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.

INA333-Q1: INA333 Filter Design

Part Number: INA333-Q1
Other Parts Discussed in Thread: INA333, OPA333

Tool/software:

Hello, 

I am currently testing my product under 0-400MHz - 100mA or 60mA - bulk-current-injection(BCI) test. I have an INA333 design for my product. You can see my design below.


As you know during BCI test that we apply current on output cables of product. I have four cables which is VCC,GND,CANH,CANL.

When I apply current on can cables, output change of INA333 is accaptable. You can see the change of graphic below. 



When I apply current on power cables or all cables at the same time, output change of INA333 has peak between 20-30Mhz (always at same frequency). You can see the change of graphic below. 



I have an input filter for INA but it is not enough. Should I add an output filter of INA? How can I solve this problem? 

I can test all your recommendation. I am waiting for your response.

Thank you.

Best Regards.

  • Basri,

    The INA333 is uses chopper amplifiers (OPA333), with integrated resistors to achieve the instrumentation amplifier function.  Chopper amplifiers use internal switching to calibrate out the offset voltage and offset drift.  This switching sometimes leads to noise tones similar to what you are seeing.  However, for the INA333 the chopping frequency is at 125kHz, so this is not likely to be an issue for you since your noise tone is at 20MHz to 30MHz.  I mention the above details because you may have wondered if the chopping noise is the issue here, but I don't think it is.  Thus, this is likely a noise pickup on your cables.

    I think it is likely that the noise pickup is a common mode noise since the cables are adjacent to each other.  By this I mean that the noise signal on the to input wires connected to the INA333 are equal to each other.  When the input signal on each input of the op amp are equal this is called a common mode signal and the amplifier can reject this type of signal (over a limited frequency range).  Unfortunately the 10MHz common mode signal is outside the common mode range of the INA333 so you will not get much rejection.  

    Even though the INA333 does not have good common mode rejection at 10MHz, you do not want to convert the input common mode signal into a differential signal.  In your design tolerance errors in C26 and C27 will convert the common mode signa into differential.  For example, assume that C26 is +10% and C27 is -10%.  This will cause C26 to attenuate more that C27 which will unbalance the common mode input (i.e. convert it to differential).  To resolve this  you should use a differential capacitor that is 10x  the common mode capacitor (see page 53 of the Analog Engineers Pocket Reference below).  The differential capacitor will attenuate the common mode signals at a lower frequency than the differential mode signals.  Also, I think you are using 0 ohm resistors in series with your input filter.  You should choose something like 1k ohm and select the values to achieve a cutoff near the bandwidth requirement of your INA.  The INA333 has a unity gain bandwidth of 150kHz, so you can set this filter far below your 10MHz noise signal. 

    Also, you can add an output RC filter and choose a frequency near your system bandwidth requirement.  For the output filter be careful to keep the capacitance low or you may see some stability issues.  Normally you would connect something like 2k x 200pF (398kHz filter).  You may not need the output filter.

    I hope this helps.  Best regards, Art