ADAS1000: clarifying when and how VCM is used, and the resulting calculations

I am trying to synthesize the information in the Rev C datasheet along with the digital lead mode discussion in follow-up question #3 in this thread and am finding that there are several details which are ambiguous. Can you please answer the following questions to help clarify the situation (note that I'm using CH0 ... CH4 to indicate the physical PGA + ADC paths in the LA, LL, RA, V1, V2 order):

  1. Is it correct that VCM is always connected to the inverting input of all PGAs except for CH0 ... CH2 when DATAFMT = 0 and CHCONFIG = 1 (analog lead mode)? If not, please describe any other situations where VCM is not connected to the inverting input of the PGAs.
  2. Is it correct that VCM is always defined by CMREFCTL bits LACM ... V2CM, CEREFEN, and EXTCM, regardless of any other configurations (and assuming the selected inputs to VCM are also enabled)? If not, please explain the cases in which the VCM signal can be otherwise defined.
  3. Table 11 on Page 36 does not include CEREFEN. How does the combination of CEREFEN and EXTCM control SW1 in Figure 66? For example:
    1. If EXTCM = 1 and CEREFEN = 0, is it still correct that VCM = CM_IN, or is it high impedance?
    2. If EXTCM = 0 and CEREFEN = 1, does CM_IN contribute to VCM equally alongside any other selections?
    3. If EXTCM = 0, CEREFEN = 1, and LACM ... V2CM = 0, does CM_IN combine with VCM_REF?
  4. My understanding is that in digital lead mode, the results are completely independent of VCM. If statement #1 above is correct, then the "digital block" is manipulating the outputs from the ADCs (which are differential with respect to VCM) so that the result is independent of the VCM signal. Can you explain what this calculation is doing? For WORD1 ... WORD3 I see how the VCM contribution cancels out in a simple subtraction, but it is unclear how WORD4 and WORD5 are derived.
    1. Is it correct that this calculation always work regardless of the VCM configuration, so long as the VCM contribution to the ADC input does not lead to saturation of the ADC? If this is the case then I can live without knowing the mystery of how WCT is calculated.
  5. Is it correct that DATAFMT = 0, CHCONFIG = 0, and CEREFEN = 0 ('Common Electrode A' mode) always produces the following:
    1. WORD3 = (CH2 - CH0 - CH1)/3
    2. WORD4 = (CH3 - CH0 + CH1)/3
    3. WORD5 = (CH4 - CH0 + CH1)/3
  6. Given that DATAFMT =0 and CHCONFIG = 0 in 'Common Electrode A' mode, should I understand that to be a type of digital mode that produces signed values in the data stream?

I think that answers to these questions should resolve all the ambiguities that I currently see. Thanks.

Parents Reply Children
No Data