• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 50 subscribers
  • Views 299 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Original question:

INA240-Q1: input filter

INA240-Q1: INA240 Input Filtering - Nyquist filtering

Jennifer Stander
Intellectual 470 points
Part Number: INA240-Q1
Other Parts Discussed in Thread: INA240, BOOSTXL-3PHGANINV

Hi, 

I am using the INA240A1 (gain of 20) in a design, after the INA240 I put a 20 kHz nyquist filter to limit noise going into my ADC sampling at 40 kHz.  Unfortunately, this causes the noise induced by the PWM switching to settle more slowly which is not desireable but makes sense. 

To not degrade the INA240A settling time, is it OK to move the nyquist filter to the input of the INA240? I understand about the series resistance issues with gain but do not have a feel for negative effects by adding a capacitor across the differential inputs and using this capacitance in conjunction with the input series resistance to form a Nyquist filter. I would not be adding capacitors from the inputs to ground, just across the input. 

In the BOOSTXL-3PhGaNInv example schematic, I see a 1500 pF cap has been added here. Is there a limit on how big this capacitor can be? I am looking at maybe placing 100's of nF. Is this OK or will it mess up the common mode rejection or other critical design characteristics?

Just curious in this same schematic an RC (20 ohms, 2200 pF) filter is added after the INA240 with a cut-off of 3.6 MHz. Is this filter important?

Thank  you,

Jennifer

  • 0
    Guru 140200 points

    Hi Jennifer,

    you may want to watch this training video from Peter Iliya (especially 5:56):

    https://www.youtube.com/watch?v=hLGN_wl-xgA

    I would limit the filtering resistors to 10R and take a C0G (NP0) differential input capacitor of some hundreds of nF. But avoid adding uneven stray capacitances to the both inputs of INA240. So, you could remove the solid ground plane a bit arround the filtering components to keep the stray capacitance to signal ground minimal.

    If the speed of output of INA240 is too low, you can easily insert a fast buffer between the INA240 and ADC and do some of the filterings arround this buffer. There's no need to drive the ADC with the INA240.

    Kai

  • 0 in reply to kai klaas69
    Intellectual 470 points

    Thank you, 

    I will move forward with a 180 nF COG capacitor across the differential inputs and understand about the resistance and stray capacitance. 

    I already have an amplifier between the INA240 and ADC. In the  BOOSTXL-3PhGaNInv design there is a filter between the INA240 and the Op-amp that drives the ADC with a bandwidth of 3.6 Mhz. Can you comment on how the cut-off frequency for this filter was determined? What is the lowest filter cut-off in this location that will not cause the INA240 common mode setting time seen at the ADC to increase. I would like the INA240 specifications to drive the setting time not the filter cut-off.

    -Jennifer

  • 0 in reply to Jennifer Stander
    TI__Prodigy 660 points

    Hello Jennifer,

    The filter you are asking about between the INA240 and  in the comparator in  BOOSTXL-3PhGaNInv design is implemented as a charge bucket filter and so capacitance value was more relevant in determining cap choice . Please see section 3 of attached link for explanation of choices made in implementation of this filter.

    Current Shunt Amplifier Considerations for Driving SAR ADCs

  • +1 in reply to Jennifer Stander
    Guru 140200 points

    Hi Jeniffer,

    and the additional RC filters at the inputs of comparators are used to absorb the toggling noise of the comparators and by this to isolate the outputs of INA240 from these comparator togglings. Without these RC filters the output voltages of INA240 would be contaminated with the comparator noise.

    Keep in mind that the aquisition time of ADC must be matched to the time constant of charge bucket filters and the settling time of INA240. I guess the 20R and 2200pF values of RC filters were more found by experimenting than by calculating to find an optimal precicision and the aquisition time of ADC was modified later to properly work with this charge bucket filter.

    Kai