If I write the following 5 coefficients (10 bytes) to Page 4 / Regs 14-23 (Biquad A), it provides a high-pass filter with Fc=71 Hz for Sampling=48000:
0x7F , 0x66 , 0xC0 , 0x4D , 0x00 , 0x00 , 0x3F , 0x66 , 0x00 , 0x00
The above only works for the full 2nd order Biquad equation: (works in the sense that I can get MATLAB to match expected results)
Text "* BQ: H(z) = (N0 + 2*N1/z + N2/(z2)) / (32768 - 2*D1/z - D2/(z2))"
Text "* Filter Coefficients in format N0, N1, N2, D1, D2"
Using the 1st order equation fails: (assuming the need to drop the zero terms = my mistake)
Text "* 1O: H(z) = (N0 + N1/z) / (32768 - D1/z)"
Text "* Filter Coefficients in format N0, N1, D1"
If I feed all 5 terms with 0's back into the tool, it works.
Question: How does the hardware determine if it's being asked to execute a 1st or 2nd order filter? Does it look for 0's for the last 2 coeff, then use N0,N1,D1 ordering?
Or, the question put another way, if the tool only gives me 3 coeff (Butterworth_1), where do I put them?
Our 1st order filter was generated in an unknown way a long time ago. If it were created today, the tool would output only 3 coeff. Part of me then wants to put these in the first 3 coeff regs. Part of me wants to put them in the properly declared positions (per the datasheet) of 0, 1, 3 with zeros in positions 2, 4.
Any ideas? Thanks, Chris
