• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 9 subscribers
  • Views 612 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Configuration of GC5016

Lech Zagozdzinski
Prodigy 200 points
Other Parts Discussed in Thread: TMS320C6455

Hi,

I need to write the software for TMS320C6455 with GC5016 in external memory space. GC5016 is in down conversion mode. The configuration prepared by cmd5016.exe works properly. My goal is to change during the work cic_dec, fir_dec, overall_gain (without AGC) and nco frequency, so most of the coefficients.  Number of changes makes impossible using the prepared configuration.

Application note "Digital Down Conversion Mode Gain Settings" gives the rules about  cic_shift fir_shift, gain_lsb and msb and so on. In my opinion, gc5016.exe gives different coefficients than ones produced by this rules.  Overall gain is the same, but registers gain_lsb, cic_shift and fir_shift are set with another value.

The question is, can I find the C code for subroutine for configuring gain registers or new application note (my AN is from 2004, 02.25).

Regards

Lech

  • 0
    TI__Mastermind 37500 points

    Hello

    The cmd5016 tool is used by most customers, and configurations are stored for common cic and pfir decimations.  The .tbl file and .anl file output from cmd5016 detail the gain method that is used.   The concept of overall_gain and manual gain settings has evolved in cmd5016.  The basic concepts are:

        if zpad is used, this signal attenuation is included in the calculation.

        if the DDC mixer is for a real signal, the tuning image rejection through the CIC filter (rejected image) is included in the gain analysis

        cic_dec and cic_shift are set to provide a gain less than 1.  The cic output is clipped at 1/2 for symmetric PFIR operation to reduce overflow.

        pfir_shift is based on the sum of the coefficients, if symmetry (preadd forward and reverse data) is used,  an 18 bit data path, 16 bit coefficients.

        the DDC output gain is a combination of fixed gain, and AGC adaptive gain.  In the .anl file the gain analysis is described, and for the fixed gain, the stage gain

       to desired overall gain is compensated.  The values for all of the gains are given in both the .anl and .tbl file.

        There is not a later application note.  The application note was written before the overall gain optimization was modified in 2006. 

    There are no separate C functions provided, customers who wanted to program the GC5016 either followed the above concept, and loaded for each decimation set, a prebuilt configuration, they can then change the NCO frequency, gain, and PFIR coefficients, based on their application. 

    The second approach is to request the use of the cmd5016 source code, this requires a software license agreement.   This can be pursued through your local TI FAE, contacting Dave Brubaker the product line manager.

    There are special synchronization sequences required to dynamically change the NCO frequency, narrowband gain.

        set the sync_select to OFF

        update the registers

        based on the SIA, SIB or One Shot update, select the appropriate sync_select, generate the sync

        set the sync_select to OFF

    The PFIR coefficients have a similar mechanism.

    Updating changes to the cic_dec, and pfir_dec while the GC5016 is running is not recommended, as the entire device needs resynchronization.

    Regards,

    Radio Joe