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.

ADS8900B: RREF_FLT and CREF_FLT Components

Part Number: ADS8900B
Other Parts Discussed in Thread: REF5045,

In the data sheet it states :-

"Select RREF_FLT and CREF_FLT to limit the broadband noise contribution from the external reference source. The device takes very little current, IREF, from the REFIN pin (typically, 0.1 μA). However, this current flows through RREF_FLT and may result in additional gain error."

I am using the REF5045IDGK as my external reference and I have placed it very close to the ADS8900BRGE pin 3. I have also connected 2 x 10uF 10V 0402 X5R Caps on this track. Is this additional  RREF_FLT and CREF_FLT required ? It does not appear to be mentioned in the REF5045IDGK  data sheet ? Also "However, this current flows through RREF_FLT and may result in additional gain error." - does this mean that there is a preference not to place the RREF_FLT ? 

Thank you in advanced 

  • In addition I have noticed in the data sheets for the application circuit that "The output broadband noise of the reference is heavily filtered by a low-pass filter with a 3-dB cutoff frequency of 16 Hz." consisting of 2 Caps and 2 resistors as shown in Fig. 105. How do I figure out if I need this or something equivalent 

    Another Question : In section 10.1.4 Reference Decoupling it states that the ESR = 1-Ω but in Fig. 105, 110 & 111 it appears that the ESR = 0.1Ω ? 

  • Hi Shmuel,

    The RC filter for the reference input is optional, but a good idea to help with overall system noise.   Since the output noise of the REF5045 is only specified over a bandwidth of 0.1 to 10Hz, there may be higher frequency noise that could show up in your measurements.  Adding this filter will allow you to reduce some of this noise if needed.

    The typical reference input current for the ADS8900B is 0.1uA which will flow through RFLT.  This will generate about 100uV of voltage error when RFLT=1Kohm, resulting in a gain error of 100uV/9V=0.001%.  For most systems, an additional 0.001% of gain error is negligible.  If RFLT=100kOhm, the system error would be closer to 0.1% which would be more problematic in a system targeting overall accuracy of better than 1%.

    Regarding section 10.1.4, use ESR=1ohm.  This has been built and tested on the evaluation board and has shown best overall noise performance.

    Regards,
    Keith Nicholas
    Precision ADC Applications