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.

INA240-Q1: input filter

Part Number: INA240-Q1
Other Parts Discussed in Thread: TINA-TI, INA240

Dear team,

In our datasheet, we describe how Rs impact the Vout accuracy. Does C1 and C2 impact the accuracy or other performance? For example, 1nF~10nF.

I made a simulation with TINA-TI as below. According to below simulation, the input filter greatly impact the Vout. Could you please help analyze this reason?

1. no input filter

2. Add filter RC, but no C1& C2. It seems that the output is out of range. So the input filter can't be added?

3. Add filters RC and C1, C2. The output is different with above picture, but it is still out of range. But I don't understand why the voltage between INP and INN is +/-15V.

Thanks & Best Regards,

Sherry

  • Hi Sherry,

    C1/2 do not affect accuracy for the simulation here. The problem you described are due to setup issues in the simulation file.   

    The biggest issue is that the input stimulus is a 320KHz triangular waveform, while the input differential filter is more than 10x lower at about 24KHz. You need to either remove the filter (C1) or lower the input speed. To the same token, the output filter will limit bandwidth as well. You could probe at the output pin for unfiltered output.

    Another issue is that you have an effective shunt resistor of R11+R3, which is 10.0002Ohm. This rails the output and contribute to the distorted output signal.

    I’ve corrected these issues and made the attached file, hope it can be somewhat useful to you.

    sbomac8a (3).tsc

    Regards, Guang

  • Hi Sherry,

    In our datasheet, we describe how Rs impact the Vout accuracy. Does C1 and C2 impact the accuracy or other performance? For example, 1nF~10nF.

    Any imbalance of C1 and C2 will degrade the common mode rejection of INA240. I have seen circuits where the common mode rejection was totally ruined only because of having common mode filters at the inputs.

    C1 and C2 must not even be exactly identical but must also show identical temperature drifts and identical long term drifts. Even the least imbalance will result in a desaster in a high side measurment. Run your TINA-TI simulation and you will see.

    Kai 

  • Dears,

    Thanks for your feedbacks which are helpful!

    Recently the customer encounters one issue which maybe related to the filters, but except the filters, there are also other questions. The issue description is as attachment, please check it!

    INA240 Current Detection Deviation Issue.pptx

    Below questions are listed in the ppt too. You can understand the questions more based on the waveforms showed in the ppt.

    Due to case's urgency, need your support for below questions before 2022/5/18:

    1)Will the common capacitors @INP/INN make the output deviation?

    2)Is there some common capacitors internal in INA240 between INP/INN and GND?

    3)Dose TI have suggestion value for the common capacitor?

    4)Dose INA240 have requirement of dv/dt between INP/INN and GND?

    5)Can INA240 work normally when the voltage of Shunt+/Shunt- as picture shown?

    Thanks & Best Regards,

    Sherry

  • Hi Sherry,

    Here are answers to your questions:

    • This use condition involves switched common mode (PWM), these capacitors can affect output.
    • No
    • For switching inputs, we don’t recommend filters due to potential poor matching; for DC or pseudo DC type applications, the value is not critical. Although a corner frequency of 1MHz or below is a good choice.
    • No
    • Yes, it works well with PWM input conditions. But other parameters such as switching frequency and duty cycle also impact output. Enough time should be allowed for the INA240 output to respond.

    Here are some observations/questions, please help clarify:

    The PWM frequency is 100KHz, and the duty period is ~2uS. The 2uS time after a PWM edge is difficult. Measurement should improve with longer duty.

    There are significant undershoots on the INA240 output, but no corresponding ones on the current probe. Do you know the reason why?

    In the simulation schematic, does L1 represent parasitic inductance of R1 or something else?

    Regards, Guang