About TI demo code:
configure_new_compensation(volatile FILTER_PMBUS_REGS *source)
volatile struct FILTER_REGS *dest;
if(source == &filter1_pmbus_regs)
{
dest = &Filter1Regs;
}
else
{
dest = &Filter0Regs;
}
dest->COEFCONFIG.all = 0;
//Configure bin 1 with the new bin 0 coefficients
dest->FILTERKPCOEF0.bit.KP_COEF_1 = source->FILTERKPCOEF0.bit.KP_COEF_0;
dest->FILTERKICOEF0.bit.KI_COEF_1 = source->FILTERKICOEF0.bit.KI_COEF_0;
dest->FILTERKDCOEF0.bit.KD_COEF_1 = source->FILTERKDCOEF0.bit.KD_COEF_0;
dest->FILTERKDALPHA.bit.KD_ALPHA_1 = source->FILTERKDALPHA.bit.KD_ALPHA_0;
//Configure filter to use bin 1 coefficients only
dest->COEFCONFIG.all = FILTER_CONFIG_BIN1;
//Configure bin 0 with the new bin 0 coefficients
dest->FILTERKPCOEF0.bit.KP_COEF_0 = source->FILTERKPCOEF0.bit.KP_COEF_0;
dest->FILTERKICOEF0.bit.KI_COEF_0 = source->FILTERKICOEF0.bit.KI_COEF_0;
dest->FILTERKDCOEF0.bit.KD_COEF_0 = source->FILTERKDCOEF0.bit.KD_COEF_0;
dest->FILTERKDALPHA.bit.KD_ALPHA_0 = source->FILTERKDALPHA.bit.KD_ALPHA_0;
//Configure filter to use bin 0 coefficients only
dest->COEFCONFIG.all = 0;
//Reprogram the remaining filter coefficients
dest->FILTERKPCOEF0.all = source->FILTERKPCOEF0.all;
dest->FILTERKPCOEF1.all = source->FILTERKPCOEF1.all;
dest->FILTERKICOEF0.all = source->FILTERKICOEF0.all;
dest->FILTERKICOEF1.all = source->FILTERKICOEF1.all;
dest->FILTERKDCOEF0.all = source->FILTERKDCOEF0.all;
dest->FILTERKDCOEF1.all = source->FILTERKDCOEF1.all;
dest->FILTERKDALPHA.all = source->FILTERKDALPHA.all;
//Configure filter to use bins as defined
dest->COEFCONFIG.all = source->COEFCONFIG.all;
change old PID parameters to new ones.
Why first configure " COEFCONFIG.all = 0" then "configure dest->COEFCONFIG.all = source->COEFCONFIG.all? Why used new SETA coefficient then change to new SETA/B/C coefficients?
why not use "configure dest->COEFCONFIG.all = source->COEFCONFIG.all " directly? just change new SETA/B/C coefficients directly from old SETA/B/C coefficients?
