C this problem be rectified.

    For Dc balance can we generate a 10khz pulse over the bus for its saturation..Going for manchester coding will make my device for that particular application. So I just want to get a hardware compatible thing. for my different application, i can use dc power line over signal

  • 0 in reply to divyanshu bawa
    TI__Guru 62910 points

    Divyanshu,

    Do you mean that you send a 10 kHz signal for some time before you start your data transmission in order to charge up the capacitors to an average signal level?  If so I believe this could be effective, but it would depend on the L/C values.  With values that are too small you could still lose the accumulated charge within a data frame (especially if it contained relatively long durations with no toggling).  (Or, if the L/C values were even lower you could lose the charge even within a 10 kHz signal period and get no benefit.)

    Are you connecting termination resistances between A and B at the end points of the cabling?  If so you might try removing these since they would impact the corner frequency of the L/C high pass filter.  Assuming your line's idle state is high, you could also try installing a weak pull-up resistor on the A pin of your receiver and a weak pull-down resistor on the B pin.  This would give a differential offset to the receiver even if the line had been idle long enough for the capacitor charged to completely decay.  If you implement this and use a fast enough bit rate (and short enough data words) you may be able to get data transmission without introducing encoding or adding the 10 kHz filler pattern.

    Max

  • 0 in reply to Max Robertson
    Expert 1325 points

    HI

    Please have a look for the signal from tx side. I am continuously sending the signal from my transmitter. at the rate of 115200.

    the inductance is 560uh and capacitance is 10uf..But at the receiver side, I am not getting any waveform.

    115200.zip

  • 0 in reply to divyanshu bawa
    TI__Guru 62910 points

    Divyanshu,

    These waveforms show that the differential signals communicate across the coupling capacitance are degrading rapidly.

    What's the absolute fastest bit rate you can use?  115200 bps is still fairly slow for an AC-coupled link like this.  Can you increase it 10x?

    Are there any termination resistors on the bus?  If so can you try removing them?

    What's the highest coupling capacitance you would be able to use?

    I also noticed that in the time between word transmissions the differential amplitude has decayed fully to 0 V.  Based on this, I think the pull-up/pull-down resistors at the receive end that I proposed previously could help.

    Regards,
    Max